UK trader in fraud arrest over US 'flash crash' | Hacker News Reader

UK trader in fraud arrest over US 'flash crash'

bbc.com

56 pointsgoodcanadian73 comments

Ok, maybe I'm horribly misinformed but I thought this was commonplace on wall street (trying to trick other algorithms and the like). Are we really going after this guy because he is just a guy and not some massive corporation? I mean it wouldn't be the first time the single person doing something is made an example of while corporations do it without repercussions but surely I'm missing something here...

Edit: ok this article (posted in new right now) http://www.businessinsider.com/what-is-spoofing-the-market-2... seems to have a lot more information. Relevant bit:

> The tactic [Spoofing] was outlawed in the 2010 Dodd-Frank regulation, but, as with other forms of fraud, it's hard to prove the trader's intent – in this case, the trader's intent to cancel the order. Prosecutors must prove the trader didn't change his or her mind for legitimate reasons after placing the trade.

Edit 2: Ok looks like regulators are just now starting to really go after these people. Who wants to take bets on the chance that ANY major financial orgs get in trouble (and no, fines that are pennies on the dollar for what they are making is not "trouble", I want to see people in jail seeing how we are trying to extradite this UK guy.)

It's the same story like that of Bernie Madoff and also the same morale. You're free to do whatever you want hustling & bustling screwing people over but if you dare to mess up with the wrong people, you're toast and you will be made an example of for everyone to see.

Not sure why you got downvoted with no reasoning, but you're right. I would bet my left testes that major money corps of the Three Sister Cities do the exact same thing but just make it less obvious and do it over longer periods of time. The whole things is rigged anyway, and the SEC is a joke.

Also, there are plenty of middle-men scapegoats to be sacrificed if the big boys get caught too. (see, gold manipulation scandal, libor, etc).

I don't give a shit about getting downvoted and I'll keep speaking my mind. As to why I was downvoted, the mindless HN swarm here can't handle any dissidence and when it encounters any opposing view that contradicts with its worldview, it attacks and neutralizes its prey as soon as possible to avoid the mental anguish that comes with challenging their preconceptions of the world around it.

You're not wrong. I came for the tech, I stayed for the show.

It's possibly because your original downvoted comment was a vague hit out at 'the man' castigating the establishment and placing this trader in the same bracket as Madoff.

It had no facts, no real argument, was unspecific, ad hominem and didn't advance the debate in any particular direction that was useful or informative.

Much like your second comment.

You could have pointed, for example, to the rise of HFT, and perhaps asked why this particular trader was being penalised unlike several larger HFT operations when he was front-running or pulling trades before they could be completed. (See Flash Boys by Michael Lewis). Then, chances are you might have been upvoted.

We can take dissonance. We just prefer it to be informative.

It is unlawful to place orders deliberately intended never to trade. Obviously, every market maker has to place and cancel orders, but every one of those orders is submitted with the knowledge that they might trade, an eventuality the market-maker has to be prepared for.

In a layering spoofing scenario, the trader is deliberately structuring their orders to make it virtually impossible for them to trade. The point of doing that is to create the illusion of volume; the orders the spoofer places with actual trading intent will probably be in the opposite direction of their giant spoofing orders.

Obviously, lots of legitimate trading strategies rely on forms of deception (most often: "These piecemeal trades are not part of a concerted effort to buy a zillion shares, I promise"). Where the line is seemingly drawn is, your orders can conceal information from the market, but they can't directly inject fake information.

Here's a detailed SEC complaint, which accompanied a settlement, that does a decent job laying out how regulators establish someone's submitting orders with an intent to deceive rather than trade:

http://www.sec.gov/litigation/complaints/2015/comp-pr2015-4....

I think what he will claim is that his orders were not "virtually impossible" to trade with. The so-called "layer" orders were in the book for a reasonable length of time. His trading program was modifying the orders to keep them a few ticks from the inside, but they could have been hit by any contra side order that was big enough to trade through all of the orders in front of his. It's not unheard of for big trades to sweep multiple price levels in the ES book.

They will have to prove that his intent was to use the spoof orders to signal other participants to trade and I'm not sure that is as much of a slam dunk as all these news articles are claiming. I.e. there is no email where he just comes right out and says "these here are the spoof orders meant to move the price etc etc etc."

