Spotting a heart-shaped cloud in the sky

Last week the Foresight Project published its written report on HFT (see previous blog Some things in life are bad…) which gave a balanced picture with carefully drafted conclusions.

However, during a presentation of the report’s findings on 23rd October, there was one particularly interesting issue that caught my attention. One of the speakers referred to a case, presented by Nanex, on natural gas futures trading on NYMEX in June 2011. It was used during the talk to highlight the risk of trading strategies self-enforcing. Apparently, the picture clearly shows that some algo(s) made prices oscillate around the fair value until someone realised what was going on and turned the algo(s) off. This algo behaviour distorted markets and created price inefficiencies.

Source: Nanex, LLC

I believe this conclusion is invalid, based on simple statistics. Tossing a coin is a random game in the sense that you don’t know whether the outcome will be heads or tails, but you do know that there is a 50/50 chance that it will be one or the other. If you throw heads 10 times in a row, some observers may start to think that there is something wrong with the coin. This effect is called Apophenia, a term coined by noted German neurologist Klaus Conrad to refer to “the unmotivated seeing of connections”. Interestingly, this seeking of meaningful patterns in random data happens surprisingly often. The chance of throwing 10 heads in a row is 1:1024. Thus, if you throw a coin 51,200 times, you would likely see a sequence of 10 heads in a row about 50 times.

The same principle can be applied to the chart above. Price changes are assumed to be random and most of us cannot predict whether prices will tick up or down in the next moment. There is 50/50 chance of the market going either way. The probability that this graph represents pure coincidence may be really small, but it only displays prices in one instrument across a 15 minute interval. There are millions of instruments traded every day so, if you look at enough 15 minute price charts, you might just find some that seem to “prove” your belief, while being nothing more than the product of pure random chance!

Claiming that this case tells us anything about HFT is a bit like seeing a heart-shaped cloud in the sky and concluding that cupid really does exist. Speaking strictly statistically, you would need to see a whole lot more hearts in the sky before you can safely rule out pure chance or the wishful thinking of a die-hard romantic.

3 Responses to “Spotting a heart-shaped cloud in the sky”
  1. C Furlong says:

    What is fascinating about many of the NANEX pictures is that they give the viewer a sense of what algo behaviour looks like. Quoting and pricing behaviour that is both outstandingly fast and often inane. I agree you cannot really draw many solid conclusions about the various pretty pictures that are drawn but they do provide some sense of the madness of market instrument pricing in the age of computer trading.

  2. Bruce Bland says:

    In my opinion it was likely to be a single algo that was responsible for the chart provided by NANEX. The algo itself entered a simple feedback loop where new orders it created where subsequently used in its own prediction system, generating a loop. Whoever started it running must have spotted it and stopped it, realising what it was doing.
    What this chart shows is that untested models do make it out into the wild even if they do get killed off quite quickly!

  3. Christian Voigt says:

    Hi Bruce

    I take from your comment, you are convinced that this particular case is caused by an algo because the pattern fits perfectly well to a reasonable explanation. However, I see no argument why this pattern couldn’t be caused by random chance. Agreed, the likelihood that it is caused by chance is smaller than winning the lottery jackpot. However, there are millions of charts created every day. Thus, who is to say this is not random chance?

    Going beyond the focus of this particular picture. I feel at the moment there is no real academic evidence on the negative impact of HFTs. Rather the opposite is true, many comprehensive studies show that HFTs improve market quality and efficiency. However, it seems that one picture is sufficient to condemn a whole industry. Has it gotten so far, that anecdotes are sufficient to define regulation for all of Europe?

Leave A Comment

Copyright © 2017 Fidessa group plc. All rights reserved.

The information contained within this website is provided for informational purposes only. Fidessa will use reasonable care to ensure that information is accurate at the time it is made available, and for the duration that it remains on the site. The information may be changed by Fidessa at any time without notice. We also reserve the right to close the website at any time. No representation or warranty, expressed or implied, is given on behalf of Fidessa or any of its respective directors, employees, agents, or advisers as to the accuracy or completeness of the information or opinions contained herein or its suitability for any purpose and, save in the case of fraud, all liability for direct, indirect, special, consequential or other loss or damages of whatever kind that may arise from use of the website is hereby excluded to the fullest extent permitted by law. Any decisions you make based on the information in this website are your sole responsibility and information on the website should not be relied upon in connection with any investment decision.

The copyright of this website belongs to Fidessa. All other intellectual property rights are reserved.

Reproduction or redistribution of this information is prohibited except with written permission from Fidessa.