List of questions asked to Matteo Gandola, Autoscalp creator and founder.
1) Autoscalp price engine logic required many years of research,
why don't you standardize the process and sell it to retail brokers ?
2) So why don't you run your own customers pool ?
3) So why do brokers offer their tools to customers instead ?
4) Brokers with hidden agendas offering standard tools to customers,
sounds quite nasty. Can you make me an example please ?
5) Your view seems negative about brokers in general , can you elaborate more ?
6) You mean that often beneath a "broker label" there is a market maker?
7) So is all about broker bad practice , is such practice legal ?
8) What you mean by brokers routing orders to external sources ?
9) How about your view on another hot topic: LEVERAGE ?
10) What is the best retailer market platform in your opinion ?
11) What you mean by broker-side created prices ?
12) What about high frequency trading and flash trading platforms ?
13) I didn't quite understand the difference between flash trading and High Frequency Trading .
Can you make me a clear example ?
14) I see you have neither facebook nor linkedin page... why ?
15) So, intellectual property is important in autoscalp but if competitors try to steal your know how?
How do you react?
16) Can you please explain an example on how you defend from procurement outsourcing
or other intellectual property theft attempts?
17) ...but if you outsource development, isn't that dangerous from a proprietary know-how point of view?
List of Topics discussed with the autoscalp team:
18) Fundamental Analysis .
19) OHLC candlesticks : a broker side convention passed as a customer convenience.
20) Technical analysis.
21) Mark to market .
22) News release deal entrapment and analyst advice traps.
23) Charts always lie .
24) "tight spreads" , false friend.
25) Leverage is the hang up game.
26) High watermark.
27) Black Scholes model.
28) Market transparency.
29) Broker fees - Software fees.
30) Manual trading.
31) Why Traders can't profit with manual trading.
32) Segregated account:
33) STP guarantees the customer doesn't receive in-house brewed artificial price.
34) Market price and broker price.
35) Why all platform show same instruments to trade on .
36) Expensive models work better than cheap ones.
37) How CTA and Asset Managers lie with the track records, how brokers lie with reports.
38) How Track Records Certifications are faked.
39) Why remote hosted trading systems , grids , mirror trading can not profit.
40) Percent Allocation Managed (PAMM) accounts : deception based lie.
41) Mirror trading : lie based on asymmetric information.
42) Pool account misuse.
43) Close all positions before Friday market close.
44) Recipes through which companies boil and lure traders.
45) Asymmetric info and restricted usage of price filtering.
46) Mind that even the software evaluating your trading system might fail .
47) Often brokers even try sell things they don't know anything about .
48) Pseudo ... trading instructor, trading coach , trading guru , trading mentor trading rooms.
49) Financial Tv Shows, trading books and their lies.
50) See some movies instead.
51) Why "Anyone can trade" is a lie .
52) How Asset Managers and Hedge Funds lie about "their" algos.
A case study: the stolen algo.
1) Autoscalp price engine logic required so many years of research,
why don't you standardize the process and sell it to retail brokers as your competitors do?
Because is my proprietary speculative instrument , not a publicly available tool. I am a speculator and autoscalp is my tool so I use it , I don't sell it. let me explain it clearly, assume you created a model seldom capable of creating profits: most retailers will try any possible trick to steal your intellectual property and use it themselves rather than allow you use it against them. Then Autoscalp is not a standard broker tool by choice: the less you standardize something the harder is to reverse engineer the process . Also most retailers fit in this frame (sketch above) and that doesn't help.
2) So why don't you run your own customers pool ?
1)Simply put: because We don't buy debts.
Running customer pools means buying debts, not only asset managers but even market-makers fail to get it as most of them have a banking background which forces them to swallow debt chains as main source of income. They buy debts on behalf of their customers hoping to control the net (often leveraged) position against their broker who does same against their bank who does same against the central bank which is NOT obliged to tell them the truth, at all ! Is a foodchain brokers aren't on top of. So Customer pools: no thanks. I like creating algos against them instead.
2) Because I decided for a start up. Most Asset Managers statistically go bankrupt after 3 or 4 years. Which is even faster than broker companies or funds average life. Most often these guys don't create solutions, they exploit customers weaknesses using asymmetric information and broker-side provided tools. For a quick reference , Asset Managers always do the same deadly mistakes , which in detail are :
- they don't invest in expertise,
- they don't invest in technology,
- they don't invest in research,
- they just boost sales to profit from cash flow and broker-side tricks.
Such idea is flawed because makes the asset manager de-facto dependable upon his broker who has specific tools to make the asset manager go bankrupt at the brokers' will : when either the customers pool becomes of an interesting size for him or the total exposure passes a certain threshold. In short, the reason why I decided for a start up is to avoid autoscalp to become part of a short lived vicious circle.
Assume you developed a system that is good at arbitrage or speculation in general, that should make you absolutely strategic for the asset management firm you work with but you would learn the hard way that if you don't bring in new customers, you risk to get fired for "lack of productivity" that would be a nonsense provided that management company was playing fair with their customers but most of times... is not . They care for the profits you produce to them not the ones you produce for their customers. This is why their average life is shorter than the broker they work with. Their business model is flawed and they know so they simply exploit any possible profit source till implosion.
Always keep in mind R&D budget is negatively correlated to used leverage: higher the leverage in a structure, lower the R&D budget.
- If a solution is efficient usually gets stolen by governments.
- If is useless and absurdly expensive is sold to banks.
- If is not working at all is sold to brokers
- If it just creates costs is given to retail customers for free.
What if you have a successful model ? You don't sell it. You lend it. Also you don't advertise it at all.
You don't want your competitors to go head-hunting your R&D team , do you ?
Luckily there is no way to overcome the algo obsolescence . Information lag is your best ally.
When they finished reverse engineering the process is often too late for them to profit with the tool they cloned.
Thus making competitors waste time budget and efforts in late cloning what you do is also a kind of profit as
they learn the hard way that playing fair and hiring your model is far cheaper than stealing it.
Is also wise you don't disclose the research effort beneath, or better : derail and deceive nosy guys toward mock-ups.
Even a mediocre Head hunter is capable of putting in line all men who cooperated with your research
and rebuild it from sketch: from the idea to the algo but again this path takes time and often is an useless effort.
[see the case study: "the stolen algo"].
3) So why do brokers offer their tools to customers instead ?
They don't. They give them the "retail tools" not the "sellside tools".
You may have noticed that Broker tools, are all similar one another because they serve a specific purpose: standardize their trades and exploit counterpart weaknesses. Often those "standard tools"
create pseudo tech-justifications in the customer mind to make customer gear up an higher and higher leverage ... say, the ultimate objective of offering standard tools client-side is to destroy the customer account. In another environment you 'd call those standard tools with a peculiar word: Trojan.
4) Brokers with hidden agendas offering standard tools to customers ,
sounds quite nasty. Can you make me an example please ?
First and foremost: most people can't get the difference between brokers (do their interest) and market makers (bet against them).
You really think that with a two hundred years old candlestick analysis can be of any use nowadays with third grade math charts painted on it?
Or even better , brokers always offer same technical analysis package since 1980s but execution is much faster than chart updating itself. Whenever a price appears on chart is late to click on it. Market is now so fast that charting has become useless. Also because you read a price doesn't mean there is enough volume to support your deal at that price level. In one word: if you see it on screen is late to trade it.
5) Your view seems negative about brokers in general , can you elaborate more ?
Years ago brokers were reliable entities that tried to execute customers orders and will at the best price. Now has become common practice for market makers to name themselves brokers and keep their internal "order pools" anyway. They can and DO profiling, act AGAINST their own customers and ultimately often inject arbitrary in-house created prices in their feed to go fishing stops from customers or delay their executions until price is bad to customers, generally speaking , brokers bait customers, thus I don't like brokerage models at all.
Fortunately if you let one of those self appointed brokers talk more than 10 minutes you get flooded by a gargantuesque stream of incompetence . Most of them are either dud promoters, or retail market makers so they like to talk a lot, lifting the doubt on what their real job is.
6) You mean that often beneath a "broker label" there is a market maker?
Yes but also that often such malpractice is hidden, often orders are sent as STP: a seemingly harmless acronym but it means orders are re-routed to external pools which offers arbitrary quotes not necessarily in line with market. Especially when customers profit. STP means sending orders somewhere outside the market maker company, which doesn't imply the STP receiver is a friendly firm at all.
7) So is all about broker bad practice , is such practice legal ?
It depends: certain jurisdictions have forbidden such practices but the problem is now about routing orders. Assume an European broker routes orders to a third party who belongs to an unregulated country... generally speaking if you don't do B-book (taking positions AGAINST your customers)
and then route deals through stp to a company operating in an unregulated country... n not only you do B-book but B stands for butchering! Sad story it takes years to spot such malpractice and as usual: muppets pay the bill.
8) What you mean by brokers routing orders to external sources ?
Some countries have no market regulation or customer protections, then is even much less likely there is any law against bad practices, at all.
9) How about your view on another hot topic: LEVERAGE ?
By default autoscalp models are not allowed to leverage, as that it is a required step for Shariah Compliance we are trying to reach. But if you are asking instead what is MY personal view on leveraged accounts... well, leverage is just a way for retailers to make you gear up debt to become a victim at the brokers' will . Notice that leverage was once 3 then 10 then 20 now 50 or even 200... leverage has grown parallel to broker-side know how. No coincidence. Mind that most of leveraged accounts get emptied within 90 days from inception. Making it more clear: serious traders and their counterparts don't leverage at all. Leveraging means indebting yourself to bet on a non necessary risk... but be sure
you 'll never see it explained this way on the brokers page.
10) What is the best retailer market platform in your opinion ?
That is another common misunderstanding : that any platform represents market. This is false. When you click on a platform you're convinced to click on the market but too often you click on broker-side created prices and there is also a long chain of big and expensive machines between you and the market. simply put, the more buttons a platform offers, the more is easily to confuse your will with your perceived projection on the market. Higher the buttons higher the possible mistakes. Just because a retail broker offers a powerful platform with more than 200 keys combinations for each order in in any possible market doesn't mean that has been done for you to profit.
On the opposite, specialists got a very limited number of buttons on their platform. Also, specialist platforms have the proper tools to avoid being boxed and spiked by broker created prices. In short, what I look for when evaluating a platform is :
<< does it include specialist tools or just a ton of buttons ? >>
11) What you mean by broker-side created prices ?
Arbitrary pricing. There are platforms that allow brokers put in arbitrary price quotes on OTC market on broker-side price manipulation against customers, fore example google for "virtual dealer" and you get a very precise idea on it...
[ Virtual Dealer is a broker-side package for price manipulation against customers. Interview was resumed after interviewer looked for "Virtual dealer" on google and got informed on it. ] ...This is really shocking .
12)So what about high frequency trading and flash trading platforms ?
That's another misunderstanding .
There is a radical difference between high frequency trading and flash trading. Flash platforms allow light speed execution of extremely high volumes (billions per second). Personally I feel that sooner or later flash platforms will lead to a huge market failure whose early warning announcements are already on everyday newspapers. Instead High Frequency Trading means doing a lot of orders to differentiate risk fragmenting operations on multiple instruments and multiple markets. Newcomers mismatch between high frequency trading and flash trading and that doesn't help.
13) I didn't quite understand the difference between flash trading
and High Frequency Trading . Can you make me a clear example ?
Flash trading or ultra HFT looks like ordinary HFT, scalping.. but is like sheep shearing instead! Problem is that people are paid to execute orders the faster they can so they always suggest you do flash trading but is in their interest, not in yours.
14) I see you have neither facebook nor linkedin page... why ?
Reasons are both quality and Information control .
I don't hire people on web and a linkedin account is mostly a way to drop you hundreds of C.V. per week, it creates the illusion in the senders that their C.V.s are being read which is not. So even just to avoid creating the illusion their curricula were going somewhere , I chose not to open a linkedin account but to outsource public relations instead.
Same for facebook and other social media : which are mostly marketeers' tools to hook their marks.
I don't need those marketeers' tricks , if you mean personally I prefer twitter .
With twitter at least you know you are speaking openly to anyone whoever he is.
Linkedin average user income is 54k $ per year, Facebook average user income is 37k $ year.
Users have a low income and you also never know who you are in communication with .
Most of times you discover is a disguised web marketeer or even a camouflaged counterpart typically a competitor. Since protecting intellectual property is my priority, social media are definitely not the right tools.
Example 1: there's a whole class of social media bloggers who offer themselves to upmarket your status with outright lies, gossip and spin doctor tricks, so I prefer doing good old background checks on managers and companies instead. Checking penal records is always a wise practice .
Example 2: if you compare linkedin facebook results with the real stuff in financial authorities records you might even discover things such as this: a person sentenced for financial scam that resells himself as transparency paladin. Also, you routinely see companies which self advertise as top notch to appear in the FSA banned companies black list instead.
Example 3: Most contacts coming from unsolicited linkedin or facebook like sites are often either fired managers, promoters or people trying to steal some know-how or intellectual properties through social engineering tactics and that is a known fact as we deal with that issue on a daily basis.
Example 4: information control: any autoscalp network member is committed to use our common account on twitter only
so we know which info goes in and out, thing which would be impossible if any of them used a personal accounts in linkedin facebook etc.
15) So, intellectual property is important in autoscalp but if
competitors try to steal your know how? How do you react?
We often let them reach a certain level of intrusion before counter acting.
The most common, and obviously hopeless, attempt to steal us something is the typical social engineering trick to offer us a very cheap hosting contract on very fast colocated machines which hides an industrial espionage attempt from a competitor. When we receive such proposals most often it takes just a few minutes to gather who hides beneath such scheme as the most likely candidate for these naif espionage attempts are a few non neutral (broker-side) trading system grids or boiler rooms such as forex promoters chains . They are the only ones not to know that they are well known for such practices. Broadly speaking , we practically reverse their exploits and social engineering attempts on a daily basis , we also engage in proactive defense against more organized dishonest competitors who are using more sophisticated techniques to bait our staff through "procurement outsourcing" schemes, that is luring people to get away with a part of code or resources etc. for a better job. Apparently it looks the best way to steal engineers and processes but in real life experience, procurement outsourcing is a malpractice very easy to defend from.
Example 1: some CIO hints.
Today Intellectual Property Protection switched from "how to spot competitors making a theft"
to "how can we bait them first?".
We are not discussing legal implications of intellectual property defense , just some counter intelligence methods
typically applied by CIOs in what is clearly not disclosed often but are everyday's practice.
The younger the startup the surest they are learning the hard way:
- bank stealing algos [is the algo owner has to prove they did it ]
- some wrench encryption scares [a bad guy using a wrench to make the algo owner confess master password]
- an usb key left in a restaurant [provided wasn't done on purpose to derail competitors]
...any structure which has intellectual property to cover immediately learns to keep a low profile on its capability
to defend and protect intellectual property and to covertly counter-attack the perpetrators rising invisible walls.