They're got something like 400 trading days worth of data to make the case with. If this spoofing case is like the last one I read, the spoofed orders aren't going to make any sense relative to the guy's assets, and his actual profits are going to contradict any thesis he proposes about the intent behind the spoofed orders. But we'll see, I guess.

Actually, what they've really got him on is him explaining the scheme in detail to a broker-dealer, describing it as, essentially, a conspiracy to manipulate the market and defraud other investors. This is much different than a case where all you have is trading records which might represent an algorithm detecting a reversing trend with no intention to manipulate. On the other hand, the CFTC is quite aggressive recently to deter behavior that doesn't seem particularly bad. Feels like a slippery slope from here to a place where breaking up large orders and routing to separate venues might be viewed as intent to defraud by concealing intentions.

This sounds like it may be possible only because of the exceptionally weak US-UK extradition treaty, which gives the USA unusually strong powers to grab people from the UK without any equivalent ability for the UK to extradite people from the USA. The one-sidednesss of this agreement has caused political problems before as it is seem as epitomising the slavishness of British politicians to Washington.

If "spoofing" was really only outlawed in the Dodd Frank act then it seems likely that most countries wouldn't have any laws against it, and an extradition would fail on the grounds that what Sarao is accused of would not be a crime in his country of origin.

I'm sure the UK Government will love the idea of sending a trader to the dogs in America regardless of whether or not this is actually illegal in the UK itself, just because it's election time and like all politicos everywhere they want to be seen as whacking financial types. But it seems to lead to a bad precedent. If trading strategies can be simply made illegal in a law so large even the people who voted on it didn't read it, then how can anyone really know if their trades are illegal or not?

The last time this was hashed out on HN, all the available documentation seemed to establish that the UK-US extradition treaty had recently changed --- to correct an imbalance that had historically favored the UK, and is now more equitable.

Equitable? Why does it matter if this is equitable?

In practice, UK citizens now have to obey US law or risk being extradited - in this case, for conducting their work legally in their home country.

This is not right.

You cannot be extradited from the UK to the US for things that aren't crimes in the UK. Check out the treaty.

This doesn't seem to me to be true in practice, e.g.

Not a crime in the UK at the time http://www.independent.co.uk/news/business/news/norris-loses...

UK citizen entrapped by US agent (entrapment would likely be thrown out of a UK court, but is now being tried in the US...): https://en.wikipedia.org/wiki/Christopher_Tappin

Even if the treaty is applied in this way, which it doesn't seem to be, US penalties are still not appropriate for UK citizens.

The UK has abolished the death penalty on human rights grounds - and yet the UK government can send someone over to the US for this punishment.

If you commit a UK crime on UK soil, you should be subject to UK punishment, and you should not be the subject of the law and punishment of any and all nations of the world.

Can you be clearer about what crime Norris is being charged with that is illegal in the US but not illegal in the UK?

Every European country extradites only on the provision that the death penalty will not be sought. It's also part of the ECHR.

Your last sentence suggests that your problem isn't the specifics of the extradition treaty, but rather the concept of extradition in general.

> Your last sentence suggests that your problem isn't the specifics of the extradition treaty, but rather the concept of extradition in general.

You misrepresent what I said. I don't particularly object to extradition for crimes committed in US law on US soil, although I do think countries have a responsibility to protect their own citizens first and foremost (which is surely the primary purpose of nations).

If an extradition treaty means I am effectively subject to inter-national law while in my home country, then I object, yes, very much.

Could you link to that discussion? The article here suggests otherwise:

https://en.wikipedia.org/wiki/UK%E2%80%93US_extradition_trea...

> If "spoofing" was really only outlawed in the Dodd Frank act then it seems likely that most countries wouldn't have any laws against it, and an extradition would fail on the grounds that what Sarao is accused of would not be a crime in his country of origin.

"spoofing" has generally been outlawed since forever - at the very least since 1934. Dodd Frank may have defined spoofing better, but "intent to execute" (which, in imprecise language is the complement of "spoofing") has been a requirement for much much longer, and in every regulated exchange in just about any country.

But even if it was only outlawed in 2010 and only in the US - Sarao was trading the CME -- that's the Chicago Mercantile Exchange, thereby putting himself in the jurisdiction of US trading laws.

