The Liquidity Lie

The same technique is worth sixty-two thousand dollars a go to somebody working a billion dollars of exposure and twenty-five dollars to somebody working two contracts. That is why the conduct actually proved in court runs at institutional size, against other machines, for a tick. And believing it was aimed at you is not merely unsupported: moving your stop out far enough to escape it gives up 0.46R on every trade you take, which is half the whole distance between the worst stop distance and the best.

Prerequisites: Lesson 25, which showed that a cascade through resting stops needs no author, and lesson 26, which showed that an order withdrawn before it can be taken is the ordinary state of a modern book rather than evidence of anything.

What a tick is worth, and to whom

The whole activity exists to capture a tick, so pricing a tick tells you who can be in it. Take the S&P 500 e-mini, because its arithmetic is fixed and public: the contract moves in quarter-points and each point is worth fifty dollars, so one tick is twelve dollars fifty. At an index level of 5,000 a single contract carries a quarter of a million dollars of exposure. Now vary only the size:

Contracts	Exposure	One tick	Two ticks
2	$500,000	$25	$50
5	$1,250,000	$62.50	$125
50	$12,500,000	$625	$1,250
500	$125,000,000	$6,250	$12,500
5,000	$1,250,000,000	$62,500	$125,000

The column that matters is the third one, and what matters about it is that it is a straight line. A tick is worth exactly two and a half thousand times as much to the participant working five thousand lots as to the one working two, because that is what multiplication does. But the cost of the technique does not sit on that line at all. The spoof order has to be big enough that the machines quoting against it change their minds, which is a property of the book and not of your ambition; it has to be exposed to being filled while it sits there, which is a real risk taken in real size; and the legal exposure at the end of it is a criminal one and is the same size whatever you traded. Those three costs are roughly flat. The gain is a straight line through the origin. So the activity does not scale down into small size and become a small version of itself — it stops existing, because below some size the flat costs are larger than the line.

Which is the answer to the question the story never asks. If a technique is worth sixty-two thousand dollars a go to somebody working a billion dollars of exposure and twenty-five dollars a go to somebody working two contracts, then the participants who do it are the ones for whom it pays, and they are transacting against the participants who are quoting at the touch. You are not in that trade in either role.

And now the part that costs money, which is what happens if you believe it anyway. Suppose the story persuades you to move your stop out of the zone lesson 25 measured — out to two ATR, where nothing can reach you. That lesson’s table already prices the move: expectancy falls from +0.75R at three-quarters of an ATR to +0.29R at two, which is 0.46R given up on every trade you take. The full span of that table, from its worst row to its best, was 0.92R. So acting on the story hands back half of everything the previous-but-one lesson found, on every trade, for ever, in exchange for protection against a technique that was not aimed at you and could not have been.

What this does not settle

That the enforcement record is the whole of the conduct. It plainly is not: proved and occurred are different quantities, prosecutions are expensive and slow, and the cases that reach trial are the ones with the clearest evidence rather than a representative sample. Everything above is about what the proved conduct looks like, which is the only sample anybody has. It would be a mistake to read that as a claim about how much goes unproved, and an equal mistake to read the gap as licence to fill it with whatever story you like.

That none of it ever touches you. It touches everybody, which is precisely the point being made rather than an exception to it. If the touch is displaced by a tick and you transact in that second, you got the displaced price, exactly like every other account standing there. What does not follow is that the displacement was arranged with you in mind, and it is the second claim rather than the first that changes what you should do.

That the rules are the same everywhere. The provision quoted above is American and dates from 2010; the European framework is separate and differently worded; other jurisdictions differ again, and venues have their own rulebooks on top. Nothing here is legal advice and none of it should be relied on as a description of your own market’s rules, which are a thing you can look up and this lesson is not a substitute for.

That deliberate deception is the only way to lose money at a level. It is nowhere near the largest way. Lesson 11 priced what an ordinary cascade costs the people caught in it, and that cost is paid on volume, every day, with nobody breaking any rule at all. A trader who fixes their view of intent and changes nothing else has fixed the smaller problem.

That any of this tells you what to do at a swept level. It does not, and deliberately so: this lesson is about attribution, not about entries. What a sweep is, when a reclaim is evidence and when it is noise, and what to require before acting on either are lesson 35’s subject, and that lesson works whether or not anybody was acting deliberately — which is rather the recommendation for it.

The market does contain people who break the rules. It does not contain anybody who has heard of you, and the difference between those two sentences is worth about half a unit of R on every trade you take.

Problems

1. Price a tick at your own size. Find the tick value of the instrument you actually trade — it is published by the venue and takes a minute to look up — and multiply it by the size you actually take. That number is what one tick of displacement is worth to you. Then divide the one-tick figure in the last row of the table above by it. The answer is how many times you would have to be on the wrong end of a one-tick displacement before it cost you what a single one is worth to the participants the enforcement cases are about.
2. Price the belief. Take your own version of lesson 25’s table, built in that lesson’s problem 2, and read off the expectancy at the stop distance you use now and at the distance you would use if you were trying to sit beyond anybody’s reach. The difference is what the story costs you per trade. Multiply by the number of trades you take in a year. This is the only number in the lesson that is yours, and it is the one that decides whether any of this matters.
3. Read one case. Pick any of the enforcement actions referred to above and read the charging document or the regulator’s order end to end. As you go, write down every distinct kind of evidence it relies on. When you are finished, look at your list and note which items could have been obtained by watching a chart. The list will be short and the point will be made better by your own reading than by this paragraph.

Sources. Eun Jung Lee, Kyong Shik Eom and Kyung Suh Park, “Microstructure-based manipulation: Strategic behavior and performance of spoofing traders” (Journal of Financial Markets, 2013), which is the closest thing to a direct look at the activity: account-level exchange data identifying who actually did it, what they traded, and how they went about it. Álvaro Cartea, Sebastian Jaimungal and Yixuan Wang, “Spoofing and Price Manipulation in Order-Driven Markets” (Applied Mathematical Finance, 2020), for the mechanism stated as mathematics rather than as a story — what a spoof has to do to the book to work, and therefore what size it has to be. Andrei Kirilenko, Albert S. Kyle, Mehrdad Samadi and Tugkan Tuzun, “The Flash Crash: High-Frequency Trading in an Electronic Market” (Journal of Finance, 2017), for a reconstruction of the most-cited day in this whole argument from the actual audit-trail data, which turns out to describe something rather different from the version that circulates.

Three lessons have now been spent on what the book cannot tell you. The next one is about the one moment when it can: price arrives at a level, size is actually spent there, and the outcome is either that the level holds or that it does not. That is the only test in this module whose result is a fact rather than an inference.

Educational only. Trading involves substantial risk of loss. Not financial advice. Past performance does not guarantee future results.