HFT, Market Elasticity, and Manipulation

Discussion in 'Automated Trading' started by MedianVelocity, Mar 5, 2014.

  1. This is a visual explanation of HFT exploratory trading and how it works. This is a simplification of the "Nanex Exploratory Trading" articles that the average Joe can understand. This isn't unique to HFT. This was done regularly by Salomon Brothers and other large players in the markets. I am not sure if it continues to this day except in HFT.

    So, lets begin... In scenario one... We have a inelastic sideways market. A steady sideways flow of 50 hypothetical buy and sell orders going past per minute that prefer to stick to a specific price. This market is very low elasticity and will not respond much to outside influences.

    Now, a bidder comes in. This bidder adds in 10 buy orders per minute in an attempt to bid price up. The market reaches a balance and refuses to go higher even as he continues to put in 10 buy orders per minute continually. The market has reached an equilibrium.

    He stops buying and price drifts back down to equilibrium. He then panics and price spikes downward sharply then shifts back to equilibrium. In this scenario, the manipulator lost money to the market. Nobody was playing against him. His attempt to bid up price out of equilibrium was his own undoing.

    However, in the real world... Price is rarely very inelastic. (Not to mention it doesn't naturally go straight sideways.)

    [​IMG]

    Scenario two... A manipulator comes into a elastic market. As he bids the market up the market responds well to his orders and continues to move higher without him increasing the amount of bids per minute he is contributing.

    After he is satisfied that he is pushing the elasticity of the market as far as it will go, he then begins to distribute. The market returns to equilibrium around when he distributes off the last of his position. It can go under equilibrium if he accumulated too much so he might sell off part of his position for a loss, but be overall profitable.

    Now, why is this all important?

    Certain HFT alogos are using elasticity testing to profit on a micro level. They are hitting the bid and analyzing the entire L2 to see if there was a reaction to their order that was sufficient. Eg... The market is elastic enough for a bigger order to be profitable. If so, the market is hit with a larger order. The market responds slightly, then the HFT program distributes off one tick higher.

    Salomon Brothers and other large players back in the day use to do this on a massive scale. They would shift the entire Forex market multiple percent.

    Now, what creates market elasticity? Passive order flow coming in through the day from people not watching the markets at all is inelastic. People trading off fundamentals and other things and not including prices in their decisions are inelastic. The elasticity comes from people watching prices and who's decisions can be influenced by prices changing.

    A lot of HFT is only profitable because people are closely watching L2's, bid/asks, volume and etc and can be influenced... Eg, they are elastic. They are preying on traders elasticity. Or in other words... Traders reaction to a rising price by buying, or a lowering price by selling.

    If nobody were to react, all "manipulative type" HFT alogos would go broke. Why is this important to the retail trader? DO NOT REACT TO TICK NOISE AND TINY MOVEMENT IN THE ES AND OTHER INSTRUMENTS... DO NOT react to shifts in bid/ask sizes. Just treat it as noise and plan off 5m charts and higher where trading is mostly fair and ethical nowadays. (As far as I know. Seen a little shady overnight stop hunting here and there by some big players.)

    Also, for those that actually do HFT and want to replicate what some big HFT's are doing... You need a LOT of capital do this kind of stuff. Millions... Also, as far as I know a lot of the tactics they are using should be or are illegal. They are definitely unethical when combined with L2 bid/ask quote spam obfuscation. That's what makes things real tricky... The best alogo's use a combination of technical analysis and exploratory orders.

    Anyway, good luck on your trading everyone, whichever side your on!
     
  2. Oh, and here is something to mull over. Short term traders that lose may not be losing to each other overall, they might mostly be losing to lower elasticity order flow.

    They may lose overall just due to the fact a good part of the market is not very elastic, and they are the most highly elastic part. That may be a little hard for most people to understand... Think it over for a while. :)
     
  3. Just as you state, I've heard that TA is used in some HFT programs. But I've also heard that it would be of no real use given the limited extent (tickwise) of the price movement that HFT covers. And then again I've read how the program can be designed to kick in on a break of support or resistance etc. I do notice that to occur, not always, but from time to time. Be that a part of a HFT program or done manually by a trader, I have no way of knowing. And I'm only guessing that it is HFT given the above normal speed of the trades following the break.

    As you seem know, just out of curiosity, how is TA actually designed to be used in HFT programs?

    My reason for asking is that other than extremely fast trades being made from time, I see no difference in the price movements nowdays from those of 40 years ago when I began using TA (just price and volume, no indicators). And if the trading based on TA in HFT programs was so well designed, I should never be able to get filled. But this is not a problem for me. So I guess it just comes down to the type of TA in the program and/or the TA knowledge of the programmer.

    Thank you.
     
  4. nth

    nth

    Or, maybe what if.. rationalization is a defense mechanism of the ego.. and maybe it's easier for a trader to blame hft or blame regulation or blame his wife or anything/anyone but take responsibility and realize that it's the strategy/tactic they use.. :)
     
  5. What HFT uses for "technical analysis" is more just short term focused mathmatical quant formulas, but it's still defined as technical analysis.