A Case Study: Aggressive Gap Fill Trading System to Day Trade the E-mini S&P

By Lawrence

This is a case study of a trading system presented in the eBook, Aggressive Gap Fill Trading System to Day Trade the E-mini S&P, written by David Bean. The trading system presented in the book is a simple one and the past performance presented in the book is impressive. But there is a catch – commission and slippage are not accounted for. What if these factors are accounted for? Is the gap fill system presented really have an edge or is it just money management edge in disguise?

Disclaimer

It is my interpretation of the system thus it is fair to say that if my interpretation of the system presented in the book may not be exactly what the author has in mind. My opinion of the system is based on my interpretation thus ultimately it is my system being tested and discussed here.

The trading system presented in the ebook is not disclosed here. Those interested can use the link above to purchase the ebook from Amazon.

Not A Review

This short ebook contains one trading system as stated by the author. But in fact it is a trading system that trades in both directions. Thus the system has both long and short setups.

This ebook is an obvious attempt to create a lower price point product to see if more people will buy this ebook and eventually interested enough to buy the more expensive book, Seven Trading Systems for the S&P Futures, coming from the same author.

My take is that this ebook does not worth its price at all because I have written about gaps and their properties for years. In my opinion, my statistics approach towards gap behaviours is more useful than a static trading system. For example, my piece on inside gap behaviour tells you more than enough on how to play the particular type of gaps. And my statistics biases on these gaps was published well before this ebook was published.

In contrast, I find the book Seven Trading Systems for the S&P Futures is more thoughtful overall and fairly priced.

Remember that if you are not proficient with programming and basic statistics, build up the required skills before attempting to trade mechanical systems. It is very important to understand that majority of the trading models presented out there are often presented with favourable conditions to improve their appeals. Not knowing how to properly evaluate a trading system in the first place will give you false confidence in them.

The Ideal Performance

My implementation of the system results in the following performance chart with no commission or slippage. The data is Emini S&P 1-minute bars in Regular Trading Session only (9:30 am to 4:15 pm Eastern Time). The plots at the bottom of the chart are net gain in dollars.

image

I split the trading performance of the system into 2 parts so that readers can see how each signal performance historically.

Notice the way how the pair compensates each other during different phase of the market. It is a good sign for a trading model because it is how a system can maintain stability over the long term.

The Performance In A Tough Environment

Following is the same system with $2.5 per trade commission ($5 round turn) and 1-tick slippage per trade.

image

Notice what happened to the long setup performance? It is cut by $10,000. It is a very significant portion of the profit.

The performance of the short setup probably catch your attention first because it crashes hard in last 2 years.

Overall the system can still edge out a small positive performance but it is no longer impressive any more. In reality, if you follow the system exactly, the performance would be somewhere in between the ideal and the tough case scenario. The tough case scenario is not the worst case scenario. I have cut some slack for this system already.

For many marginal trading setups, there is a drastic difference between the ideal scenario and the tough case scenario. This is why I always backtest my trading models with slippage and commission enabled. It tells the truth. It tells you if the trading model can still perform in real life when slippage and other issues are bounded to happen.

The Impact Of The Money Management Rules

The trading system works pretty well even when the stop loss is tighten by 20%. Performance starts to degrade though as it approaches the target size.

Following is the performance of the system when 50% of the stated stop loss is used. Commission and slippage are excluded.

image

The long setup can no longer perform. Yet the short setup is not affected much by the tightening of the stop loss.

This tells you the 2 trading setups do not behave the same way. And it is absolutely normal because markets do not move up in the same way when they go down. It also tells you the long entry is less than optimal because it depends heavily on the money management edge to function.

Summary

The ebook Aggressive Gap Fill Trading System to Day Trade the E-mini S&P presented a nice concept of gap fill and illustrated the point that it is a profitable system under certain circumstances. The statistical biases is strong as I explained in my articles here long time ago gaps are things to watch out for. The system, however, is not robust as it cannot withstand the impact of slippage and commission.

Using this system as an example, I am showing what has to be done to evaluate the validity of a trading system. The stress test on various aspect of the trading setups reveals potential direction for improvements. This revelation is important for all system designers – make the historical backtesting more realistic and accept the harsh judgement from the results can open doors to much more profitable trading ideas. It is also better than wasting time and money on ideas that are marginal.

Share

Comments
  • smilingsynic September 8, 2014 at 7:54 pm

    Well done here.

  • Kel108 August 16, 2015 at 1:52 pm

    always amazed at how many publically traded systems charge monthly frees and high round turn commissions but present their historical results without commissions and slippage.

  • You must be logged in to comment. Log in