Poor competitors are no threat at all as naive competitors always engage in same naive techniques :
social engineering, business intelligence tactics, competitive intelligence "by the book" which are useless nowadays.
Even more evolved humint tricks such as procurement outsourcing (stealing managers and their know-how) or
captive workforce demand: force the business owner hire from a restricted pool of pre-corrupted prospective employees
have poor results as CIOs are aware of such possible behaviors.Also alienating managers bribing them could be expected
and pre-contractualized against in advance. Whistleblowing or its forbidding can be easily fitted in any contract and
the employee signalling the perpetrator might even receive prizes if CIO policy allows .
So always assume there is a budget range to move within: when someone tries to steal an intellectual property
or a trade secret has to expect to invest at least 15-20% of its value otherwise what he gets is either a mock-up
or probably a trojan or even an outright jail ticket. Because believe it or not,
CIOs have quite fun at creatively baiting thieves.
Also cassical schemes of biased "disclosure... compliance... due diligence..." and the likes are often just mere hooks
in order to force further disclosure of proprietary know-how don't work as are spotted rightaway.
Tech eavesdropping comes in too many flavours to count but what you aren't usually told is that
it has some sense only if counterpart isn't specialized in counter intelligence tactics otherwise is just a waste of time.
CIOs always assume You can intercept any data today, one of their main task is increasing the cost of the
counterpart to gather executable intelligence rather than hiding stuff under the carpet.
Letting counterpart "steal" what is supposed to in order that counterpart I.T. are paid for reaching their goal
without exposing the real intellectual property is being protected. Also shredding in creative ways
things that could be profitable to counterparts is a way to make them lose money rebuilding the real product.
Make the core of the product unintellegible and incomplete or incomprehensible or even plain derailing them.
Counterpart attempts at identity thefts to replace access credentials has proven to be very unsuccesful
even that's not what media say , but media talk about what their marketing needs to sell you the latest
IT defense product, not the truth. The truth is that the average guy being stolen a credit card,
not a company being stolen a secret, as the CEO being stolen the "main password" (if exists)
is an extremely rare event. Stealing the right identity in the right moment to access the proper valid credentials...
in a moment where the core intellectual property is vulnerable is quite unlikely to happen, aside from the cost to do it.
As it implies the credential owners to have done some absurd mistakes meanwhile often those lucky shots are just baits.
One thing CIO and CEO often share instead is the attitude to be extremely revengeful against unfair behaviors instead.
Even reverse profiling a person's life to infiltrate in order to gather information, or to blackmail
works only if the mark has no idea his information is turned against him. Otherwise deceptive behaviors are
the standard not the ecception. Devil's in detail in these cases and those details make the mark aware
of being under surveillance, be it humint,comint,elint whatever then the mark starts making it up rending
the whole intelligence effort not only worthless but misleading also.
A wise mark never discloses the awareness of being under surveillance and that's a fact.
Thus diversions such as : false meetings, false approvals , false appraisals, false certifications, claque clapping,
pseudo VIP meetings, political or reilgious belief exploitations, pseudo research tokens and bogus compliance prizes
or even biased statistics evaluations, are all expedients which fail if the mark expects them.
Socially engaging marks, personally or socially, force them react is another unilateral illusion from who
has never been engaged in counter-int ops to cover intellectual property. You can't expect one to
show you what he is hiding as he is trained to continue business as usual while calling in the CIO's team
which runs covertly the matter to adjust the situation before the intellectual property theft is accomplished.
Capturing intermediaries through bribing or information sniffing, infiltrating company personnel...,
covert blackmailing to extort informations are all known and expected techniques and every CIO specialist
knows how to train his partners against.
Some years ago it was in fashion litterally the physical replacing of proprietary hardware to clone
some highly sensitive or valuable infos, which on and off had some results but nowadays is far more complex.
As any physical data support self encrypts as any manomission attempt is not spotted but just suspected or
triggered given certain behavioral thresholds are met.
Typical low tech ideas of getting employee drunk to make "big mouth sink big ships" worked in the past
while today are useless if not ridiculous as who is in the knows is trained not just to fight against information leaks
but also to reversely exploit tricks such as blackmailing or even honey traps of every sort.
In short any known technique triggers a predefined decisional path as reaction, not just an alert
reversing the chessboard against the thief himself often with the result of exposing such behavior.
Any person having access to restricted know-how are also trained to lie at a professional level
to compartimentalize any piece of sensible information or even derail and deceive
and often any valuable information is deceived while compartimentalizing it and that's by design.
Taking the intellectual property owner by surprise is a truly harsh task as all of his partners are aware of such risk.
Also thieves looking for the weak ring in the chain often exposes the one wisely -choosen- to appear the weak one.
So, there is no added value in provoking a behavior if that is expected by the mark.
Always mind that if a thief is spotted he is often for sale, for both sides.
Tarball building or social profiling or even the true metadata building
composoed by sniffing packet trafic, sniffing accounts, intercepting reports , linkedin contacts or tweets...
are most often counter-bait hooks since beginning and that's by design of any wise CIO disinfo campaign.
From a CIO point of view: social media are a way to make thieves expose themselves, not marketing tools.
CIO exploitation of reverse profiling of competitors is very diffrent from the marketing guy's one.
Also targeted issues against the mark, ranging from outright lies to public defamatory intents
or misappropriation of thir party works often revert against the thief
as if the thief tries to resell something must prove he has the preparatory work for that
not only the nominal intellectual property and faking research results in a hurry isn't that easy.
Also a wise C.I.O. might corredate a ready on the shelf stealable research with fake scientific results,
not by chance and fun happens when the thief tries to prove ownership.
P.R. techniques from rumor milling to red herring only work if public opinion can be esily biased
which often is but P.R. is a double sword as makes the perpetrator vulnerable to reactions as is exposing himself.
If CIO knows his job the perpetrator is victim of a counter-campaign too at the end of the loop
and that happens by design, not by chance.
Of course all classical guerrilla marketing techniques you can read on any marketing
manual are expected and debunked before having any actionable intelligence produced.
Burning a person's cover while a person is infiltraring is a standard technique and applied
far often than one might think so much so that: "exposing the thief while stealing"
has become a standard procedure nowadays.
The real fun stuff comes when everyone who has got stolen gets aware who was the thief.
As if reward is single sided, punishment can be at calss-action level instead.
stealing someone's intellectual property in the 2020s is not so easy as it was in 1990s.
as a proper counterintelligence strategy baits thieves at every level , engaging them first.
That makes an active Intellectual property protection strategy a winning move,
but all that is far from being either innocent or cheap.
Example 2: How you discern if those algos are original or if that person is a true partner of Autoscalp .
The Guidelines to spot a Fake Autoscalp Algo are very simple:
Any site o promoter saying they can offer you our algos: is lying .
We have many ways to spot fake algos or fake representatives...
using our names on their tools or trying re-brand what we do as theirs etc.
a) We don't disclose who we work with (NDA+NCA),
if they say they work with us, they are lying.
b) We have no promoters,
so if they say they work with us, they are lying.
c) We don't disclose our algos,
so if they say can show it to you, such algo is not ours.
d) If you can see the algo running on any platform or on any algo-grid,
is definitely not ours as we route the algo feed only and allow no external access to them.
e) We don't use external website as antennas, so if is not on this site expect is NOT us.
f) We use no facebook or linkedin thus if advertised there such algo is NOT ours.
Social medias are definitely not the right tools to manage Intellectual Property, as explained in answer 14.
g) Any site quoting us cannot bear a license as our license (NDA+NCA) explicitely forbids such disclosure.
h) Any site offering you our algos, is lying as our license (NDA+NCA) explicitely forbids such practice.
i) Any site trying resell you our algos, is lying as our license (NDA+NCA) explicitely forbids such practice.
Example 3: Case study: Why don't we focus on bots as every trader does ?
Because bots are just ... bots.
To focus: when you think an algo you think this:
When we think algos we think this:
The image shows an automated chain of a generic car production company.
Note: in a 4 football fields wide car production company there is just one human,
in 1980s were 1500 people working there
The production chain works in such a way that: robots control ... robots and when a robot breaks up
gets replaced by other robots... who repair the robot and reorganize the production chain accordingly.
Aside from the social implications such as "you maybe just don't want people to know that":
The question to focus is: what is the human doing there ?!?
1) He is forbidden to access the robots (would just create problems).
2) He has no direct contact with the production chain (he would just slow it)
3) He has no idea how cars are built at all ( manual work is not his job)
He is there as I.T. Security : to make sure no one steals the algos.
Because, being the robots a generic model anyone eventually stealing the algos
would steal the whole firm. His job is make sure thieves just steal... mockups and trojans.
Now you know why, for us, robots are not important: algos are .
16) Can you please explain an example on how you defend from
procurement outsourcing or other intellectual property theft attempts?
Yes, if you are aware that often your competitors engage in such schemes, for example, you always put low level executives names on front end papers and when a competitor think he's stealing you some core people, such an experienced trader, is more likely what he steals is a webmaster. As he will be offered a very high salary from the counterpart for re-building the profitable algos he allegedly has taken with him when leaving us , he'll never resign from such chair. Fun enough, the likely outcome is that he will destroy our competitors' trading revenues with his over paid incompetence. Trying steal us people, know-how, algos etc. is not so easy as it seems. We have plenty of resources dedicated to pro-active intellectual property defense.
Some intellectual property defense tactics examples: We call it switching to "hello quota" mode:
We wait till the thief passes a certain threshold,
then make the thieves become a source of income rather than a cost.
Example 1: We put some algos "almost" on display so when dishonest competitors ask a hacker to steal it,
the hacker gets paid for stealing that algo (is not his fault if the algo doesn't work).
We are happy, Hacker is happy, employer could go bankrupt... but that's secondary for us: hello quota.
Example 2: "someone reselling the stolen algo can be used as market without knowing":
As in certain times the algo is likely to be putting up liquidity on the WRONG side
and nothing is more compelling to traders than a broken market to sip liquidity from.
Believe it or not the counterpart liquidity pool get merciless sucked up till the last drop at light speed:
Example 3: "Retaillation is a sport whose participants are unaware of competing":
but if is a sport it means is bound to get bets on it (direct and indirect ones) and every bet has a market.
Thus , the more market participants think are smart, stealing the algo and using it for their own profit,
the more they act like one entity and the surest their loss. Assume is not so vanilla as it seems:
a stolen algo might work flawlessly 99% of time but that 1% of times it says "who is your daddy"
to the thief and whacks his whole account in an incredibly small period of time: hello quota.
Example 4: "transform a passive intellectual property theft attempt gone dud... in a sales channel":
after a few months the moment of reckoning comes: the thief running the algo admits
it doesn't work as expected: he wasted budget, wasted time, wasted resources...
as the loss typically reaches 6 zeroes and cold sweating becomes the norm,
panic reaches the boiling point: "F***.... This way We go nowhere, let's give 'em a call !"
which We answer as : hello quota!
17) ...but if you outsource development, isn't that dangerous
from a proprietary know-how point of view?
We have many concurrent ideas in various development stages, even if when one turns out to be unprofitable is always taken to the final stage. The dud ones play the front-end mockup role, that is to derail infiltrates and lead them to dead ends or wrong conclusions, in a word: deception.
Reason is simple: the harder is for a competitor to spot the right haystack in which to look for the right needle, the less cost he would have had if he played fair instead. We are extremely efficient in splitting development into very tiny tasks which are eventually outsourced to different independent companies so that only the project leader knows what the final model will look like. Worth noticing the project leader rotates too when his task is closed, that fragments information even further.
Assume a coder is trying to reverse engineer one of our models from the portion he was working on: he would get just a lot of garbage as the core part is never outsourced. Provided that the part he was working on was a profitable one because that is not a fact , is his assumption and the difference can be staggering. Assume a decent trading system today is around 25000 lines of code, just changing one of the external environment variables,not even the internal ones, radically changes the output.
Broadly speaking, we apply many intellectual property defense tactics... including many pro-active ones, most of which are unseen to an untrained eye. I must stress this: Intellectual property is our core value and must be protected at all costs >>.
End of first part of interview with the autoscalp creator, Mr Matteo Gandola.
in part 2 interviewer requested also the autoscalp team to join a shared open discussion about the common mistakes in trading and automated trading and autoscalp team network members promptly answered online from their various locations around the globe. The interviewer then collected all answers and ordered such concepts by topics to make the reader an easier understanding of the key points that were discussed.
The Interviewer did a summary , a synthesis by topic and asked to add examples where possible.
While developing the main autoscalp engine and some of its modules and strategies we got in contact with the very deep logic of a price engine and found numerous issues which could be just common sense misunderstandings, lies, malpractices or even just bad data interpretation but all ultimately is concurring to confuse or derail trader minds, what follows is mostly derived from our personal experience on forex and indexes.
So let's go through the most common trading lies and mistakes we found...
[Interviewer did a summary , synthesis by topic]
18) Fundamental Analysis :
it is based on theoretical concepts that have evolved since 1929 to justify crashes (backwards). If applied to forex market it did not even work in simulations since 1960s. It lost coherence as most of the "static analysis" model assumptions got obsolete faster and faster. So go figure out if they have ANY sense in a world where trading speed doubles every 6 months.
Example 1: self-referential and second guessing .
Fundamental analysis is too often done to justify political choices not to inform on market status. Some states are not quite transparent , as it looks from their balances hence it happens that on and off their major variables are simply artificially inflated or deflated for political targets to such an extent that their currency rate and their stock exchange indexes simply don't follow those data at all. Reason is that such data is not computed but politically cast and derived from its own objectives too.
Example 2: M3 and M4 releases are purely "nominal" and used to trigger sentiment driven market response rather than explaining the real liquidity of the State publishing them.
Example 3: fundamental analysis implies a wrong assumption: same political move brings same economic results, which is wrong.
If a central bank promises an intervention too many times there is a point where market doesn't react at all at such announcements.
Example 4: reaction times changes depending on operators' size hence same news can be bought and sold at different speeds making a fundamental data release completely useless if not dangerous from a self-confidence point of view.
19) OHLC candlesticks : a broker side convention passed as a customer convenience.
The real reason people still uses OHLC models is because they are given such tools for free from their counterparts (banks and brokers). This alone should make one think twice before using such tools. Also you can decide a time frame but forex and indexes don't care about time frames at all . Those instruments don't expire . The only OHLC candles fair usage left today is in option trading because they carry an expiry date and hence time based analysis has some sense. Banks, brokers and market makers exploit this asymmetric information thoroughly giving final users OHLC based analysis tools for free which they know to be ill fated from the roots. between the time passing and an instrument price there is an almost perfect statistical independence rather than a relationship. OHLC analysis tools are also free because your counterpart wants you to use them as the most likely output of such "analysis" is a coin flip. The more people do coin flips, the better for the brokers who hedge the spread among them... Now think is OHLC candlesticks a "convention" or a "convenience" ? This is also why we have then created a proprietary alternative to the common "OHLC candlesticks". We called our tool "binary pricing model" which is not OHLC related and is not time based at all.
