Updated Intraday Trend Systems

Having gone over the counter trend approach and finding I could not produce good results on any system, I turned my attention to the systems which try to capture some part of the intraday trend. A simple system was developed which uses the same entry methods as always. The system was optimized for one 18 month time period selected from the years 2000 through 2012 (no 2013 data). The optimized strategy was then applied to historic data from 07.01.1999 through 09.13.2013. Below is the chart from the TradeStation performance report.

This above equity curve is based on data which is 90% out of sample and using $10/rt commissions and $25/rt slippage. This is a reversing strategy, but it also has stops and will stop trading if the daily loss exceeds a loss target. With a 1.15 profit factor, this is not a particularly compelling strategy, but it certainly passes the first and most important test of being able to produce positive returns outside of the design period.

Next the daily returns from the strategy are screened with a regression filter. The purpose of the filter is to determine whether a day is a good trading day before the day begins. These calculation are all done in an Excel spreadsheet. After filtering the trades, the equity curve looks like this:

The regression filtering in this case is done on an two year walk-forward basis. The regression filter is applied to each calendar year based on filter criteria from the preceding two years. For year 2001, it is only one year because the only trade history available is the year 2000. After that, it steps forward two years at a time. The backtest results has no losing years. Most importantly, the current year, 2013 performance is up about $5K per contract and it is entirely based on strategy design which does not include any data for the current year.

This looks good, but it's still all history and hypothetical and experience teaches us that (1) prior performance whether real or hypothetical is go guarantee of future results and (2) returns are much more likely to be lower than higher.

The next step is to begin reporting results going forward on the website, but I expect this to replace nearly everything that is on the current SIFTradingsystem.com website as far as daytrading systems for the ES e-mini.

Review of the ES1 system

The poor performance of this system the past two years is the primary reason for the general overall systems performance. ES1 performance chart. This trading concept dates from the beginning of 2006. It has done well in every year since except for 2012 and 2013. The concept is very simple- fading breakouts. The chart image below illustrates how this works (click to enlarge):

The poor performance could be a result of some fundamental change in the market which cannot be accounted for by looking at market history or it could be due to hindsight bias in the system as it exists or both. To address these questions, I decided to go back to square one, so to speak, and rebuild the strategy from scratch and optimize incrementally and only report out of sample results. This is known as walk forward optimization.

My optimization process is a little different than some in that it has two distinct parts. First is the part of the strategy that determines when and how trades will occur within a given day. In the case of ES1 this is mostly determining where to set the buy and sell points. For this study I used a version of this strategy that has only one optimizable parameter (the width of the channel). The more variables you have to optimize, the easier it is to producing a winning strategy, but you can overfit the system to the data which we avoid by using just one parameter and by using only out of sample performance for hypothetical results.

The second part of the optimization involves deciding whether or not to take any trades at all on a given day. This is done with a regression filter which relies on a lot of previous market data  to predict whether a given day is likely to be profitable or not. The regression filter is an array of technical indicators such as moving averages, stochastics, calendar effects, and so forth. Without the filter, this system is not and would never be a profitable system. 

The method I've been using for the regression filter has always used nearly all available data in the calculation. For ES systems I've standardized on starting in the year 2000. I would then use all data from year 2000 up to the end of some past year, typically 2010 or 2011. This base period was used to "train" the filter. Any data following the training period was used as evidence of good performance on the part of the filter. If the out of sample filter performance was comparable to in sample performance, then the method was consider valid. Of course, there is always more out of sample data to come, and this is always the best test. Lately, it has been failing this test.

For this study, I decided to use only the period from 2009 forward. The first reason is that 2008 was a very unusual year and I was concerned that it might be having an undue influence on the filter. The second reasons is that I thought some of the very old market data might no longer be relevant. 