We can't legislate that you don't spoof - it'll simply be an arms race. If you want to participate in a market without spoofers, simply require a deposit for offers cancelled within a certain period of time.

The problem is that the market is only capable of defending against what it is used to dealing with, if we "ban" scammers we end up with a market susceptible to scammers. If we "allow" everything we'll end up with markets that can withstand a sybil attacks. As another poster said, the "correct" metric is accepted offers, not outstanding offers. ("Correct" is complex of course...)

It's like passing laws making web-scraping for email addresses into hacking. It feels like it'll make us safer, but because "only criminals have guns" it just ends up shooting a few people in the foot and weakening us overall against actual threats.

> We can't legislate that you don't spoof - it'll simply be an arms race. If you want to participate in a market without spoofers, simply require a deposit for offers cancelled within a certain period of time.

That's like saying "we can't legislate vendors don't do a bait-and-switch" or "we can't legislate against false advertising". We can, we have, and it made life better for society at large. The false advertising arms race is still on - but the collateral damage to consumers is significantly smaller than it was before those laws were enacted.

We can legislate "don't spoof", and we have (for some proverbial we), but any law is useless without enforcement - and there is no enforcement for these regulations. Additionally, at present the SEC is a part of the spoofing arms race, and is massively (and some say willingly due to regulatory capture) underfunded and incompetent in that race with no intent to win.

You want to legislate it so that it has teeth? Publish pseudonymized order flow six months after the fact (changing identifiers every day, for example), and give someone who can use this data to prove spoofing 10% of the proceeds of any fines levied thanks to their work. (Sure, pseudonymity is hard, and may be impossible. But 6 months after the fact is enough time for that to matter little in case of trades).

You want to stop it without changing enforcement mechanism? Charge for unexecuted orders, or otherwise limit their numbers. In fact, some european exchanges have very stiff penalties if you have >10:1 ratio of orders:fills - and spoofing there is almost nonexistent.

But SEC behaviour indicates that they have no intention of fixing that aspect of the market.

> It's like passing laws making web-scraping for email addresses into hacking. It feels like it'll make us safer, but because "only criminals have guns" it just ends up shooting a few people in the foot and weakening us overall against actual threats.

No, it's not. Exchanges are few and (supposedly) extremely well regulated. The number of "bad actors" (spoofers or otherwise) is also small and under strict regulation.

Your claim that it is is nonsensical. An equivalent claim is "It's like making shooting people in the streets a crime. It feels like it'll make us safer, because only criminals will shoot people but it ends up weakening us overall against actual threats". We have made that a crime, we've done what's needed to enforce those laws, and it has made us stronger, not weaker.

> No, it's not. Exchanges are few and (supposedly) extremely well regulated. The number of "bad actors" (spoofers or otherwise) is also small and under strict regulation.

Almost everything you said is directly refuted in the article, or by some simple reasoning. Exchanges are not well regulated, as this happened for years, and they appear to have come down on the smallest offender. This is feel-good enforcement, and in no way comes with even an attempted guarantee that the behavior has been stopped. You're free to assume it has, but that's a probably niche view among people investing with their own money.

> Additionally, at present the SEC is a part of the spoofing arms race, and is massively (and some say willingly due to regulatory capture) underfunded and incompetent in that race with no intent to win.

Impossible mandates have a way of sapping your will.

> You want to legislate it so that it has teeth?

No, giving a bad law bigger teeth seems like a poor idea.

I want to examine the root cause of the issue and fix that once and for all.

The issue appears to be that someone decided that you should be able to trust my signals, "because trust, mkay". So inherently we've got a situation like poker where I benefit if you misread me, and we've got a re-active nanny trying to make me play nicely...

If instead we said to you "his buy/sell orders aren't an appropriate signal for you to base things on because they're easy to misread", we'd be done.

Then if you wanted a system where people didn't game it like this you could do what I proposed earlier and simply make each order cost a small amount if cancelled early or whatever. You'd know precisely how much (the early-cancellation fee) you could trust my signals. And then if enough people agreed, they'd start only accepting bids that had a deposit and you'd have your way, unlike now where you only think you do.

> An equivalent claim is "It's like making shooting people in the streets a crime [...]