Demonstration: you can debunk OHLC candles yourself. Just reverse the relationship between time and price and you see such relationship does not exist at all: OHLC candles represents time on the X axis ,that is how many kilometers Earth moved around the Sun : in one minute it moves 1800km . Now reverse the relationship: Is eurusd price sensitive to how many kilometers did Earth moved in the last minute ? NO ! Time price chart does not represent a function just a convention. So what is the real relationship between price and time ? Perfect statistical independence. Yet you see the price/time charts on almost any technical analysis models based on OHLC candlesticks .
Example 1 : eurusd doesn't expire at all , no reason to analyze it in "time slices" . Also you arbitrarily choose your time slice... another mistake. You decide "a time slice exist" and even choose one of them... but still eurusd price doesn't care about that choice (see demonstration above).
Example 2: if you trade an index , your stops are triggered on price levels , why do you execute signals based on the time passing instead ? The broker side hidden reasons and moral hazard in the above statement is of such enormous size all brokers refuse to answer all stating that putting time on charts is a "convention" but they should say "convenience" instead.
in OHLC (Open high low close) candles you can decide the time-frame but this means you also unilaterally decide to add the open and close values to the price feed depending on the frame you choose .Then is you who paint the candles colors to be red or blue not the price feed.
Then you can have a fast frame blue candle or you can have a slow frame red candle...while price did not move a tick at all. Result is you add 2 unilateral variables (open, close) and then build the OHLC candle.
20) Technical analysis:
biggest and most hidden problem is that indicators cross one another for the PASSING OF TIME not for price action reasons for the above quoted "OHLC convention" .
Example 1: on all standard charts have you see the X axis indicating time, perfectly in line with the "OHLC convention" above and thus making any further analysis (indicators, oscillators etc.) totally useless .
Example 2: even if a price feed is blocked and sending no prices, charts still update part of its data set and as time passes : price stays steady but time moves and then... indicators cross themselves, giving you a pure "algebra" driven signal which has nothing to do with market price at all.
Example 3: all technical analysis myths are disinfo by the book : not only pushes traders to the dangerous game of forecasting
(which typically ends in losing their unilateral bets against reality) but what technical analysis also pushes traders to forecast are
the wrong datapoints:
Most of guys in the wall Street guessing game always try to forecast "today new york closing price" as if it were a kind of Graal .
Meanwhile the ones in the knows are perfectly aware that is much easier to forecast the OPENING price instead,
(and set their premarket trades accordingly so that when NY market opens their trading is already ended and positions are squared)
...but you are not supposed to know... and the reason is so simple is disarming: because if you know... is UNFAIR for banks.
Since they seldom trade on premarkets... they want most of muppets be exposed so can flash all muppets stops
to position themselves on the right side.
Conclusion: technical analysis is a proper counter-information tool knowingly biased and working for "the ones in the knows".
21) Mark to market :
just because market today closes at a certain price, it doesn't mean there will be enough liquidity to close your orders at such price tomorrow. If an important news is about to be released your clever mark to market price is totally useless and just confusing as real price moves away from yours. Real market price is where bigger orders are concentrated , not mark to market.
Example 1: What if you had to close a deal and if your position were heavy ? You would discover your real match price to be at the most robust level, not at the price you are seeing on the platform.
Example 2: If you priced your eurusd position "mark to market" yesterday evening, such price doesn't make any sense ahead of a Non farm payrolls release today. Then your mark to market price is wrong today because was computed with ... the OHLC convention quoted above.
22) News release deal entrapment and analyst advice traps:
the few who know the news act earlier than you do ... you often see "informed" price movement ahead of important releases.
When the news release happens the real thing that you see on screens is market dismantling any position structure which blocks its way. Is in the very nature of the market to swallow latecomers positions and chew the news traders. Is all about asymmetric information. Banks and Brokers suggesting customers to trade the news are not very transparent on the concept above.
Example 1: liquidity traps .
Liquidity disappears just before news so you can't use the news at all , if you try to beat the market your stops get triggered at light speed.
a technical analyst takes 2 days to come up with a decent paper, a fundamental analyst takes 2 weeks to get its report , while a Keynesian takes one month to produce its analytic tome. That would be alright if the information they were trying to analyze didn't evaporate in the 15 minutes BEFORE it was even released. So while reading those papers always mind these guys are paid either being right or wrong but they are paid more if they bring in customers.
That is: most of analysts are disguised marketeers.
23) Charts always lie :
Because charts can't account news release.
Because charts update when time passes not because market moves.
Because they don't take information from Order book.
Because if you see it on chart is late to use it.
Example1: linear regression indicator is 107 years old... you really want use such thing in the HFT era ?
Example2: if any broker offers free charting packages, just imagine why he gives you, his counterpart, such a present. In other environment such "charting package" would be called Trojan .
24) "tight spreads" , false friend:
Spread should represent a difference between bid level and ask levels instead it shows a synthetic bid and ask aka: "instant price" showing that bid and ask are close one another. Which is just an artifice.
Example: during a price spike , market is where the spread is . Distance from buyers and sellers can be wide, as they can be concentrated on different price levels while brokers still they sell an "artificial" fixed spread to... bait customers' orders and their stops.
25) Leverage is the hang up game:
Leverage is an unilaterally exploitable broker-side hang up game. No matter the platform, you just got a very partial view of how a leveraged account runs while, on broker-side, a myriad of broker-side tricks exists with the sole purpose to transform an ordinary speculation into an asymmetric betting game which always put traders in the gambler's fallacy casino problem. With today's broker-side tools for price manipulation the sad untold reality is that as soon as you start increasing leverage, account evaporates. This is not related on which model or strategy you use at all. Is the leverage model which is a scam and flawed by design.
The more you are on the left of this chart... the faster your account is wiped out
by the greedy counterpart on your right. That is the real meaning of leverage:
while you might think you use leverage to invest,
your counterpart perceives you are buying a higher and higher debt.
The higher the leverage, the faster they hang you up with your own money.
Notably also, the reciprocal of average account life is exactly the sector concentration:
the lower the leverage used the more important is the player. No coincidence.
Example 1: if in a month you loose 50% of your account at leverage 20 , just to recover you have to do +100 % probably using a higher leverage.
That means you will likely destroy your account just trying to go break-even. That is often not very clearly explained... as it means a sad fact: once you start losing using leverage, to go back at break even you need to double the leverage and more than double the risk... again and again but the profit ? Not a penny. While the loss : the whole account or worse.
Example 2: what happens if customers don't get their account scooped up in a few months ?
Broker changes its trade engine and replaces it with a more profitable one. While your strategy or market knowledge stays the same. Sadly, they don't advertise this on the media.
Example 3: always assume most leveraged accounts get emptied within a few months from inception.
Independently from the strategy used , the broker used or even the skills one can develop on market. Most of times disasters comes from using leverage.
We use this simple formula to explain this little known point to our peers:
- at leverage 5 expect lose all money in 24 months
- at leverage 10 expect lose all money in 12 months
- at leverage 20 expect lose all money in 6 months
Example 4:There are others kind of leverage, indirect leverages for example.
"Collateral transformation" being the most common, often proposed by custodians or brokers. The guy asks you if you want to "grow faster" proposing you a collateral transformation through which you hire the right to sell 4 apples depositing just your fresh banana... problem is that the four apples have gone off (toxic) but when you discover that is too late and your banana is lost too, that is nothing more than a bucket shop scam just re-branded as "collateral transformation".
Example 5: the muppets butcher house chain: instruments used broker-side to trip customers most hardened stops:
kamikaze (leverage 400) are used to nail CFD (leverage 200) to nail Futures (leverage 20) to nail options (leverage 10).
Delivery (leverage 1) can't be nailed, this is why they don't sell it.
As it just takes a milliseconds to snipe a stop... tripping stops apparently far away from market price looks expensive but it is not,
where there are no orders it can cost dimes and produce millions, broker-side obviously.
They know how muppets are to react, where they are regrouping, their propension to counter-snipe..etc.
Just the most informed counterpart can win at this game. To make it clear muppets are the dinner.
If you noticed leverage is working reversed, to create profit broker-side and is not coincidence.
Example 6: a typical retail trade engine setting:
|month result:||trade engine behavior:|
|30% loss||fake assistance / help / offer courses / guidance etc. to avoid account closure from panic (avoid awareness).|
|20% loss||just some slight help to create confidence as the muppet (you) is doing his job: self-destroy the account|
|10% loss||typical situation. No intervention needed: illusion that your skills are improving and you can "turn profitable"|
|0%||you receive many warnings promotions, invite to roadshows,webminars,courses: suggesting to trade more|
|5% profit||you receive a dedicated account manager whose task is make impossible for you to profit in the long term|
|10% profit||you account starts receiving "strange" execution spreads,lags, deal suspensions, maintenance down-times...|
|15% profit||stop picking /account suspension / legal tricks / abrupt margin calls / creative vengeful ideas (*)|
(*) mostly waivers: "unfortunately is not broker's fault if happened ...[force majeure list of plausible excuses] and you received the margin call" .
Expect a couple of unilaterally (broker-side) exploitable news events a week, also common is the exploit the Friday close / Monday gap.
account manager's job is to manage the transfer of your account from you to the broker , not the opposite.
Assume the price engine to be hostile and the account manager NOT to be your friend but a broker's employee
Example 7: "Most hedge funds fail: their average life span is about five years". Source: Financial Times
Autoscalp prefers not to comment that extract as it speaks by itself.
26) High watermark:
apparently ethical for customers in reality it works against them. As it forces traders to work for "free" when their profits are not growing. That creates adverse selection: good traders don't accept to work for free. So if a trader accepts it you can be reasonably sure he is to loose the customer's principal.
A typical disaster occurs when a Fund does an unexpected -25% while its Statute was expressly forbidding managers to lose more than a certain threshold and the Fund abruptly find itself both without proper management to handle the crisis and without rational chances to recover. Unfortunately most of times Funds have to learn the hard way that high quality managers are not waiting in line to enter a structure which left 25% on the ground, whatever their statute says.
This happens because most funds are created with a logic flaw: they try to forecast ROI and ROE projections on expected sales for compliance reasons, rather than create scenarios based on management stress behaviors (Fund statutes rarely include exit strategies for "panic scenarios" ) which in fast markets are the norm rather than the exception. Most of times when a stressed manager is renewed the trust, he doubles the leverage and completely destroys the accounts especially margin ones . High watermark plays an infamous role in this because even if someone could enter the Fund with the will and capabilities to manage the crisis but had to start with a -25% to recover, is quite impossible he will get any performances out of it as customers come first and he would get performances just producing more than 25% or even worse you see a completely incompetent person faking an uncommon skill accepting such responsibility as he has no idea of the real burden he is to carry. In both cases the outcome is disastrous.
If high watermark is part of the core structure of a Fund, to us the Fund is badly designed and bound to create greater losses and doomed since inception.
Usually we consider this simple base assumption: the more a Fund is client friendly and promoter friendly , the more hostile to managers, the more is likely to crater abruptly.
The best solution would be banning high watermark since start as trying to remove it while the Fund is running would make customers unhappy, thus forcing all stakeholders to rethink what is the purpose of the Fund, if any.
27) Black-Scholes model:
is a model taken from the gas physics and badly adapted to finance as implies "random distribution".
Example: how a gas randomly distributes into a room , is not necessarily related to how volatility behaves , being the volatility often unilaterally manipulated in an extremely speculative manner through a variety of tricks, the most popular of them being: asymmetric information and price lag on news release.
28) Market transparency:
Banks don't disclose where they are positioned by statute but banks and brokers can see and profile your trades and act against you. They also can change the rules without informing you as you probably signed a contract that allows arbitration against you and modifications on their side without noticing.
Example 1: outsourcing the control of ECN pricing to a third party entity which is in a country that has no market regulations at all.
Example 2: Special orders.
Assume each ordinary order is a text string with tags, with a syntax similar to html. Strings carry all infos to match the order in the exchange: index name, size, limit price etc. Special orders tag are small tags, almost invisible, put inside that message that give an ABSOLUTE precedence on all other orders . Assume a new tag or "feature" is developed for a certain exchange and that if you add this short tag to your order messages <TTL>2</TTL> it means that if your order is not executed within 2 milliseconds is canceled. Since the exchange objective is to match the highest number of orders possible,this order jumps all other orders queue and is executed first. If you ask ten thousand market players to raise their hand who knows about things like this TTL tag it comes out just a dozen do. Most often are the guys who developed such "feature" for that exchange.
[see the documentary "the wall street code" where this is explained very clearly].
Example 3: price apparent univocity
Standard trading platforms are designed to get you enough information just to pay banks their interest for the time passing rather than allowing you to really track the market. Which means if they convince you to look at a chart that you configured yourself, they profit for sure. You can't and... is by design.
Most of times you think your trading model has one user. That's what you see one user one password. Meanwhile unless you answer all questions below you have no idea if price is true or not.
There is a whole class of hidden users in providing you the price feed
So before telling a price is true or false must know:
Who is the real user
Who is the apparent user
Who is the hidden user
Who is the promoter
Who is the informer
Who is the provider
Who is the sponsor
Who is the debunker
Who is the slave paying for it
Who is the master cashing in
Who is the thief squeezing them
who is the market rerouting money from
who is the market rerouting money to
who is the darkpool lurking in between
who is the broker creating your pool
who is the market maker matching which orders
who is the owner of the unmatched orders (error account)
who is the carrier lagging which feed
who is the price engine concentrating which data
who is deciding the level of hositlity of the price engine settings
who are the guys building which feed
who are the guys reselling which feed
who are the guys obstaculating which feed
who are the guys lagging which feed
who are the guys supervisioning which feed
... believe it or not, those users never coincide and everyone has his preferred price to suggest you trade at.
So the comfortable red and blue square you click onto is not representing
the real price at all: is lagged , flawed and arbitrarily injected 100% but its "dressing" is so standardized you don't even notice that.
Also you are NOT given the tools to question it, that is also by design.
So, the more you self-convince the source is fair without checking, the surest price feed is flawed instead .
Clearly, any subsequent analysis is just useless since dataset is tainted.
Example 4: exotic places, exotic banks, centralized scams.
To get which countries... just list all countries capital by concentration of banks respect to inhabitants, keep the top 20.
