HFT in the Eminis?


Jun 21

Here is a conjecture for how the HFT in the e-mini S&P 500 may be trading:

  1. A group of technical traders find a predictive pattern in the market or trade in similar fashion.
  2. Those traders act in unison at similar times or prices in the market. This creates an imbalance of buying with predictable stop prices and on book limit order prices.
  3. If the traders are correct then the market will move to the limit order take profits. The HFT detect both (1) imbalance in entries and (2) the stacking book. The HFT will  temporarily add additional size to the same price levels. This signals to other HFT traders where to trade from. This group of correct traders will not be able to get filled at the proper price.
  4. If the traders are wrong then the market will not move, if the market does not move to their target quickly then the HFT traders will add large size opposite their orders and buy/sell aggressively counter opposite. They will be forced to close at break even or worse prices.
  5. In either case, the HFT will load the book and execute aggressive buy or sell orders to run the stops.
  6. The net effect in either case is that becomes very difficult to get the good fills.
  7. As soon as the stops are run, the HFT will remove their limit orders and the market will resume the normal action. With no trade imbalance, the original market intention plays out with only strong hands left.

Other possible HFT patterns:

  1. Similar to the aggressive orders near targets, another set of HFT traders perhaps sensing what the more aggressive traders will do will pull the liquidity. This creates a vacuum like effect which allows the market to trade up to stop levels with little volume taking place.
  2. The above is also very effective method for getting winning traders to close their positions because they will see their profits evaporate and may get scarred out.
  3. The HFT trader knowing they caused the action can take advantage of both 1. and 2. to get good prices.

If conjecture is true, how can the small trader minimize HFT risk?

  1. Use larger stop losses. This requires taking a bigger risk.
  2. Try to trade using limit offset from what would be the original order. The downside is you may not get filled.
  3. Don’t display orders in the book. This is probably not going to help the small trader as in aggregate the positions can be inferred from the entries.
  4. Minimize optimization.

All of these “solutions” unfortunately minimize the profit potential. Do you think the HFT’s are active or is it just an excuse for losing traders?

Beyond the HFT risk, the small trader must also deal with the larger trader, who we’d like to track, who due to the HFT arms race, has most-likely became very adept at concealing their intentions by using the following strategies

  1. Only buying or selling when there is sufficient opposite side buying or selling the market in attempt to keep the volume delta, difference buys and sells, neutral.
  2. Only buying on new lows or selling on new highs.
  3. Not displaying liquidity and instead using iceberg orders to take positions.
  4. May have ability to create patterns in the market under specific conditions. Primarily overnight or near market closes. Primarily with intention of slowing down or shaking out weak hands trying to trend follow or trying to hit a target at or near market close.

As a tape reader, I am able to read and predict the market with high accuracy most-likely for similar reasons as to the HFT. However, the HFT has an unfair speed advantage plus they most likely use limit orders to additionally manipulate the market. Why does tape reading work? Let’s imagine there is some predictable pattern in the market. A systematic trader may do a lot of work to discover this pattern and then they will trade on it. If the pattern is very profitable, we can imagine that they aren’t the only ones to have discovered it. Other firms will be trading on it as well. So, instead of doing the hard work, all we have to do is figure out if traders are buying or selling and use that information to take out small profits from the market. Of course, it is one thing to be able to predict the market but it is another to be able to trade it profitably. The latter is much more difficult.

About the Author

The author is passionate about markets. He has developed top ranked futures strategies. His core focus is (1) applying machine learning and developing systematic strategies, and (2) solving the toughest problems of discretionary trading by applying quantitative tools, machine learning, and performance discipline. You can contact the author at curtis@beyondbacktesting.com.