Wow, no. Equivalency fail. Hitler much?

But though, if you trust in a prohibition on murder to keep you safe and refuse to defend yourself, perhaps there is some small amount of similarity...

Is the argument in your second paragraph still valid when we replace the word "scammers" with "murderers"? If not, why not?

I'm not the author of the original comment but I think scams are a lot more subjective and often are the result of a series of completely legal actions. The only identifying attribute of a scam be the intent of the actions, which is subjective. The act of murder is illegal in its own right, although sometimes justified in the legal system.

I think the original commenter's point was that if you take for granted that people will act ethically with no ulterior motives, you lose a healthy skepticism that is very beneficial to a market.

Are most analogies valid if you replace one party with nazis?

Murder isn't an economic action bordering on indistinguishable from normal interactions. Publishing some bids you don't really want accepted is.

Here are some copies of the complaints from our friends at the WSJ:

http://online.wsj.com/public/resources/documents/SaraoDOJ04-... (DOJ complaint)

http://online.wsj.com/public/resources/documents/saraoCFTC04... (CFTC complaint)

Reading through these it's fairly obvious that Sarao's layering was manipulative and illegal (it boggles the mind that they only caught up with him now), but blaming him for the flash crash is ridiculous over-reach.

Another factor to consider is Sarao's internet speed limitations. It seems that news outlets are reporting that Sarao "worked out of his house in London"... well I lived in that part of London and in 2012 it was hard enough getting anything near 60 Mb/s downstream, 10 Mb/s upstream and 15ms ping, so how the heck could he run a HFT operation?

Even if we consider the possibility of co-location it costs millions for this sort of access, and the capitalisation of his firm looks like it is sub £20mil between 2010 and 2011.

I just can't see how this adds up.

Colocation cost a couple grand a month. Its almost exactly the same price I paid for tier 1 internet colocation during the first dot com boom.

Well since we're talking about the e-minis lets assume the CME GLink... according to CME Group it would cost $12,000/mo with a one-time install of $2,000 and a minimum commitment period of 12 months. [Source: http://www.cmegroup.com/globex/files/connectivityoptions.pdf]

You then need the hardware and technical expertise to pull off the install... so lets say another $20,000 to $25,000 for hardware and I'd estimate $2,000 to $5,000 in technician charges.

So total outlay just to get the hardware set-up would be about $168k to $176k. Then there is the dev costs and I shudder to think about what is actually required here bearing in mind I think I do some pretty advanced retail trading in LISP.

Now, my point to all of this is when I step back for a moment and ask myself would someone outlay circa $150k on what is clearly a strategy that is blatant market manipulation, it defies belief that someone smart enough to pull it off would be dumb enough to try. I would even go further that this is especially true since there are many other opportunities to make honest money from the market trading legally.

You are assuming he is using the full connection and not using a hosted one. There are lots of options for spreading the cost of that around. See http://www.cmegroup.com/globex/trading-cme-group-products/tr...

If you read the actual complaints you will see it is explicitly outlined that he was using FCMs and was buying the trading platform. In fact one of the pieces of evidence that is mentioned are emails asking for feature requests from the software vendor.

Even if you grant the 150k capital outlay, the complaint alleges that this was a scheme worth 40 million dollars, a business opportunity many people would jump at. Further, the amount of orders he was placing would put a much higher capital constraint against his margin and clearing risk management than on the colocation.

My main point is that people act like getting colocation is the big barrier to entry for HFT and it simply isn't. Quite the opposite it is a business where most of the technology has been commoditized and the margins are razor thin. The barrier to entry is coming up with a trading strategy that works well on a risk adjusted basis, and spoofing may be one way to accomplish that.

Firms much larger than this one have been accused of similar behavior in the past.

I'm amazed that "spoofing" is illegal.

If your high frequency trading algorithm is so dumb that it can be tricked by an order that is then canceled, you deserve to lose money.

Anyone doing actual analysis of companies and trading accordingly would be completely unaffected by a spoofed order.

What sort of analysis are you suggesting? Because common analysis of supply and demand for instance is very much affected by a spoofed orders. If an exchange says that 10000 people want to sell at X and 2 want to sell at X-Y that is a pretty good mechanism for price discovery (something we want the markets to enable). That means nothing in the face of rampant spoofing. You might as well get rid of public price feeds at that point.