In all these countries banks form their employees to look customer friendly while what they are really
committed and motivated for is just to empty deposits on an industrial scale. Banks provide their employees specific
marketing courses to cover what their banks are really doing. When engaged at a business to business level
often don't even fake denial and explicitly admit their true purpose which is emptying the customer accounts
in any possible way. While in a healthy market: speculator is king, bank just executes,
in these countries the banking system is so deeply corrupt and corruption is rooted at legislative level so much
they succesfully reversed the market paradigm to fit their criminal banking cartel in a perfectly lawful way,
allowing banks to keep customers at bay, while emptying their accounts through legalized commission chains and
keeping customers in the dark of their true intents (typically banks there are just engineered ponzis or bucket shops).
Their "Asset Management Branch" is often just a perfectly legal bank compliant cover-up, to hide their purpose:
whacking customers (aka muppets) accounts.
Those banks don't care at all about Strategies, Algos,Risk and the like they call it compliant if they profit
non compliant if they dont. Since their real objective is to shred muppets accounts by any mean:
typically inflating commissions on soft dollar based schemes.
When all goes south the bank typically blackmails the customer
either he accepts a draconian reduction on its capital or will be signalled to his home country financial police.
This paradigm has been running for a century. Is so standardized there are companies selling courses
to properly train bank clerks in that subject with "special courses" which should be named:
"how to keep muppets at bay and prevent whistleblowing and get a bonus on that too"
but wouldn't be that great for marketing purposes... would it ?
29) Broker fees - Software fees:
Since inter-bank market pays you to put orders into it , why you have to pay to access it ?
Example: brokers ask you to pay a fee for accessing market instead so it should be called "usage of software fee" not brokerage cost. While brokers get both: volume fees from inter-bank and software fees from you, You even buy the risk in between. To insure they don't lose.
For same reason, brokers call their customers traders rather than customers. So customers have the illusion of being a dealing party rather than a final retail user, .The illusion to be dealing with the broker rather than buying his product is very powerful , so customers think they are partners, not targets. They think they are partners while they are just marks. If you ever worked in a broker firm you call the customer a "trader" during the cold call but when he hanged up you call him "muppet".
30) Manual trading:
Your order gets routed at least in 5 different machines without any human counterpart before finding a match
Example: why brokers avoid explaining is you vs machine ?
You would NOT put orders manually at all if you knew your counterpart were a 3 story wide hyper fast machine hardened in a server farm bunker. Would you ? So they try make it look "human"... after all no one likes playing "Kasparov vs deep blue" if is not strictly necessary.
31) Why Traders can't profit with manual trading:
78% of leveraged accounts is emptied within 6 months from their opening. Enough said.
Example 1: this is why brokers use "push & pull" fast sales : to avoid customer awareness before is too late.
Example 2: vertical sniping : "Scheduled Maintenance" warnings appear on platform: on some customers, (not all of them) , when their position is flattened. Those customers might say, no problem I am flat, I can wait... while other accounts are not aware they are forbidden from entering market and thus broker-side position book is getting thinner and thinner, at a certain point broker-side engines can crush the remaining accounts in position at no cost spiking out the remaining positions inside the pool .
Example 3: What a market maker advertises vs real message beneath it.
|apparent message||real message|
|your main problem is self discipline||your main problem is leverage|
|your main problem is trading system||your main problem is leverage|
|you can manage leverage||you don't need leverage but we don't want you to get it|
|improve your system||trade more, loose all before you learn something|
|use multiple instruments||dilute concentration so you don't learn a specific know-how|
|use time based charts||ignore market dynamics for same reason|
|use exotic instruments||get spiked out faster and at the minimum cost for us|
|how to trade the news||you can't trade the news we give you lagged prices|
|use stops, limits, complex orders||so we can profile and snipe your positions better|
|...our expert says||if he were a true expert wouldn't work for a fixed wage there|
|...our strategist says||if he were a strategist would be forbidden from disclosing|
|we are brokers..we trade with you||they are market makers (taking positions against you)|
|learn from track record||create yourself the illusion that past performances will repeat|
|learn from others||if they were right they would not talk but run their fund|
..."never talk about the fight club" also applies.
Example 4: how a bank screens traders :
most investors think "big banks names hire the best traders". Wrong. Just ask a HRM guy to debunk that myth
and expect great laughs when coming to details. Want to know what a big bank looks for ? Here it is:
|Bank HRM requests :||what they really mean|
|a nice looking Phd fully cased and chiseled||no market expertise|
|with a self-contained ego, cooperative in teams||no personal initiative|
|smart, cheerful, brilliant, elegant looking, good P.R. skills||not outstanding personality|
|some robust publication history as background||no operations capabilities required|
|a sturdy sense of belonging , complicity and sharing||no leadership|
|oriented to client relations||not oriented to profits|
|deep knowledge of all financial instruments||not specialized on any instrument|
|attitude to rise doubts in front of unexpected situations||not a solution provider|
|better a don't ask don't tell attitude rather than improvising||no risk taking attitude|
|one who avoids setting up his little kingdom of power||no organizational capability|
|with a strong know-how on disaster management||no customer empowering skills|
|who isn't shy of asking senior advice when has ethics conflict||no whistle-blowers|
|clean personal background (no bankruptcy etc.)||no market experience|
Given the reasons above now try ask yourself : does the bank really care if the "trader" is capable of handling
a hostile market in a turmoil situation with a complex exposure ? No way.
Conclusion: if the guy is good or not trading the market is completely irrelevant for the Bank who hires him
as his real task is not to profit but to mantain customers at bay.
Now, just think the other way round : what would be a a good trader doing in a bank ?
That is maybe the right question to ask but is the one that the HRM guy will never answer
as he is paid from the bank... not to disclose but most important the HRM guy is there to make sure
you are not aware that there are alternatives and "what is good for the bank is good for you".
Example 2: trading competitions.
You win if you take a risk so absurd to beat the top 10% of wise traders.
and this alone would explain why they are useless.
Since is in fashion that some "not so wise" head hunters try to use trading competitions as pools
adverse selection applies. The winners are not the top traders just the wildest.
Also an extract on Forbes surfaced that considers trading competitions a valid source of screening
when some "professors" (who typically never traded in their life) pick brains for big companies.
In our opinion flaws in evaluating trading competitions are so huge that make the competition useless.
We found mistakes such as:
1) <<Eenergy companies don’t want to hire riverboat gamblers>>
2) <<prospective employers are not looking to hire speculators
but people who can reduce their risk>>
These misunderstanding are so huge that we dedicated a whole think-tank on it.
Since banks evaluate risk from their point of view which assumes borrowing
money to be "necessary" while it is not and mitigating risk which implies
avoid taking profits as a mission as bank's job is to keep customers at bay, not make them profit.
If you judge trading competitions with banking criteria you are doing nonsense since the very beginning.
Since western banks are biased to judge anyone not borrowing money not to be worth investing on
(compare that to mudaraba which instead widens the funding if project profits reduces funding if it loses WITHOUT indebting the entrepreneur in between) .
Thus any trader being evaluated with western banking criteria rather than market point of view... is a gambler.
In short: anyone not in line with "buying debt" and "keeping customers at bay" is a gambler.
Just for a reminder the worst thing to do to a bank is "crowdfunding"
not because your idea is silly or risky but because the bank doesn't get a penny out of it.
Thus until evaluation methods don't get in the post flash crash era too...
trading competitions are nonsense.
Also worth noticing that if you "use speculative instruments to reduce risk"
you are a speculator, no matter how you justify it to the risk manager,
also being evaluated how you mitigate risk on a ... speculative instrument is
so absurd that just a professor that never traded in his life might not see the flaw:
Don't want take risks ? Stay off speculative instruments and that's it.
32) Segregated account:
Bank is held responsible for mismanagement but not for broker bankruptcy or market failure...
Example: many big names on the newspaper today who "failed" to fill collateral properly and their customers went boom, not them. Also in certain jurisdictions some accounts are insured for a nominal value not the entire account value in case of broker bankruptcy.
33) STP guarantees the customer doesn't receive in-house brewed artificial price:
but says nothing on the counterpart (kind of risk waiver).
Example: STP means order is routed "as is " outside the broker but doesn't guarantee the customer that the counterpart the orders are routed to is transparent, at all .
34) Market price and broker price:
Always assume a broker is your counterpart, not the market.
Example 1: in broker contracts you must sign a clause where you expressly allow him to position himself and make money against your positions (and specifically you allow a conflict of interest) so he is allowed to give you any price he believes fair, not necessarily market price.
Example 2: the six S of broker-side pricing: sniping, skewing,slippage,spreading,spiking.spoofing.
This is how market makers make money today. Thus being price manipulated against, for example, sentiment driven traders ,
they can forget profiting using any trend following model: whatever the code, the platform, the market, the instrument.
Price is unilaterally manipulated before reaching you, not after you see it.
If a trader is cashing in a profit he doesn't look at the tiny fraction of pip stolen to him so watchdog can't say nothing and banks cash in. That is not a secondary issue because this trick alone accounts for trillions of deals per day where that tiny "undisputed" percentage goes directly into bank safes.
This happens either because Watchdog can't track if every single deal closed at fair price (imagine the cost of that kind of supervision) and because a trader should signal such behavior to the watchdog, with the only result that the profit he was cashing in has to be frozen until dispute is resolved... and also because customers inadvertedly sign a clause that allow the whole chain of price makers to post him the price they think fair. So if you have : broker ,prime broker, market maker... and whoever up the order pipe making the price on your order close, be sure someone is cashing in your pip fraction.
This trick alone speaks trillions a day but it seems no one complained about as this was assumed to be a a structural issue in the market flow but is exactly where HFT comes in. Also this explains why banks don't want HFT : they run the risk to lose the tiny pip benefits they have been cashing in for decades.
On and off you see a market neutral strategy advertised as top-notch panacea while being it applied through long chain of intermediaries such as a market maker a broker a prime broker and a bank before your orders reach a match. What is believed to be a market neutral strategy is just a contradiction in terms, since none of those counterpart is neutral at all.
explain oil price manipulation process why standard centralized oil pricing fails and where (example Futures) :
The basics of oil price manipulation are quite simple, assume oil price is:
q is manipulated by retail brokers / stock market / base currency micro fluctuation
z is manipulated by market makers or clearing houses / secondary news Releases
y is manipulated by central bank / OPEC / Rating Agencies / major news releases
x is manipulated by multiple government pressures or OPEC important decisions
k is manipulated by UN / world powers balances shifts
Examples of price manipulations:
k passed from 1 to 0 for Russia-Ukraine uncertainty
x passed from 3 to 6 for First Gulf War
y passed from 4 to 7 on an OPEC release
...q passes from ? to ? with no reason as is arbitrarily home brewed broker-side on their own advice and unilateral convenience.
In none of those manipulations is included a real will to buy and sell oil barrels with the due reciprocal trust among counterparts a true market would require to work properly.
Oil is perceived as a currency and has same kind of manipulations currencies run into.
Thus CL price is not resembling true market situation just order netting of some unbiased orders mostly systematically harvested in American managed concentrators, orders which stay "anonymous" for some reasons. To participate to that concentrators you have to stick to the concentrator's rules which are not necessarily in line with fair market behavior: if certain banks can't profit from your orders, your orders are not routed to the market at all. Which is also evident in analysis tools which still slice price in "time-frames" while in real world I can have a 6 months political turmoil creating same price movement of a 15 seconds market panic. Or I can have 2 oil tankers waiting to dock at a fair price and the price being spiking for "lack of offer".
Useless say, market matching machines are perfectly aware that those prices aren't sliced in time frames.
Thus marking centered oil CL price de-facto an obsolete "fake and artificious market" price where a Canadian server-farm producing tons of orders out of thin air backed by soon expiring paper barrels or even by doubtful collateral to be heavier than the official statement of the ministry of oil of an Arab country unilaterally and abruptly deciding to cut production. Often oil is called a "politically cast price" rather than an order match driven price.
Consequences: all dis-informed counterparts using "standard analysis tools and standard oil price" pay the bill of an incredible complex network of inefficiencies you call centralized oil pricing. Inefficiencies which ultimately are the reason why we had to create our own internal oil price feed, which is a proprietary tool, not a freely available indicator , is NOT an opinion... but an algo cluster running in a dedicated server farm and thus we use that proprietary feed among other tools to increment our proprietary algo effectiveness.
Other examples of oil price manipulations:
Another common misunderstanding: oil price is standard , price is centralized in Chicago and NY... spread is fixed etc.etc.
Oh well, not quite. Same deal on oil has a spread costs of 50$ in NY , 30$ in Rus , 10$ in Abu Dhabi and can be even spread negative elsewhere.
Now real issue is probably what you call a "standard price" depends on both control on spread width and volume stability which are definitely not standard and can shrinks and widen either on geographic position AND deal volume or underlying availability or lack thereof.
Often depends who you are, who you know and how good your results are... then you might have access to a choice price not a standard one. Since micro-pricing is taking over (thanks to contract speed always increasing and you can see real-time offering from allover the world on your screen) centralized market makers are losing efficiency respect to local pools. Oil is a sector where market micro-structure is fracturing its vertical market event which is probably having a major impact in the years to come. This is why we expect oil to be one of ther first commodities to have its price redefined by peer to peer networks rather than leveraged bank derivatives.
Example: if you can post your oil barrel price real-time you can choose not to sell it at a bank imposed price.
Other more hidden price manipulations are a daily practice :
If you trade with a tight spread, you think you are trading real-time... sad story. Reality is instead that wider the spread longer the time it takes to go back break-even once entered market at the (spread width) loss.
(assume 10$ spread = 5 minutes lag).
Oil spread on same oil barrel amount is:
-1$ other troubled countries
-10$ other oil for food countries
-20$ embargoed countries
That means not only price is not centralized anymore but
some countries ask (+) money to trade oil while other countries (-) offer money to trade it.
The simplest conclusion being: with 25$ uncertainty area on a 60$ oil barrel price: maybe would be cheaper for the whole market to admit there is no more centralized price rather than faking selling as commodity at a "standard" price which just represents political decisions and not an underlying good price thus carrying an exploitable inefficiency:
if you insist to assess a wrong value of a good and the counterpart knows, you give the counterpart the chance to speculate (arbitrate) against you with your own derivatives.
the real problem is that is politically too expensive to admit there is no more centralized oil price anymore but, guess what, market knows already.
35) Why all platform show same instruments to trade on :
Because Western banks are filled of euros and dollars only . In Eastern countries instead you can use other base currencies.
Example 1: Any Russian or Chinese broker can buy and sell yuan as base currency but a western trader can't. Again, you guess why.
Example 2: If many indexes have same daily excursions differentiating in many indexes ...is not real differentiation. That is apparent differentiation: it just increases commissions on the other side of the trade.