So starting in 2009, the single parameter was optimized for the first three months using only those three months. When the optimization was done the resulting trade sequence was used to design the initial filter. Then the initial optimizations were applied to the next three months which were not in the initial sample (2nd quarter of 2009). When this was done, another 3 months was added to the design period and the results were then applied to the next quarter. This was done for the entire year of 2009. This gave me three quarters of out of sample results. From this point, the in sample period was increased to 12 months going forward 12 months at a time until sampling was done through the end of 2012. This left all 2013 as out of sample. From this work an equity curve was pieced together which used only out of sample results. The chart below is the result of this process:

What the chart shows is the same deterioration in performance since the beginning of 2009. Below is the same chart, but using the actual strategy which is being reported on the website and has been for quite some time.

The two charts do not match up exactly in time or in magnitude, but it makes the point that while this type of trading worked in the past, it is not working now and is not suitable for trading. It may at some point start acting like it has in the past, but only time will tell. I have tried other means to see if could revive this trading method and while I can get the past couple of years to stop losing, I haven't been able to turn them upward. And even if I could, it would require the use of up to date market data, which still leaves one waiting to see if the redesign will hold up. 

That's enough on the trading method for the moment. My next post will be about the other way to trade the ES, and that is to try to capture some portion of an intaday trend.

RECALCULATING!

Because of the lackluster system performance across the board the past 12 months I am in the process of an extended review of the methodologies in system development. The primary focus will be making sure that out of sample performance from the design point holds up year after year and that everything is re-optimized at least once a year, but not to include data from the current year. The first system to undergo this review will be ES1 which was the first system ever developed. If this system cannot hold up under more rigorous backtesting and walked forward optimization, then the counter trend systems such as this may just have to be taken out of the mix altogether. Poor performance of the counter trend systems has been the biggest point of failure in the current environment.

More to follow....

More on Alligator drawdown

The chart below shows the backtest equity curves for ESCombo 1.0 and 2.0. They may be instructive relative to the current Alligator drawdown. The ESCombo 1.0 development was finished in 2009, but there was a pronounced drawdown at the end of equity curve back then that I was unable to get rid of. This was somewhat disappointing, but eventually the system started to move up and it became publicly traded in 2010. It didn't do much for a while, but then moved up, but within the past year it experienced a it's biggest drawdown ever. Since then it has moved back up to make a new equity high just recently. This kind of thing could easily happen with Alligator, so anyone contemplating getting off the train might find that things got a lot better after they did. Obviously, no one knows, but these things do not move in a straight line. Everyone must decide on their own what is the best course of action, and live with the consequences.

Gap Filling on the ES

In looking for reasons why the Alligator strategy is in a slump I've reviewed long term market behavior for gap fills. What it shows is that the amount of potential gain from fading a gap (selling an up gap at the open or buying a down gap at the open) is dropped to historically low levels in 2012. This is a little odd, because there has been a reasonable amount of volatility in the market otherwise. The Alligator system fades the gap almost every day, but because market behavior has changed, opportunities for successful gap fill trades have been curtailed. There is certainly no way to predict whether this is the new norm, but it seems unlikely to me.The number is moving up slightly for the past few months, a trend which I would like to see continue.

Comments on ES1

Systems have been in a drawdown for the past few weeks. Part of the reason is a recent drop off in performance of the ES1 system. ES1 is an important component of everything but Alligator. This system has three modes which are differentiated by the size of the stops. The predominant mode used is the one with tighter stops. It's returns are lower over the long term, but it's also less volatile. The ESCombo system has the version with large stops, so it has been helped somewhat lately as the system has begun to turn around. The charts below tell the story better than I can.

Alligator Slippage

A while back I increased the slippage plus commissions on the reporting for this system to $45 per round trip. Since this system has fallen off of late, I've been reviewing everything including the slippage and commission impact. You can see the effect from the chart below, which is for the past 12 months, more or less. Since this is having such an appreciable impact, the effort to switch from market orders to limit orders is going to be accelerated for this system. It's difficult to judge the impact of this before hand, because of partial fills which can occur with limit orders. However, since this means you are getting your order filled without your limit price being cleared, the effect should be positive.