I think that there is an argument that spoofing should be legal, but we would need to de-anonymize the trades so that you could develop some digital analog for reputation. That way if I see an order from a known spoofer, I can discount it.

He's arguing that people should be trading based on an assessment of the actual value of the assets behind the stock, rather than the price that other people are trading the stock at.

Isn't that like arguing that we shouldn't have markets at all? The primary function of a market is price discovery. If we could all agree on the value of things, today and in the future, things would be a lot simpler.

But that price discovery should reveal the price that the underlying security is worth.

If shares were only traded based on what other people trade it for, isn't that the blind leading the blind, and the price it is traded for will have little grounding in reality.

I have several issues with HFT, however 'spoofing' as it is described sounds like it should be perfectly legal to me. If a player wants to trick the market, it is an irregularity that should be capitalised on by someone with a better understanding of the true value of the security

> But that price discovery should reveal the price that the underlying security is worth.

What does that even mean divorced from supply and demand? A share is worth precisely what someone will pay for it. You may think it will be worth more or less later, but right now that is what it is worth. That is the price discovery.

I think there is an argument for allowing spoofing, in that it is very hard to enforce.

But be clear, spoofing is bad for ALL market participants, not just HFT because it makes the supply/demand calculation more imprecise. This lack of precision leads to less fine grained pricing (in the form of a widened bid/ask spread) which hurts long term investors more than anyone.

You are ignoring the temporal element of the market; in the short term, prices are set by supply and demand, and trading to match supply and demand is exactly how market makers earn profits. The true value of a security can only be determined over a long period of time, which is why retail investors should pursue a "buy and hold" strategy. This is exactly what Warren Buffet does. The market is an ecology; the behavior of the ants is generally not all that relevant to the lions.

You can still gauge supply and demand -- but you would do so using actual trades, and not unfulfilled orders. This is how pricing works in most markets, by analyzing transactions, not offers.

It seems to me that a law that hinges on proving the intent of the trader seems destined for failure. It looks like this guy got caught by his incriminating emails and not based on his actual trading.

I like your idea of de-anonymized trades. Perhaps it could be something like the blockchain ledger.

> You can still gauge supply and demand -- but you would do so using actual trades

This actually happens a lot in some venues, especially dark pools. This leads to pretty high bid/ask spreads and lowered liquidity.

> It seems to me that a law that hinges on proving the intent of the trader seems destined for failure.

Which is the primary argument for making it legal in my mind, because a law that is unenforced only penalizes honest market participants. Another option would be to dramatically increase the cost of the punishment.

Another big problem is that even the idea of "intent" is in question when it comes to algorithmic trading. Absent a code comment //spoofing goes here or an email it seems like you are getting into deep philosophical territory.

> I like your idea of de-anonymized trades.

To be fair, there are lots of problems with it. Big block traders would absolutely hate this as it would give market makers such an advantage. Tracking reputation would become yet another advantage of sophisticated market participants that would be largely unavailable (or expensive) to other participants. Further, nothing would prevent reputation gaming, and detecting it may be even harder than spoofing.

Basically anything that increases the bid/ask spread is likely to benefit market makers at the expense of the rest of the market and simplicity is a huge virtue.

This is exactly the case, and it's why "painting the tape" has rightfully been illegal for quite some time.

The criminal complaint states that the trader was using an excel spreadsheet trading product, from a vendor, that was specifically augmented with order types that would automatically place orders for him at price points away from the market.

Does anyone know what vendor he was using to build this product?

Probably something like Stellar (Stellar Trading Systems) or TT (Trading Technologies XTrader).

He paid a company called Edge (http://edgefinite.com/) to extend the functionality of TT (https://www.tradingtechnologies.com/products/trading-analyti...).

Page 49 of full CFTC document - https://www.dropbox.com/s/ego1blhnylgs7sx/Appendix.pdf

"I asked Edge to design 3 more functions specifically to help try and hide my orders from these people. I do not know if this can be described as HFT, to me it is just giving me the ability to have some extra functions that my base trading software (TT) does not give me..."

From the actual complaints, it's not clear to me that Nav Sarao is being charged with helping cause the Flash Crash, but rather for flagrantly executing a spoofing strategy during the Flash Crash.