In above chart you see six months of the daily price range (high-low)/close for 6 indexes we track . We know that they are all equivalent one another but probably you did not. It means that adjusting the trade size allows take the same risk on each instrument . Those indexes are properly weighted at source, this means that apparently you differentiate more if use plenty of them, reality is sadly different. Anyone carries same risk provided size has been adjusted correctly.Assume though that each index should be representing a different market but again, they are all alike and perfectly normalized, thus those indexes are simply artificial and don't represent real market which by nature is a bit more chaotic than a perfectly normal distribution . That is , most indexes are synthesizable changing the parameters in the same function.
anyindex = k(deltaFX+LambdaVIX)
36) Expensive models work better than cheap ones:
False , if "the best trading model" existed market would have already failed.
Models applicability always changes with market conditions.
Example 1: back-testing a model shows what a model would do if in the past price was "fair" and market behaved correctly, which is NOT. Back testing of same model in different market situations and timing gives different results: because back testing is a model too.
Example 2: Just because a positive pattern happened earlier there is no guarantee it will repeat again in same fashion, the opposite is almost certain instead. The most informed counterpart has the right tools to profile and exploit locally forecastable events and turn them unprofitable to the little informed counterpart.
Example 3: seeing we didn't come up with an "instantaneously profitable model" a guy decided to fire us for
"lack of productivity" while real issue was he didn't even get what we were doing there. So, in return, I wrote him a post-it note in which I wrapped in my office keys and resigned. Guy asked what it was, I answered i wrote on post it, reason why they were to fail and the date of their bankruptcy... and their bankruptcy happened.
Lesson learned? Expertise value can not be univocally determined by an unilateral evaluation especially if the competence of the evaluator is doubtful: often you see the use "compliance" as an excuse for incompetence.
On most algo optimization routines, whatever the algo beneath being machine learning/neural etc.
you see that "the target variable is specified by the training data set optimization menu"
but that is backtesting flaw by definition! In real market, You can't decide what to optimize
and much less when . In an ever evolving market moving faster and faster against you,
all variable must be kept efficient : optimizing one means disbalancing the model.
Which is the reason why stock pickers algos, fund selectors algos and "top traders" (humans or not) always lose in real markets.
The typical outcome being : "the algo had a flawless trade history and a great backtesting but sucks in real market".
37) How CTA and Asset Managers lie with the track records, how brokers lie with reports:
Asset managers and brokers etc. are perfectly aware that any person can ask "customized reports to prove their profitability".
Hence showing the potential counterpart only profitable trades or just profitable accounts in support to their alleged capability of producing profits is a common malpractice very difficult to debunk.
Example: assume one guy shows you his best reporting , a very profitable one... but that could not necessarily be the only account he has with that broker or maybe he might have many other accounts with different brokers and just show the best one. This is commonly known as "asymmetric information".
Example: how some ways Brokers lie with their Reports:
what you get in the reassuring word "REPORT" hides a sea where plenty of stuff can be hidden in plain sight, masked and distorted .
Practically the more powerful a client is the less true his statement is likely to be: who owns the database own all its universe.
First and foremost there are people who are trained to alter data-sets so that they are customer friendly although,while not necessarily resembling facts, to even be law abiding and is more than 50 years that these tricks are run and the best exploiters have always been been big banks thus research has gone deeply in the direction to engineer them rather than ban them. Facts are then mixed in ways so complex, artificious and artificially cast in complex systems connections that are simply hard to explain , much less debunk or investigate. Some known examples follow.
A broker has the unilateral power to access and manipulate many databases thing which by itself seems harmless creates almost impenetrable barrage of data to any customer or audit much less watchdogs, either they already know where to look or waste incredible amount of money solving nothing as all goes hidden in a few bytes in tons ot gigabytes .Shredding big database in tiny actionable intelligence has huge costs and , that's even mode absurd, doesn't necessarily cositutes a proof. Thus, at his unilateral will, the broker can: derail customer awareness, hinder watchdog control capability, slow customers trade closures, cross customer positions, take private agreement with customers concerning close-to-close netting and hidden leverages or arrange artificially "disputed deals" etc.
Assume a customer does high profit but want to mask his account as total loss. Since is the broker who is the only one allowed to distribute the track records might engineer close-to-close in such a way that a profitable account is apparently losing or vice-versa. But to discover that you would need access to all customers accounts to prove which deal has been sipped from which account, prove which account is connected to the customer, in which way it sips the profit from the main one and all that can just happen with database access not order by order (which would be an absurd effort by itself) but order status by order status for each single order (a single order can switch status from: pending to partial fill, executed, suspended, partially closed at match... and these statuses are variable too.
One might think than when an order is closed... is closed. While after 50 years of "position engineering" it is not.
Even on closure orders were closed or just coupled , netted internally, cleared externally, or just paired etc. etc. there are a dozen of order statuses not just "position closed" and some "suspended order" statuses are very profitable by definition by one one part or another. For simplicity just assume somewhere while the order changes its status someone can sip a tiny fraction of its value now scale it world wide... Trillions? Exactly.
Although you don't read that on the news media often. Do you ?
Another way is external rerouting: temporary rerouting (parking) orders outside the customer account and make them get back later when netting is favorably or negatively adjusted (Cached STP). Just using different network speeds(lags) for different customer can provide that resut...
or ""dump the parked wagon on the worst possible price and smash customer account".
Another way to show a thing and hide another are exploitable flaws in margin agreements: which might show a thing on report while the margin agreement was changed or otherwise engineered and not reported in meantime. A similar way is to formally declare a leverage while a hidden agreement uses another one or spilts fraction of leveraged exposure on many sub-account (hidden deleveraged mirroring) .
Also noteworthy, most brokers have a 5% of deals in the "under dispute" status but those disputes can be true or artificial as a customer might like getting his account frozen for many reasons... frozen until allocated after dispute is settled with customer, who might be in ties with broker to keep that deal "under dispute". Settling a dispute usually takes months while that position in meantime could be allocated to other accounts and provide shadow liquidity which any wise broker knows is one of the most reliable sources of revenues.
Another way is simply switching the account number from one customer to another, (read only copycat account) while one is aware of that the other customer is not . So the losses are real and checked, just belonging... to someone else. As the real loser account would not be informed thus unaware of the fact that his account is used to mask another account profits: the masked account could show losses which don't belong to him. Also that is almost undedectable. Thus the profit maker might use the loss deal on an account to show his losses (cross close). Again, you you would need a double cross check reports and expose that to prove those deals belong to another customer and again register is broker hands.
asset management model is de-facto a hidden mortgage scheme but is so nasty reality no one even speaks about it. Assume you're good and profit 2% a month keeping half profits. After 5 years the part money you left to your partner costed you 400% of your profits. Would you really give your money to be managed by someone who is gearing up 400% debt to cover his profits? No way!... Exactly.
Upon doubt always call the certifier (broker) who often simply denies he can certify the trader...
A wannabe trader to us: "I can prove to have a good track record with profitable and robust back-testing..."
Our guy: "yes yes ... you can -con- ahem but how many accounts you have with your -accomplice- ahem broker ?
Our guy immediately called his broker on phone:
His Broker returned : "I can't disclose our customers report...unless is our customer asking his report"
Our guy: ok but can you swear ME that is the only account he has with you... and that you are the only broker your customer works with ?
Broker: <<I am not swearing you anything, sir>>
Our guy to the wannabe trader: then your customer track record is worth zero to us as you might have (and probably he has!)
tons of accounts with dozens of brokers and you're showing us just the best one.
In short, you can't rely on counterpart evaluations unless you have absolute control of the information reaching your desk.
First sight detectable, ( yes visible! ) structural flaws that tell you , your Asset Manager is probably lying to you and why:
|If what you see from your Asset manager is...
||he is probably...
|| What you should see instead:
a typical 1990's noisy trading room full of people computers and phones and guys seemingly greedy typing on keyboards in a rush
|A small room with couple of tech savvy guys
managing a remotely collocated server
|Worldwide Marketing "pull" campaigns on many media||running a
|No commercials at all. Very low profile to defend proprietary tools from competitors.|
|Nice screens full of charts and indexes||running a Bucket Shop scheme.||No charts and no indexes. Market is so fast can't track it that way.|
|First person you meet is a smiling marketing guy||running a Ponzi Scheme.||First person you meet is an intellectual property lawyer.|
|Big Events where promoters and the like are
invited to visit the company have lunch,
speak with managers
|running a Ponzi Scheme.||No parents day, even temporary visitors must wear badge and are allowed just visit a few, not core, areas.|
|Very wide asset class and incomprehensible
structured finance.Instruments are chiseled one another.
|running a Soft Dollar Scheme.||A very narrow, dynamic basket of liquid assets
clearly understandable by the average joe
|Reports are compiled using many "Greeks" and
|running a Soft Dollar Scheme.
||He should just care for alpha and risk exposure.|
|People oriented to client relations||running a Ponzi Scheme.||Finance and tech geeks fully committed to market only.|
|If you say don't understand their report,
they blame you and tell you to ask an expert
|running a Bucket Shop Scheme.
||If you don't understand their report, they blame is their fault.|
|Clean personal background (no bankruptcies, no complaints etc.)||a newbye with no market experience.||Complex personal history, some failures and many job switches and litigation record.|
|Asking info about him ... banks truly love him
||running a Soft Dollar Scheme.||He is not known neither popular in banks.|
38) How Track Records Certifications are faked.
Originally it was an audit company who certified your deals . After Lehman , Andersen and M.F. scandals surfaced certification has just become a pyramidal scheme with a bribe on top. Since it has become very common practice to rely on small companies in a former soviet block country, where a simple accountant can also have the Notarized Power of Attorney. Since those countries are now part of the European Union then to "certify" a track record all it takes is an accountant and some stamps. Also is again you telling your broker which deals to send there and be "certified" which ones not and cheer on top: who certifies the certifier ? Its country regulations. Certified track records are worth zero
Do a quick comparison yourself between a guy with a "certified" track record vs a guy with a twitter real-time published track record. Which one you would trust more ? To us publishing deals real-time on twitter is more robust than showing some "certified" ones which are picked up one by one the week after and also manipulated from a third party whose clear intent is to prove profitability.
Why certain funds still use certified track records to evaluate people ?
Probably because their customers are not ready to ask a top down audit to the whole management chain. Telling customers that control process is flawed from top is not a nice commercial for the whole commission chain: the bank, the broker, the fund and the manager and ultimately the customers.
Go exploit that flaw. The real point about certification is self evaluation. Faking a good result implies more risks than rewards. Try join a demo (fake money) live trading competition if you can stay on top 10% of the participants you are self-certified and most important you you're playing fair with yourself. If you're fair others will follow you is just matter of time not of "certification stamps".
39) Why remote hosted trading systems , grids , mirror trading can not profit:
The grid owner knows which systems profit, which systems just create volume, which systems lose and even when are about to do it. While you don't have a clue of any of those data.
There is asymmetric information between the trading system grid owner and the retail user who "hires" the system. Way too often the grid owner has a preferred broker also . This fact alone speaks why one should not use such public system grids. Grid owner also knows which clusters of systems are profitable in a given market moment, which ones are temporarily losing instead. Such information is restricted and often sold to his broker or to professional customers but not to the retail system users .
Example 1: why Forex hosting is a scam:
If you run your strategy on a third party hosting service, they know your positions and your margin, they can counter-profile your strategy , resell it , redistribute it and most important they can widen the spread against your specific account if they know your strategy is about to hit. They can widen spread on market closing times or delay your order close a few, very expensive seconds, or most often delay your executions on important news releases addressing the issue as "force majeure" , then they can delay your order routing and... you also pay for that "service" .
Example 2: mirror trading is not what it seems:
There is no such thing as a public profitable strategy. Reason is so simple but noone seems see it: if a grid published a strategy who profited say 1% a day, in 2 months they would send the ECB bankrupt.
Notably enough if your strategy is profitable there are investment banks piling up to run it putting solid gold on it so why publicly disclose it on a grid ? Mirror trading is nonsense.
Example 3: dedicated virtual server running your strategy has a few unsung flaws :
did you know that if the dedicated virtual machine crashes is your fault.. by contract ?
Mind that either broker and their customer are third party to the server farm hosting the system.
Also if you never tried to recover data hosted on a remote virtualized crashed partition you have no idea how complex and slow process it is. They say you can backup from previous machine state, well, provided crash is not hardware otherwise be ready to pick your number and wait for weeks .
Example 4: you can be : cloud / colocated / hyper-fast / run your strategy totally in RAM
whatever... and still have 2 pips spread that is 4 minutes lag from real-time price. Say:
if you maybe save 4 milliseconds but you are lagged 4 minutes and no one explains this clearly
unless you enter in arbitrary pricing deep logic which includes, STP external routing, level by level cleaning,
A-Book, B-book, SIP pricing ,consolidated feeds....
to make it short: your stops are being sold to low profile HFT platforms who sip them one by one as a job.
Example 5: the algo bait.
We successfully insured some algos and other intellectual properties against intellectual property thefts, copyright infringements, cloning etc. through a specialized company cooperating with our C.I.O. in supervising and investigating those issues while in the meantime we decided to test the opposite situation:
as a test we have put a deliberately mock-up algo in the cloud and some algo grids deliberately labeled "autosalp" and it has been forcibly stolen in less than 15 days while even being replaced with garbage in another grid.
Proof of concept:
Autoscalp is switching back and forth from Data center in a box solutions / micro data center solutions...etc.
while remote Cloud has been dismissed in an earlier stages.
Quoting our founder:
"once you enter the cloud you know your intellectual property is at risk. Is impossible to insure your algos if you put them in the Cloud or a Grid as since you do it they are more or less "public domain".
In a sector where most profits reside in asymmetric information, ultimately if you enter the cloud or an algo grid you are practically taking the unilateral bet that there is no one is waiting to steal your algos , that the counterpart hosting company is playing fair and the grid manager isn't interested in cloning your algo or otherwise reducing your profitability altering your code and that he is not teaming up with your counterpart (bet it a bank broker or market maker) .
Once an algo is in the cloud there are many eyes on it, none of them yours.
40) Percent Allocation Managed (PAMM) accounts : deception based lie.
Often sold by firms in close ties with a brokerage firm, PAMMs are sold as "a way to mitigate risk", instead it's a model that fragments a retail account and its margin even further. We assume that very few retail accounts profit, PAMM is just a way to distribute losses without concentrating them on same instruments and to make smaller and smaller accounts so that margin calls get triggered faster. Aka "diluting losses on a longer period" , thing which delays customer awareness of the real risk he's taking.
41) Mirror trading : lie based on asymmetric information
Mirror trading practically means cloning trading signals but is dressed up as "executing same strategy of the best traders". Mirror trading suffers a known asymmetric information problem depending on customer localization. If counterpart expects your click in a given moment can derail prices for few seconds when your signal appears (he knows it first because signal appears on HIS server before than your remote client). Is that's legal ? Depends from country to country, but way too often you sign a broker agreement which specifically allows price skewing practice, which is a kind of arbitration.