From the complaint:

35. According to Consulting Group analysts and the expert, that day, SARAO was active in the E-Mini market on the CME, and contributed to the orderbook imbalance that the CFTC and the Securities Exchange Commission have concluded, in a published report, was a cause, among other factors, of the Flash Crash. Among other activity, SARAO used the dynamic layering technique extensively.

...

36. SARAO's use of the dynamic layering technique was particularly intense in the hours leading up to the Flash Crash. SARAO used the technique continuously from 11:17 a.m. until 1:40 p.m.

Huh. And, from a much earlier source:

High frequency trading did not cause the flash crash according to a joint report by the CFTC and the SEC. The staffs of the two agencies concluded that a large fundamental trader's order to quickly sell 75,000 CME S&P 500 mini contracts (with a notional value of over $4 billion) created a "liquidity crisis" in the CME E-Mini futures that caused the price to drop more than 5% in four-and-one-half minutes during the most intense part of the episode.

Seems like an open glaring flaw in the trading systems if it can be manipulated in such ways. Why not focus on fixing such flaws instead of legislating against such trades?

So, the law says it's legal to place order and then change one's mind and cancel it, but is not legal to place order with the intent to cancel later? Seriously?

Seriously. It seems like there are many very easy technical fix the markets could implement --

- hard-cap the number of cancellations a trader can order

- penalize traders if they have too many cancelled orders (delay further transactions)

- add a cost to cancellation

Why does this require police action? Is the exchange just using the police as a bulldog to avoid having to implement technical fixes for these problems?

This would most likely have the effect of widening the bid/ask spread. Market makers wouldn't be able to react to changing conditions in the marketplace as quickly, and they would therefore set bid lower/ask higher than they would in a friction-free market.

Most of those do exist on exchanges. The mechanics usually involve a fee attached with your fill ratio (ie the number of orders/number that actually trade).

The thing about spoofing is that it doesn't require tons of cancels. Done right it falls well within the bounds of normal trading ratios.

In regards to the law, one is the normal behavior of market makers submitting their quotes, then changing the prices when the market moves.

The latter would constantly chase orders with cancels with only microseconds in-between them.

The best firms can respond to a tick in under 10 microseconds, so bad behavior really can't stay on the book for more than 10s of microseconds.

Nanex recommended the following to fix the problem back in 2010:

1. Quote and trade data must be time stamped by the exchanges at the time it is generated. This will ensure delays can be detected by everyone.

Reasoning: Changing the procedure to time stamp at the time a quote or trade is generated is a near trivial exercise. It probably comes as a surprise to many that time stamping isn't done that way now.

2. Quote-stuffing should be banned.

Reasoning: It is a manipulative device designed to overload the quotation system. Quote and trade dissemination (data feed) is a finite resource, and should be treated as such.

3. Add a simple 50 millisecond quote expiration rule: a quote must remain active until it is executed or 50ms elapses. If the quote is part of the NBBO, it may be improved (higher bid or lower offer price) at any time without waiting for the expiration period.

Reasoning: The exchanges must protect the integrity of the National Best Bid/Offer system. What is the point of having a National Best Bid/Offer, if not everyone in your nation (apologies to Alaska/Hawaii) can reasonably execute a trade against it? 50ms is approximately the time it takes light and electronic communication to travel from New York to California and back. It is impossible to transmit information any faster. This rule would not limit quote/trade rates. So long as trades are executing, quotes can update thousands of times a second. Only a small percentage of quotes today would be affected and the potential for catastrophically high rates would be eliminated.

Source: http://www.nanex.net/20100506/FlashCrashAnalysis_CompleteTex...

My 2 cents... the CFTC/SEC/DOJ have suffered too much embarrassment to ever adopt these sensible recommendations. Sadly I think we'll need to suffer a few more flash crashes before something is done.

> 1. Quote and trade data must be time stamped by the exchanges at the time it is generated. This will ensure delays can be detected by everyone.

Quotes and trades are typically generated on different systems. There are CAP ramifications for implementing this that may be much worse than the fix.

> 2. Quote-stuffing should be banned.

If you mean quote stuffing as DDoS, this is banned and is trivially prevented by the exchange. If you mean quote stuffing as in layering, that is illegal and is not trivially prevented.