Example: 200 users want to execute a deal in same moment, server knows and spread widens when signal appears, all user ask same price in same moment, get slipped and executed at the worst bid-ask price among market makers giving a net profit to he mirror trading server . After that, prices resume normal flow.
42) Pool account misuse:
Some broker companies made even loans out of segregated customer pool accounts . Collectively using unaware customers segregated accounts as a bulk to guarantee their personal speculations. This makes segregated account totally useless. As if broker goes bankrupt or his speculation goes bad, in either cases the unaware clients forming the pool go bankrupt for reasons not even remotely correlated to their account performance . Often an account is segregated for a nominal value, which can be far smaller than real account size.
43) Close all positions before Friday market close:
Just another unilateral lie passed as a convention . Most banks and brokers warmly suggest to close all positions within Friday close.
Apparent reason: possible liquidity spike on Monday morning or price.
Real reason ? Next week they should pay to open them on real market.
Example 1: there are pools that allow trading on weekends. Probably yours ... doesn't.
Example 2: to barely track 1 single index at Friday close, at least 8 fast computers are required . Knowing that , think about this: a medium bank offering 500 instruments with a 20000 customers should have 20 server farm computational power (20 department stores filled with computers) and that just to make sure their positions are handled properly at closing hour or during news releases. Which is not what happens. That would be expensive, for the bank.
What happens instead is that most positions are merged in bulk positions that are wildly chopped on market closures by all other players , day after day.
44) Recipes through which companies boil and lure traders :
Creating strategies, signals and models makes you very active in spotting asset managers and brokers hooks, gives also some hints on how and where to check if a broker or asset manager is about to go bust and his level of moral hazard. First and foremost: always ignore linkedin and facebook gossips as all those gossips are most likely "piloted". They are called pseudo network for a reason: the most informed part decided which information to disclose to whom.
So in our view, the most common ways a trader is "cooked" are :
- Recipe 1: the maverick: pump and dump.
- Recipe 2: the scapegoat: blame the trader.
- Recipe 3: the bouillabaisse .
- Recipe 4: the aquarium: baits, whales, sharks, squids, piranhas, shrimps.
- Recipe 5: the "pssss.... hint!" aka the hype shot.
- Recipe 6: Paint a high-tech name on the box and sell the package with a brick inside it
- Recipe 7: The pseudo due-diligence (all arrogance, no substance).
- Recipe 8: how to spot fake multi billion dollar business guys.
Recipe 1: the Maverick: pump and dump .
ingredients: a rampant ego-driven trader with a few customers, a big multinational , a capital city.
a) A corporation gets you a 2 months full-paid tour of the city making you feel a V.I.P.
b) corporate managers make sure you spend all salary for office equipment, rent, hire fast cars etc.
c) they make sure you feel busy having fun in the meantime the corporation manages your customers
d) company abruptly fires you and send you home even without the customers you previously had
e) company then even blackmails you to shut up or signaling you to authorities for malpractice
Fortunately , most companies doing these tricks end up in this black list .
Recipe 2: the scape goat: "blame the trader version",
which is in fashion nowadays in big corporations.
a) the lie: corporation says very unpleasantly that your strategies just don't fit and you're not adequate
b) the bait: then they change their mind but you will have to work for a very small pay
c) the prelude: they shift tone from hard denial to sweet approval but you must accept the job quickly
d) the deluge: many CVs of their employees flock in your email: they know company is doomed so they look at you for a new job. At that point become aware that company is about to implode.
e) the implosion: their customer care gets slower and slower while customers queues get longer and longer. Disturbing silences in their reports ahead of the unavoidable boom.
Useless to say, at this point is too late to abandon ship.
Recipe 3: the bouillabaisse :
a) the hook: first the trader ,usually a boiled promoter almost to implode, signalled to his sharks from the broker who has access to his account balances by definition
a) the bait: an end client, dubbed trader to avoid him understanding the trick, is emphatically and triumphantly invited to join a surely effective team of seasoned traders for his own broker
b) the trap: the trader has been pinpointed because he is about to go bankrupt
(his broker has specific tools to make sure either it happens and... WHEN )
c) the trigger: when the trader is about to go bankrupt , the team surrounding him reveals their real nature , which is clearly working as I.B. for his broker who will defend the clients interest,
that is the "trader" finds among a pool of sharks dividing his client portfolio.
d) the kick: when the trader is in the most vulnerable position, his broker takes out credit lines and the team forces him to leave the facility. Consequence is the traumatic learning he was simply sharing his office with his enemy. Of course his remaining customers pool is "inherited" by the broker.
Recipe 4: the aquarium : whales,sharks, baits, piranhas, squids and the fee pools.
Is a very rare event that a trader profits but that can create huge costs to his counterparts. So various techniques are used to eat the whale alive. Many tools are available to an experienced counterparts so that he has just to choose which one best suits the profile and the moment:
a1) sharks: many traders start loitering around the whale offspring making him an hostile playground until his implosion for either starvation (too many losses) or suffocation (no new customers) in both cases the sharks will exploit the unavoidable deluge of remaining angry customers .
a2) sharks(alternative): a wise counterpart knows that managed capital has a structural limit either top and bottom. Soon making the mark gear up too much capital, debt and leverage to an unmanageable level will make him implode on his own weight,which is quite easy owning the right tools. At that point any sharks around will get his remaining customers for free .
counter-profile him and take him out of market one stop at a time, like many small piranhas bites.
Any broker-side platform has plenty of tools to hide this specific (very common) malpractice.
c) the bait:
another unsung technique is called "holiday baiting" that is to offer him a holiday in specific period of the years, for example, in july . So that when he goes back to his desk in august , will use some renewed enthusiasm and will trade heavily but being the market orders list very thin,
his stops will be easily taken out from his greedy counterpart. All at once, if counterpart is a heavy market maker; one by one if the counterpart is smaller and on retail side.
d) the squid: a tentacular ego-gifted pseudo-super promoter that will downplay all you have till you sign at his terms. Often too self-centered to really know what he is trying to sell you as his main task is create confidence and confusion, he is often unaware of the real core variables in the company he works for but smart enough not to make questions. The best way to counter this guy is to reply with a counter-sale technique. Something like: "You are making me up man, so I'll contact your company in one year, If you still work with them I'll rise my price 20%. If they fired you I'll propose at today's price instead". This way you sell him while he is trying to buy you at discount price creating him a very difficult solution as he must not justify the missed sale but ...that he's been sold too .
e) Fee pools: aka the mis-configured funds. There are plenty of ready on shelf funds that are simply a way to self-destruct capital through shredding it into very tiny bits these funds are called "fee pools" , most often these hardly detectable bad configurations are supported by promoters, brokers or whatsoever ring in the commission chain as they are paid on the volume not on the fund performances . Most of times those funds have been tailored not to make the capital owner aware of such scheme until is too late. A wise mean of judgement is : if it's on the shelf and ready to be used and warmly suggested by a promoter... most of times that's not what you need. These fund designs are quite often exposed when is too late, also the counterpart designing the fund has quite often a seasoned experience in creating invisible money traps.
Another way to discriminate is that if you see that the fee structure is way too balanced toward sales ,
it's likely that's a "fee pool". This is why a wise practice is impose your own fund structure rather than using an existing one. Just because some Fund rules existed before you doesn't mean that is in your interest to keep them. Another very subtle way to keep the "fee pool running" is asking same advisors and consultant to modify it to (apparently) suit your needs, often turns out that you can't obviously rely on counterpart consultant to do your interest but this kind of conflict of interest is often to subtle to be detected.
Also mind that for them is everyday practice: to open, run, blow and care for the bankruptcy too.
Fee pools are a product of a self referential industry: where the wisest option is to stay out from.
Recipe 5: the "pssss.... hint!" aka the hype shot. A classic boiler room scheme, just one to one version.
Typically a one to one marketing driven hook publicly self-advertised as a new liquidity injection in a clearly boiled fund which is privately disguised and promoted as "not completely doomed" in order to leverage trust with a semi-confidential information from a pseudo-reliable source that a salvage or even a full recover operation is possible, in order first to find then to motivate and ultimately to hire high profile managers for a leverage buyout, such move creates enthusiasm in the newly hired management and while the fresh guys try to tap the flood with very elegant, creative, optimistic, brand new ideas... other more informed guys leave in silence and bring away with them any possible source of liquidity before the atrocious reality of such impossible mission surfaces from the sadly empty pond.
Recipe 6: Paint a high-tech name on the box and sell the package with... a brick inside it.
After 10 years stock market rising would be time to fragment and dilute shares to avoid getting caught in correction, meanwhile some smart folks are instead asserting they will manage the sale of huge orders sizes practically building huge bulk packages of boiled stocks in a nice looking brand new box and sell them to even bigger guys.
Since such practice is clearly forbidden by market watchdogs (aka pump and dump) so someone went up with the brilliant idea that if you call it dark pool ... it does the trick and now plenty of those pseudo-dark pools are surfacing but the big box they sell is nothing different than the trick of a nice box with a brick inside it .
that is likely belonging to a forex chain instead. After a short generic intro about himself his great achievements as a Venture Capitalist starts his flood of questions , broadly professional at beginning completely exposed at the end.
How many people are there ?
>normally 5-6 but can grow to 300 depends who is connected and what's the reason :
a sudden market spike or think-tank session or developing ideas etc.etc.
>How many algos you own ?
Too many to count and mutating too. With a properly locked intellectual property on the algo generators. Estimated worth between 12.5 and 19.5 millions in 3 separate evaluations... but you're a V.C. should know that.
how many accounts ?
...how many customers?
...what is the pool size ?
-...At that point is fair to return a laugh and hang up.
What a strange "due diligence". Guy was didn't seem interested neither in our know-how nor expertise or research objectives, staff profiles, co-brands, co-partners, technology , revenues, network, our history...nothing .
This guy used a "Venture capital" name for another purpose (red herring scam technique) .
He was practically asking us if we do retail brokerage but it took him 5 minutes.. No we don't.
5 seconds including hang-up.
From then on on we started answering cold calls with this "disclaimer", hope you find it useful too.
- We create algos.
- We are in R&D
- We don't buy leverage.
- We don't manage third party accounts.
- We run No Deal Desk, and no I.B.
- We have a very strong legal department
...and that is where usually the Forex cold callers hang up. 20 seconds, not 5 minutes.
When in doubt ask experts and always start profiling the counterpart: what kind of fish does this seller resemble ...is a trick that helps. Keep in mind that a wise counterpart often mixes those tools making them hardly detectable, so don't expect to unveil a "fee pool" scheme just reading the Fund statute. Also reality says that highest paid brokers have the best lawyers thus evaluating sector transparency on the conviction rate would be a mistake: the best defense strategy of white collars is using more white collars.
Recipe 8: how to spot fake multi billion dollar business guys.
1) profile the status vectors: [ location, transport, watch, restaurant, companion, clubs] If just one of them stinks, he is definitely a scam.
Status is like a bubble , if you see a hole, it means the status you are shown is fake.
Anyone with a real VIP status knows that any flaws lowers his class one notch and the risks associated with that.
So, to get if a guy is seriously engaged in billion dollars transaction profile his "status vectors" without being noticed.
lives in a fancy house in suburbs of a non core city, has hired a sport car, wears an appariscent watch, goes to doubtful quality restaurants
has a rag doll beauty as companion and hangs out in night clubs. His status speaks "fake". He probably lies even talking millions not billions.
Instead what about a guy with ordinary dress and boring life but... with a private jet? Different class level? Exactly.
2) check the status he offers you: as status calls status . You either get an all inclusive long weekend in 5 star hotel with charter plane ,
or a short weekend in 4 star hotel excluded drinks, with touristic class ticket. Different stories.
before a quote even hits the market. A newbie always expects both sides take a risk when dealing: wrong.
Sellside always manipulates the price at source before the opportunity hits the market. [Caveat emptor].
Banks and exchanges seclude the profitable assets for them and just share the ones bearing biased risks
and then let the traders take unilateral bets on them .
If you know how deeply prices are manifactured at source you know what not to touch...
If you don't... anything goes but you think you are deciding wisely, instead you are running a second best
on a biased price whose added value has already been extracted at source and transparency? Long gone.
-Why We don't do stocks and their derivatives:
In1986 you could look at EPS and the like but after dotcom crash,
after 2008 leverage implosion and after 2011 flashcrash
the outcome is that price virtualization took the lead.
Which means arbitrarily casting artificious prices to protect
the company value has become standard practice
Government injections to protect main indexes is an even more known issue
"If an index shows a trend on a logaritmic chart is 100% a ponzi" (think S&P500 after 2008).
After 2011 they tried blaming HFT rather than hyper leveraging but then they quitted...
If you notice ULLHFT is just done on the buyside of the stocks
[be it a sell or a buy order] but not on the sell side which is manipulated
at source before it reaches the order book: by either the market, the nomad, the banks
or even the company itself through "comfortable intermediaries"...
If you see the real stock pricing passages to get from buyer to seller
today it looks like a blueprint of a 1972 TV: is complex and definitely not transparent
chain of micro lags ... Not qute the "fair market" you hoped for.
-Why We don't do Funds:
Because is a Monopoly. Assume you have a pie of 4000 funds to choose from
but you know that most fund are run as a monopoly... yes really,
owned by a single (well known swiss) bank who indirectly controls the whole
Fund market, its pricing and "apparent delocalization", its trends and subtrends... anything.
Is rational to expect that also the whole sellside pricing structure to be done in-house.
So, if you know that you also know that their data engine can easily
manipulate the sellside prices at source:
Hence even developing the best A.I. augmented "fund screening software on market"
is an exercise in futility because as soon as you start "picking" they start "gapping".
If you start hammering they start lagging ... theirs the engine, theirs the game.
Yes, it's a game. Who control the sellside price engine controls the game.
Every dealer knows the trick, Muppets have no idea instead.
-Why our algos just cover eurusd oil and gold:
in short: because We are fully aware of sellside price manipulation-
Notwithstanding the forced brainwashing done by banks on "pseudo-transparency"
and "compliance" in "standard asset pricing model" (which was invented in physics to justify a random
gas distribution in a room and badly adapted to market and NEVER modified since).
If banks write market rules and You believe them , they win. Those arent market principles,
just comfortable conventions to cast fractional reserve driven, leveraged, arbitrary pricing.
Done to justify and uphold their compound interest, fractional reserve and over-leveraging which allow
unilateral price manipulation of prices at source and, yes, you call that "fair equitable market".
We don't. So, We just use instruments whose sellside manipulation of price is the hardest on planet.
Eurusd, oil and gold: to derail them for a just while takes many governments treasuries committed to it.
Also, multiple governments seldom agree on a pricing structure for a long period of time and that
helps prevent sellside manipulation even further.
Fun fact: in many exchanges gold oil and eurusd ... still have a pit, not and automated settlement of prices,
and notably the persons running the pit aren't employees of the exchanges, are government employees...
sounds nasty enough?
If not... just think about what allowing a single bank or market own the sellside pricing of a whole sector means:
they can litterally cook "pseudo prices" internally and noone can question that as they are making the rules.
Exactly like creating...
CDOs (they know the Fund expires)
CDS (they know what is inside the Funds have no value)
and not only they can set (biased) leveraged side bets on them both but can also create from nowhere...
Syntethic CDOs: fund of funds ... and then sell "shares" of them both to the muppets at full price...
See the movie the big short more or less is all about about of synthethic pricing : make believe.
45) Asymmetric info and restricted usage of price filtering:
This should summarize many points :
Most price engines can allow or deny price and spike filtering , this is done arbitrarily depending on customers liquidity , available collateral and of course, his importance.
Similar phenomena are "Micro Netting" aka "beat the closing" ,
naming convention depends on the market you are referring to either FX indexes etc.
We are talking about the infamous market sync , the moment where each platforms ping others to be sure they are all showing same price and it happens: every half hour on exotics and slow indexes every five minutes on fx majors and fast indexes and in certain cases even every minute , on the minute on the most heavily traded instruments and while major news are released.
Example 1 :
broker-side price engines don't weight bid and ask waiting orders to be executed ad price components because, theoretically such orders might be removed , so instead of the order book, the median price of the last minutes of trading is used as price source. Reality is that order book would give out information to customers and market makers prefer not to. So , what you read on price when the sync price occurs is the median deal price for each five minutes, which as also based on the wrong assumption that bid and ask are constantly separated by a fixed amount and that each order off that bracket, whatever its size, is fair to be ignored.
See the tight spread issue we spoke earlier. So, median price sync hides not one but three lies. Adding time rather than book values, forcing price to look steady and faking bid and ask to be in equilibrium , practically adding time mixes things like in the OHLC candles paradox named earlier.
most of times the synchronized price you read on platform is built this tricky way:
counting deals by passing of time rather than their size. This means that the price you see on platform is maybe formed by the last 50 very small deals properly delayed and diluted in time rather than the huge order standing far off that quote . Some small orders diluted in time and ignoring a huge order among them (definition of median) is obviously totally wrong from a market structure point of view but such fake price allows the market maker to spike out customers stops at minimum cost as the market maker knows the price is showing is slipped or skewed and such slippage will just end when others will force its price to sync. So it allows the market maker to hunt stops until the sync moment occurs. Filtering market-book information and reverting it to a time-based information is a unilateral choice, unbiased and profitable... just broker-side, like the OHLC candles named earlier.
broker snipes your stops if he sees a counterpart at your stop level, not if your screen shows you that price. This alone explains why market sync and synthetic prices are such a tricky thing.
46) Mind that even the software evaluating your trading system might fail .
Evaluators often fail to properly represent a system profitability and confuse the trader who then becomes over-confident in his new wonderful freshly developed system. System evaluating software packages are models themselves , hence they often fail too.
Example 1: Some system evaluations are particularly intriguing. it shows 33 % monthly profit on paper trading but this system performs just 3 % monthly on real market.
Too many differences between demo and real system evaluation to be explained verbally, more or less:
quality of execution, quality of order opening/closing, direct and indirect slippage and down-time.
To translate a "demo system" in the real world we prefer apply this translation for each deal done :
realprofit=demoprofit/4 -2*spreadcost -volumefees -tax if system survives deserves further evaluation.
Example 2: way too often is the broker who is giving you their trading system back-testing package because is broker is interested boosting over-confidence in your system and make you trade more and faster, not because your system is good but because they profit on any deal you do either you get a loss or a profit.
47) Often brokers even try sell things they don't know anything about .
This is why is hard to spot where is the catch because quite often the catcher himself has no idea where it hides.
Example 1: "the retailer" : summary of a real phone call about an attempted sale to us.
Seller: - we can offer you ...low latency... ultra fast platform ... proprietary liquidity..
Our Guy: wait wait.. How much you pay me for the additional risk of executing my stops FASTER ?
Seller: - No no, Sir , you are to pay us for getting you our service...I assure you is the best ...
Our Guy: ..aha. You execute my stops faster and I should pay You ? go figure !
I keep my money, thanks. Go scam someone else! Bye .
Example 2: "the server colocator": summary of another phone call about an attempted sale to us.
Seller: -- our colocation plan includes server colocated in broker-side server farm
ultra- fast connection to virtual machines locally connected to the broker..
Our Guy: you can't be serious saying you sell ultra low latency hyper fast colocation , that is milliseconds lag... to people that you know got 1 pip spread--- as it means 1 minute delay, c' m on...
Seller: -- I didn't get it ... we offering you a dedicated environment , micro wave carrier with a dedicated channel and satellite relay... a hyper-fast transmission and environment...
Our Guy: Wrong! Listen up . Assume you got 1 pip spread you have not five or fifty millisecond not even a few seconds away from market price , you are lagged at lease one MINUTE !
Seller: -how do you know that ? I didn't know myself about that...
Our Guy: Of course, you are not supposed to know...but you sell it anyway, right ?
Example 3: a corporate seller trying to sell his "Trading specific business intelligence package"
Seller: --...our Business Intelligence package is a specific tool that through a wise analysis model gets you ,
the final user, strategic valuable information about your competitors and their...
Our guy: Wait...wait I already spotted a nonsense every two words...
Seller:-- please feel welcome to expose your view...
Our guy: OK. First problem:
If a global analysis model existed big data would be shredded and sold in profitable small "end user friendly" chunks not in analysis packages.
Another problem is that such info would be sold at high price from known, well informed parts with a pseudo-social front end to other maybe less appealing companies with huge funding beneath them. If you think that harvesting public or semi-public data and building a model around them is giving you decision ready data-set you live in the 1980s. Any decent company with a decent CIO knows that is wiser to disseminate discordant infos rather than coherent ones as that simple move transforms data miners in garbage collectors, as far as I am concerned your Business Intelligence falls exactly in that category. Then my conclusion is that your Business intelligence package is a pseudo big-data harvester casting conclusions from garbage collected data and also runs an analysis on that, which is self referential to say the least>>.
Seller: (upset) --sir, are you...ahem a competitor of ours...how do you know all that stuff?
In those cases it took us 30 seconds. to spot the trickster and 15 seconds to get you aware of such trick. Real point is: will the seller ever be aware he's selling a scam or he'll retire after 40 years in sales with his golden watch without having ever known what his real job was ? Because that's what some companies are after : making sellers totally alienated and unaware of the real product they are selling so that they don't even know their true responsibilities. That guy was selling a risk and asking us to pay it.
48) Pseudo ... trading instructor, trading coach , trading guru , trading mentor trading rooms:
They should stop giving others advice until they lost at least 5 millions , of their money.
No matter if they got a Phd or whatever preposterous piece of paper they bought.
Unfortunately Mr Market doesn't get scared by such papers.
Most of times trading coaches are employed by your counterpart (broker) so watch anything they say. Is also quite common for them to use coaching sessions to sell you something. Strange enough: they are always employees, not financial entrepreneurs, even just for that they have no idea how to make money, of course, except selling such "coaching" to you. They are paid to hook customers and make them spend more possibly on their firms but they are NEVER paid because they make their customers profit. Why ? Here lies the paradox, which is called adverse selection:
if one were a good speculator he would never accept a fixed salary to explain others his own skills!
Example 1: assume you are a big customer and after joining such course you understand that the instructor is really valid in making you profit , then you would pay solid gold to get him into your firm . Hence, if one is a trading instructor and still sells courses means he's totally incompetent.
Example 2: Reality check against con artists:
Since banks seldom disclose statements of their customers, track records are easily faked and penal records get cleaned every five years, there are a few practical ways to spot con artists but we know that the average life of an asset management company (4 years) or a fund average life. We just split it in two the expected counterpart company life expectancy for each of the below statements we discover true.
- his personal background is consistent of previous failures
- starts postponing appointments and is full of due date advices
- accepts any task but closes none and just delays others
- starts changing his phone numbers like a girl changes her clothes
- starts appearing on tv and newspapers for reasons more consisting of self-advertising rather than true interviews
- uses red herring marketing techniques sponsoring fast cars and taking pictures of him surrounded by hot girls
- uses mesmerization as leadership technique (to be surrounded by yes men rather than true partners)
Typical usage of the above method:
If just one of the above points is true : expecting bankruptcy in 2 years.
If all six are true: expecting bankruptcy in a few months.
If none of the above is true... but something doesn't match: better check again.
Also a bit of common sense also helps in reality check: is normal that ex partners will hate you. The opposite is not normal instead.
So if his ex business partners ...still love him... well something just doesn't fit the frame and the information you are receiving is likely to be biased. Broadly speaking, most of these guys are so self-centered in mastering communication skills that simply fail to cope with reality checks.
How not to organize a trading room:
Assume you visit a trading room and you are astonished by their efficiency:
- broker-client relation is reliable
- information chain is perfect
- corporate commitment is stellar
- compliance is peculiar
- service is flawless
- tasks are sharply defined
- risk management is outstanding
Still you feel those guys have no idea which business they are in . So , when you disclose your view on them
to their C.E.O. explaining plain and simple that he's likely to go bankrupt soon, he doesn't seem to care.
Then the real issue surfaces: traders who lose money are rotated, while the heads controlling them are not.
This creates an adverse selection: one who learns too much is fired as much as one who loses too much.
Diverting responsibilities toward subordinates inadvertently transforms them , a profit source,
into an useless bad management damping mechanism, which ultimately is a true highway to hell .
A healthy rotation is to change 1 team leader every 10 traders hitting the road but you'll never see this applied in banks, asset manager or funds. Apparent reasons are, being afraid to lose customers, lose know-how, lose volume... real hidden reason is they all prefer a disastrous structure implosion rather than a healthy rotation.
After all they are not owner of the structure they work in, they are just managers and they just care for wages and bonuses. If the structure implodes they'll migrate elsewhere ad restart the loop.
Synthesis: if a trading room structure resembles a yes-man-chain, disaster is imminent.
49) Financial Tv Shows and their lies:
...why financial tv shows are just commercials.
- TV programs don't host you for free .
- Most fund managers are forbidden by contract disclose what they do, go figure on tv !
- To be hosted as "guest" you must pay. Higher the sharing , higher the price .
- For the mere fact of being in TV he needs customers and he does not trade.
the cover up:
- when one says "sell" on tv he could be buying instead , so that :
if he is right you join him , if he is wrong he profits.
- If the guest was previously wrong not only he never says so during the tv "interview" but also removes youtube videos about the previous one.
- Neither guest nor anchorman are connected to a polygraph but are both... in sales.
the hidden sponsor:
- If a show hosts you free is because your view coincides with their sponsors' one.
- If you are invited in a tv show you must early submit your view for "supervision".
- The anchorman binds you to switch subject when he (his sponsor) asks you to.
- You are invited again if the sponsor is happy, not because you're a good manager.
Example 1: you'll never hear this on a financial tv show:
"last time you said a thing so absurd i got 2000 people wishing to ask you damages..."
Example 2: you can't sue them for attempted market disruption or disinformation
Example 3: just google for "native content"... that's the commercial they sell as news.
A peculiarity worth mentioning when speaking about "trading books": there is no trading book at all.
Any succesful trading book would be forbidden knowledge and big players know that.
No trading books are really about trading, what they are about is lucky shots of rookie traders
who after (eventually) gaining some dimes, reinvented themselves as librarians.
That's just how information works today :
if something really works and could hurt banks: goes under wraps in case you didn't notice.
Banks don't like when you profit, so they give you bad intel when creating your knowledge basis
so they avoid the risk. They like to keep your ideas in a "safe banking environment" so you just
get efficient.. in buying debts, not profiting.
Think about it, how come there are restricted access libraries ... for economics !?!
Sounds quite strange, isn't it ? Denied patents, sealed patents, or forbidden knowledge and the like
Some special permissions are needed to read certain books, some even dating 500 years or more.
Example: the very concept that views economy as an electric field, is forbidden knowledge.
You didn't know about that. Why? ...You see no books speaking about it ... but it works
just need some fantasy to re names banks "accumulators", companies "engines", liqudiity "electrons" etc.
why they don't say it works...? Because is their job. "If it works hide it before it gets to the public".
You are not in the need to know , how to profit, you are supposed just to gear up useless debt instead
and keep trying so you buy more debt: that's their work. keep your learning curve flat.
By absurd: if you trading book gave you good ideas, banks would lose money if you used that.
If everyone were using that model : the compound interest based model at the base of the
banking system would collapse instantly. :
the innovative idea at the beginning of the XX century was at the core of every bank,
a thing which today we simply call "ponzi scam".
That banking model is today commonly known as "ponzi scam" but they don't write it this way
in economics books, they mix it up, they split it so you don't get it.
isn't healthy for (what you think is) economy you really understand how it works.
Both fractional reserve and compound interest are the scams every bank relies on,
then after 1910 any book about "economics" talks about "banking economics" instead.
Is different. Thinking as Bank-centric is useless on market-side:
you need to deal with the (unfair, biased, very informed) market not with the "compliant" bank.
Hence the "trading books" We suggest are about strategy, NOT banking (pseudo) strategy:
Sun Tzu "The art of war"
Von Clausewitz "On war"
..."Psychological operations in guerrilla warfare tactics".
Anything written by someone who won a hostile in a dynamic environment goes.
Anything written by some bank clerk or librarian faking is a succesful trader does not.
Market is not " a bank compliant , equilibrated exchange of goods for values"
is a fight. An unfar, biased game against the most disinformed counterpart.
Game changes while you play, rules changes , definition changes, values change
so much so that the "fog of war" is a proper way to define a true market environment.
Just start naming that correctly, can find the proper book. What you need
is not about trading, is about fighting an uneven asymmetric war.
Same strategies that work in asymmetric wars work in trading but nobody will ever says that aloud:
Trading the market is fighting an asymmetric war.
Just trash the pseudo-banking strategies every muppet sells,
replace it with real experience of real generals matured on real battlefields.
That is exactly what AI is doing these days to get ahead of humans.
AI has not same reading limitations and information access limitations humans have.
It can opt-out moral constrains and ethics, can access any available info instead.
It can easily discern what is PR from what is real info, what is disinfo what is counterinfo, can you ?
Probably AI will write the best books about trading (is said a smart AI is running S&p500 since 2008)...
but you won't have access to it.
How's that for your "trading knowledge" expectations?
50) See some movies instead [links below].
51) Why "Anyone can trade" is a lie:
Banks and brokers are fully aware that trading is difficult.
Is the hardest game on planet and the most iniquitous game possible.