> 3. Add a simple 50 millisecond quote expiration rule: a quote must remain active until it is executed or 50ms elapses.

This may or may not have CAP implications. I'm curious why 50 ms is chosen? Click traders won't be able to respond to this (and you've already mentioned Alaska/Hawaii) so it won't be "everyone" in any case.

Finally what does the NBBO have to do with CME trades? These were futures contracts that have no legal requirements with regard to the NBBO. They do of course have a correlation with the NBBO just like equities exchanges in Europe and Asia do. Do you intend to regulate the venues globally?

Apologies for any confusion... the recommendations were from Nanex [http://www.nanex.net/] and not my own. I quoted the recommendations directly from the Nanex report 'Analysis of the "Flash Crash"' which I will link here again for intellectual honesty http://www.nanex.net/20100506/FlashCrashAnalysis_CompleteTex...

What is the definition of "layering" here? I've seen this label applied to a technique that is used to ensure that orders are at or near the front of the queue on a price/time priority exchange as the market is moving. Essentially, this allows you to hold the option of whether to cancel your orders or leave them in place to facilitate your hedging as the market moves in a given direction; do you believe that this technique is or should be illegal?

Layering or spoofing in this sense is putting in many orders (or less commonly a few very large) orders at different "layers" of the order book with the intention to cancel them before they can trade. The purpose is to increase the perceived supply/demand and manipulate other market makers.

The problem is that the term quote stuffing is used to describe both this and the practice of spamming orders into an exchange (which I don't think happens much) for nefarious purposes.

What you are describing I've heard as layering and or stacking orders and is not illegal assuming your intent is to have those orders trade.

As always, definitions are important.

What you are describing is manipulation of what I've seen called "book" or "depth" micro price; the sort of thing that got Trillium in trouble (http://www.reuters.com/article/2010/09/13/financial-trillium...). I would expect everyone to take bid/ask + depth data with a large grain of salt if there are no accompanying trades, and use it only in a defensive fashion.

As for quote stuffing, this sort of thing was happening by accident in the early to mid 2000s due to the failure of some exchanges to upgrade their hardware infrastructure to handle legitimate volume. In this sort of environment, it would be easy to get away with spamming/DOS'ing the exchange to put competitors behind, but I doubt that this lack of investment still persists. Might still be possible in less-developed markets though.

Zerohedge gloating:

http://www.zerohedge.com/news/2015-04-21/step-step-guide-how...

Is there a difference between Layering/Spoofing and "quote stuffing?" I thought quote stuffing was a technique intended to overwhelm analytics software with volume of ticks rather than manipulate interpretation of current supply/demand dynamics. I suppose they are variations on the same theme.

I'm not sure I even see how layering and "quote stuffing" are related. Here's what Zero Hedge says:

In other words, 5 years after Zero Hedge first explained precisely what happened on May 6, 2010, the CFTC finally admits that the flash crash was not due to Waddell & Reed, but due to HFTs and quote stuffing.

That doesn't sound true at all. The point of the layering strategy is entering huge orders (not lots of little ones) that can't trade, and that's what SEC/CFTC said years ago as well. Immediately and automatically canceling the orders isn't required for layering to work; the layered orders are several price levels away from the clearing price. The point is to sustainably convince market makers of a change in the market's direction, and then profit from its correction.

As I understand it, the point of "quote stuffing" is literally to disrupt the market; it's akin to DDoSing the exchange and profiting from the change in behavior.

It's tricky, however, to nail this down, because the term "quote stuffing" was invented by people examining trading artifact data to explain anomalous graphs; the effect and even intent of "stuffing" is all supposition, and I guess the term can mean whatever Zero Hedge wants it to mean.

Layering does not have to be done with huge orders, lots of little orders work fine as well, as long as you can cancel them quickly (in some ways that is better because on some exchanges you get fills before public feeds).

Other than that I completely agree with you. Some people used quote stuffing to mean a DDoS style attack with orders (which if it ever happened was never wide spread and is trivial to prevent on the exchange side). Others do use the term for spoofing or layering.

How was he able to single handedly build up the volume necessary to affect what is literally the most highly traded futures contract?

I imagine he wrote some awesome C++ code plugged into MetaTrader or TradeStation, but certainly they have checks in place to ensure that he can't just place tons of orders that fake the volume of a whale without having lots of proxy servers that also place orders (which would also require proxy accounts). If so, wouldn't he have to have multiple accounts with his broker (I assume Interactive Brokers, given they have the only reasonable rates of all the retail brokerage firms).

Can someone shed some light? I know a little about finance and I'm genuinely curious how some random guy could possibly pull this off without some complicity with his broker.

Wait. So the apparent "crime" this person committed is the literal definition of playing bluff.

How is that even a crime ??

It seems the financial system is screwed up if someone playing bluff can crash it.

Every one of us has a vested interest in stable financial markets.

I am for any law that helps make financial markets serve businesses and society and not high-stake casino players.

We need laws like this because of the endless supply of greedy assholes who want to game the system. This extends to other areas like systematic tax avoidance used by rich individuals and large corporations.

Wait.

So you are saying to solve the problem of rich individuals and large corporations from using tax loopholes, we need to implement more laws ?

Do you seriously think that solution will work ? When was the last time you though looking at shitty code "hey ! do you know how to solve the problems in this shitty code, write more shitty code !"

I do not think the problem is as complicated as it is made out to be. The tax code is already freaking huge ! no human nor even the people making them can read all of it.

Every time you try to "fix" the problem by introducing more laws you make it easier for rich ppl to find new loopholes.

Its the same thing with MPAA trying to stop ppl from torrenting, or ISPs trying to block websites.

Interesting, but I can't see a solution in your reply.

Unless you are saying we should let the market be free, so we can have insider trading and big corps that pay less tax than a janitor (in $ terms not just %)?

Laws can be reformed. Ideally laws are tidied. To reuse your analogy: "hey ! do you know how to solve the problems in this shitty code, <<refactor>>"

yes ! refactoring.

I am not going to talk about my solution since its way too radical for HN. I do not think a libertarian approach will work.

What you want is to prevent having any single subset of the human population from having too much power. Any single government ( cough US, China, India ) or company.

We want the capacity to dynamically react to events much quicker.

Look at the largest companies on the planet, Its almost impossible for them to react to changes in technology etc.

And companies are not even that big compared to countries.

How long do you think its going to take the US,Indian,Chinese govt to implement a voting system that takes the same effort an facebook like ?

I know it sounds crazy, but its technically possible to do it.

If you break up all these large countries into small countries it wont be possible for things like wall-street to exist in the first place. Local voters can see their elected representative rather than have someone 1000s of miles away making decisions for them.

I derive these types of thinking from my experience in software and algorithm. when was the last time you saw a algorithm that does everything ? the best ones are small.

It also doesn't take away the capacity for humans to embark on large projects. If you can get concussion you can do all sorts of cool projects.

Anyway My thinking is really radical and its going to be 100s of years before we are there.

>> I am not going to talk about my solution since its way too radical for HN. I do not think a libertarian approach will work.

Libertarian seems to have different definitions depending on what I read about it, so I will just have to imagine what you mean :-o.

>So you are saying to solve the problem of rich individuals and large corporations from using tax loopholes, we need to implement more laws ?

Not just implement, but pursue and regulate aggressively.

>Every time you try to "fix" the problem by introducing more laws you make it easier for rich ppl to find new loopholes.

Not if you do it right. I don't know about the US, but the problem in the UK is that the same big accountancy firms that advise high net worth individuals about tax evasion also advise the UK government about tax laws.

Obviously they have a limited interest in producing strong laws that are actually effective - which is not a surprise when so many MPs, party funders, and Lords benefit from their services.

The real problem is that you can't make this work in a plutocracy.

It's a political issue caused by a deficit of real democracy and representation, not a legal issue caused by poor law making.

I'd guess Wall St works a little differently. But it sill seems as if there isn't as much oversight as there might be.

akhatri_aus's post has been incorrectly flagged as a dupe. If a mod is reading they may wish to look into that.

Check the comment history. Dupe is from another thread.

Fair point. It is a thread about the same topic (one that didn't receive a lot of activity). So I don't really blame them for trying to post it here.

Also a fair point. I guess the HN dupe algorithm is imperfect.

Reply on news.ycombinator.com