Example: in major trading corporations every trader must undergo a very intense trading course to forget
all "common knowledge" about trading. Plenty of ethical issues surface as soon as that trader realizes that
in a big corporation he can make money either profiting and losing customers' money.
End of second part of internal interview .
|Boiler Room||The Bank||Margin Call||Rogue Trader||The Humminbird Project|
|link to core scene||Youtube link to movie trailer||link to Movie trailer||link to Movie trailer||Link to movie trailer|
The movie title speaks for itself: how a Boiler Room Con works.
Forget movies like "Wall Street" and see the real thing inside out.
See this one to get how asymmetric information conflicts work.
"blame the trader" game in fashion in any TBTF firm near you.
This movie is about fast monopolists and slow muppets..
About the risks of not reading disclaimers.
About the risks of trusting technical analysis geeks.
About not doing proper due diligence
About good faith and rookies expolitability.
|About what happens if you fail to hide a good idea with proper I.P. tools.|
If these above were almost fiction movies the following ones are history instead:
How they do it? They forbid access to real knowkedge and just provide "banking cooked ideas" to masses.
The very existence of restricted access libraries containing economic books speaks by itself.
[ just see part 5 if on a hurry]
Summary: how they convinced the most important banks that they could beat the market is still a mystery
(nonsense in term for anyone working in it) but on and off some freshly brewed PhDs really think he can do it
and usually it ends up costing trillions to ordinary citizens and the future generations.
Who profited ? Banks who lend the money to cover such catastrophic holes.
in other words: if market crashes produce bank profits indebting others.
Nowadays the situation is different as flash crashes happen so fast that
not even banks can profit from that. You might want to see this clip
The BATS IPO Crash from 16$ to zero in a few milliseconds
and listen to this famous interview where a guy decides to speak clear about the industry:
trading , ethics ... and what is the real deal beneath it.
To get a glimpse on who's running the show nowadays you might also want to read this book.
The following two documentaries explain the ongoing situation with incredible transparency:
except maybe for the little known "special orders" issue but that is not "common knowledge"...
The following is a very interesting piece of investigative journalism: aired on Russian Today
it clearly explains that Watchdogs have absolutely no idea how markets work today
and it seems they are much more worried that this doesn't get on everyday newspapers.
For how strange it might look to a newbie ...you get more about the risks of over-leveraging
from this song that from your broker's footnote.
professional, amateur and addict attitudes to risk taking and the relative (harsh) consequences.
summary of the above documentaries:
Before you try beat the market with your laptop,
give a look at who your counterpart really is. Is not another guy with a laptop , most likely
is something huge like the server-farm at the CME data center server facility in Aurora. Illinois.
which btw has a dedicated full scale electric power plant . It has thousands of square meters of
ultra-fast computers totems (aka big irons) to stay ahead on market and to connect them with other exchanges,
they don't even use cables...
they use a dedicated microwave antenna relay so they get 1-2 milliseconds faster than normal internet wires.
Now you know all this things are working AGAINST YOU. Enough, Right ? No !
You still miss the part that there are a few guys with special orders jumping the order queue.
and that what follows is not the electric scheme of a 1978 telefunken color t.v. but
the most detailed picture ever taken of how market really works today (thanks Nanex for existing!).
Quite noticeable : HFT are not lagged at all while others...
<<well they have no idea of what they do so ... who cares!>> being the best quote to describe the situation.
Since 2007 market is flawed by design to allow banks keep the status quo: not to make market participants profit, ever.
Real reason this thing is surfacing is not the flash-crash not the one trillion dollar cost each flash-crash provides
but that there is quite a situation going on: as banks can't control market anymore. Banks are being surpassed.
They are not efficient enough , neither in trading nor in execution or algo development and even less in risk management.
Example: the price you see now,already produced tons of money to someone else 29'500 milliseconds ago so...
<< c'm on kid , stop dreaming you and your laptop can crack this rigged market... ok ?
Your broker calls you trader in the cold call but after he hanged up he calls you muppet>>
...and that was the most honest comment ever heard about the today pseudo-free market .
52) How Asset Managers and Hedge Funds lie about "their" algos.
First and foremost, nomatter if you are floating on an exchange or not, if you are a true profitable company, you should be already aware that beating benchmark is hard and in the long term it never works.
Secondly: creating algos is incredibly time consuming: if you think you can't code, if you code you can't trade, if you trade you can't think. Anyone saying can manage, trade and create algos... is a sad clown.
In 2018 average hedge fund performance was -6.7% [source: New York Post].
So just to reach breakeven it just means you must be +6.7% above the average which means you
either are in the top 20% of hedge funds or are trying making it up with tricks: leverage, soft dollar, bucket shop or a plain ponzi and with these premises, expectations fall to disaster level and is a statistical certainty such Fund will end in the -70% range within 4 years, another sad fact.
All retail broker Ponzis blow, all Bank ponzis blow, even central bank ponzis will (in a few years) just do bigger blows at the end.
Once the make-believe loop is started there is no way back: once you start making it up
it will ALWAYS get worse. Everyone knows, noone cares. Why ? They all profit.
Sow hy these things still happen you might ask...
Because Mr Market is very creative in recycling same scam dating back to the early 1900s.
When "Ponzi" (aka fractional reserve, compound interest indebting of masses ... etc.)
was a financial "innovation". More or less process is always the same:
disinform the buyside, get them under a monopoly, indebt them with their money,
so will stop making questions. Banks know this trick very deeply and exploit it every day.
In fact most muppets have no idea how hard developing an algo is.
They think the average bank guys knows how to do it, wll, He just knows how to get his wage
out of your pocket, nothing about how markets crush every algo every day.
Curiously even the typical Fund customer or even Fund partners have no idea about that.
If you knew how to develop a succesful algo would you REALLY work for a fixed wage
and enrich someone else's bank ? No way. Not difficult to understand but hard to admit.
Is socially efficient (read politically correct) no muppet is ever told about that.
How fun is that?
To run a profitable algo: you must expect to fail 400 prototypes in the process.
Hint : Von Clausewitz understood war as an information problem:
information redundancy, saturation, bias, overinformadion, disinformation, counterinformation,
lack of information and misrepresentation and fog of war.
Also all of the above are subject to lags. So, in each profitable model all piece of information
must be trimmed in each moment in each strategy and checked for consistency in the market.
In short: if you think your Fund "just developed a few algos but good ones"... be sure is 100% make-believe.
and if someone revealed the attitude to lie to you, why should you believe his algos ?
Sad truch is that the ratio of profitable algo vs losing algos is 400:1 .
Probably some smart pseudo tech guy in the "Fund R&D department"
assured you he can cope he will beat the market, beat the benchmark and whoever in between... and this is where it gets fun spotting the liars:
So, how about focusing on the lie itself to spot the flaw at source?
The typical I.T. lie is easily spotted when you ask the guy if he can "personalize" the algo:
if answer is yes, algo is 100% a fraud.
Once an algo is "live" can't personalize nothing. AI can hardly keep it running ... Sad truth.
Another tool of the trade to check for fraudsters is ask a "small adaptation"
can you edit the algo to respond to this ... performance / index / market...
If answer is yes, algo is a fraud.
No way you can adapt let's say 16gb algo running live you change just one variable:
mayhem can happen.
Another typical psaudo-algo fraud is social investing which is a nice way to rename
social engineering against muppets. Banks know that, brokers know that... muppets don't.
Social investing means putting all eggs in same basket , then whack the basket, brokerside.
and blame... the muppets ( with force majeure clause
Another very human attitude to say "yes We can manage that" because is nice to the ... muppet. If you ask you Fund if they can develop a new algo for the next trimester or even some faster way if they come up with "yes , can do in two weeks" , algo is a fraud (and they know that). These things are simply impossible: flawed ideas and waste of time.
Let me add our Expertise on that :
since 2003 We NEVER succeeded to apply one algo to a different instrument or market.
That is the reason We run 1200 algos: because they cover all possible market statuses happened up to now: including unexpected news release, flashcrashes, exchange faliures, regulatory changes, monopolistic positions... of the 3 instruments We cover [eurusd, oil, gold].
Our view: if an Asset Manager or Fund runs less than 400 algos for each instruHent he covers: chance he bankrupts in 4 years is >90%.
Another was to spot if your Hedge Fund / Asset Manager is making it up is to ask transparency on the algo process: just ask them the very unpolite question ( but is your money so you can ask) to disclose how the algo work: if he discloses that to you, either he is making it up or they were really given "something" from the real algo owner who usually is not the last muppet in town so probably gave them a fake mock up that in case of disclosure any thief would hang himself up with.
Typically, marketing guys are extremely proficient in pretending but at the third technical question they sink in their own lies, takes less than 2 minutes to spot fraudsters if the guy knows what to ask.
Now, play the opposite role instead: think as the algo owner , the real owner.
The algo owner has a very precise objective in mind: make sure no muppets learns nothing to protect his know-how and doing that
baiting, hijacking, spoofing, derailing, deceiving,delayng disinforming, mocking and counterinforming are the norm... anything goes.
He must become a Counterintelligence manual on legs just to keep his intellectual property.
That's how hard it is today. If You weren't told about that... either your Fund is smart or has, unfortunately, no tech to hide at all!
Another way to see if your Fund really owns what he is selling you is if at a certain point of the discussion "Non Disclosure Agreement" or NDA comes up as an answer. Can't go past that.
If you can pass throught the NDA, whatever the trick you use, be it money, promises, anything: if you pass through the NDA screen and he does disclosure, the Fund is a fraud and they own nothing: they would never , never ever disclose you the real process.
In short: if it took you years to develop an algo would you allow any muppets on planet to join it so that ANY decently capitalized counterpart could make you all a herd and skin you all?
Anyone in the algo business knows that: Muppets always pay the bill. Algo owner always sets the rules. Brokers and banks just play middlemen. At core there is Mr Market: let me repeat that, at the center there is the Market, not the customer, not the bank, much less the broker.
Putting customer at center is the kind of lie just a marketeer can come up with. Doesn't work when market eats customers alive.
In conclusion: the algo owner is a firewall separating the Fund from the disaster.
They know, you don't. They don't want you to know that.
A nice example to focus:
think a 4 football fields wide car building firm: how many people you think are working
in the building process? In the eighties it was thousands... today, just one . All robots.
If production line gets halted, self-reroutes. If a robot breaks, other robots fix it.
You know what is the only remainig person in the fabric doing? Security.
Is making sure noone gets closer to the robots and steals corporate secrets (algos).
Competitors might reprogram other robots with same algos, steal the whole production chain.
That is why the security guy is not even allowed to see the robots upclose himself.
Now think again: is the Fund really owning algos they sell you?
A hint: the car dealer has no idea how the robots built it.
He is not in the need to know that no humans at all built the car he is selling you.
A case study: the stolen algo.
A Bank decides to clone an algo as its advisor says "reverse engineering that algo is clearly the cheapest way" exploiting the self-assumption <<we have the right I.T. guy to rebuild the model better than the original >>.
Which is obviously a false statement: if their I.T. guy is so good at trading...
why does he work for a fixed wage... in a Bank??? A sad story is likely to unfold in two possible scenarios:
The bank hires the I.T. guy to reverse engineer the algo (clearly not in good faith). He then clones the algo but is noteworthy that either he fails or succeeds the Bank is never told about it. He will ask the Bank to use the algo in any case . As he would instantly get fired if he admitted the algo he hardly cloned doesn't work properly.
The Bank then tests the cloned algo which typically starts profitable then loses more and more as the time passes because the cloned model, without receiving the original updates, loses the alignment with the optimal market configuration it was designed for. The more the time passes, the higher the losses for the Bank, the more customers complaints start flocking.
Outcome: after a disastrous threshold of losses has been reached the Bank fires the I.T. guy and might even call the original owner of the algo asking for help . That is like stealing something and then even blaming the owner because the stolen good is defective, thus the Bank might receive no answer at all or even get sued for theft.
The I.T. guy successfully clones the algo , which works flawlessly and create profits but when the I.T. guy realizes the real potential that the algo has , he immediately resigns from the Bank and starts selling "his" tool to others including the Bank competitors: other banks . Typically done through an intermediary.
In any case, after a few days that he halted the algo from receiving its original updates, the algo gets misaligned from the market situation and then the algo starts creating more and more losses to the Bank, then the outcome is the same as the case A.
In both scenarios: playing dirty against the algo owner always plays against the Bank itself.
Is not just an intellectual property issue, is also an entrepreneurial mistake:
From an entrepreneurial point of view: you can't take out the entrepreneur from his own business idea in the initial phase, as this means dooming the business to failure since inception, an event which any startup expert would clearly define not just impossible to rescue but probably even impossible to salvage.
If you take out the entrepreneur from his venture you set the basis for a disaster.
From the ROI point of view: playing fair with the algo owner would have been cheaper faster and smoother to the Bank and probably much more profitable. This is why an NDA is an useful tool if properly applied and is working for both sides of the deal as it sets the basis for reciprocal respect, the track to follow and the milestones. Typically the NDA is also an useful cluster contract for future undisclosed agreements too.
From a stake holders point of view: things get even nastier. Since no NDA has been signed is definitely not in the algo owner interest to keep secret the hostile behavior perpetrated by the Bank or, why not, ally with a few brokers competitors to make the bank liquidity get wasted. If you know and algo is flawed you don't need a bank liquidity to whack the bank.
You can capitalize the flaw and short them with their own money (snowballing their margin).
Set aside the responsibility the Bank would publicly face if going to court for cloning the algo! That is exactly the kind of drama Bank stockholders hate as it clearly speaks management failure. Another nasty outcome, managers won't probably admit the flaw since they know
They stole the algo, if the algo owner killed the bank... will they disclose they knew it? No way.
and that's as nasty as it can get if you steal something you have no idea how it works.
From a Clan dynamics point of view:
In some sectors you hava a clan or families controlling the output of a stream flow good (plastics, oil byproducts etc.). The algo creator ultimately was creating profits for the clan, not for the bank. If the bank speculates against the algo owner, endangers the family.
The bank stealing the algo from the owner plays against the clan itself thus the bank might risk seeing all their deposits withdrawn as the attempt to steal the algo is discovered.
Also as soon as the algo owner reports to the clan of having been stolen the algo from the bank, this might shed light on a deeply hidden opportunism: same bank who stole to the clan once, might do it again in the future and could have already played unfair in the past thus that is important as the clan is now aware that such behavior is likely to occur again. Which might turn the Clan not only into choosing to suddenly close all the deposits in that bank but ultimately creating to the bank a major market shock, leading to a massive shorting of its stocks.
All of the above cases are far more expensive for the bank
than respecting the intellectual property of the algo owner.
At autoscalp we create our proprietary tools, our speculative instruments ,
what you have just read above was an essay of our personal experience,
not an advice. We don't offer financial or trading advice of any sort.