Trading Strategy Performance Review – Part II

reviewIn this week’s earlier post, I shared the summary statistics of the trend-following strategy that I have been trading over the last three months.  Following that I spent some time reviewing the screenshots of individual trades.

In this article, I share some of the points coming out of this review.

Documentation of Trades

As I have been trading the strategy, I take screenshots of every single trade to allow me to review the trade further down the line (see example) – to see whether the strategy could be improved and/or whether my trading performance (i.e. my execution of the strategy) could be improved.090

The strategy is a basic trend-following one, trying to identify trending market conditions, looking for a pullback in price before getting into trade.

In addition to screenshots of completed trades, I can also analyse trading signals generated with code on my TradeStation platform – which is very similar to reviewing actual trades.

The strategy & goals of analysis, work completed

To reiterate – the strategy is not an automated one, however it is partially based on code.  Setup signals are generated using EasyLanguage code run on the TradeStation platform.  The signals were then either used for entering trades or skipped.  Trades were entered, managed and exited manually.  The code also gave the stop and target levels for each trade.  For the most part the trades were managed in a very consistent manner.  My goals in reviewing the trades were as follows:

  • Can the performance of the strategy be improved by altering the setup criteria?
  • Can the % of false positives be reduced  (‘false positives’ are trading signals that I skip)?
  • Can the performance of the strategy be improved by altering the trade management rules?
  • Can the strategy parameters be altered to become more suitable for automated, as opposed to manual, back-testing projects?

Here’s what I did:

  • I divided the screenshots into winning and losing trades.
  • I reviewed the classification of all the 200+ trading signals over the past 3 months and categorised them into valid and false positive signals.
  • I compared the profitability of valid and false positive signals.
  • This generated several ideas for what variables could drive performance –
    • the price of the entry from the fast moving average
    • the size of the pullback in price
    • the sharpness of the pullback
    • the degree of pullback already recouped at the time of entry.  These were the most logical and obvious points to look at.
    • When analysing trading signals, the above were the factors that I kept on using in judging setups.
  • Wrote further code to collect information for measuring these variables for each trading signal.
  • Looked for correlation between variables and profitability.

 

Findings & Conclusions

Unfortunately, after going through all these steps, I couldn’t find any correlation between the above factors and the profitability.    I also couldn’t find any difference in performance between valid signals and false positive signals.  Yes, I was able to articulate in more detail what i like to see in a setup – that was definite progress.  However, the expectation that I would have more profits (or less losses) when focusing in on the ‘top quality’ setups, just didn’t eventuate.  The results from trading those desired setups are still more or less random.

I then also refined the code so that the trades would be entered earlier – but this also didn’t create a clear headway.

Put this strategy to one side for a while

I think in terms of using my time effectively, I might just put aside this trading strategy for now.  I would love to work on it further, once TraderDi becomes available to work on it together, given that she has been trading these strategy profitable consistently for many years.  It would be good to review the generated signals with her on a weekly basis and adjust the code accordingly.  On my own, I don’t seem to making headway with it!

Thus, although I managed to squeeze a profit out of the market trading this strategy fairly consistently, I have not been able to show consistent profits with it in a simulated, nor in a back-testing environment. Hence, I can’t say that I have an edge with that type of trading.  I feel the setup has become clearer in my mind, and I have been able to articulate it better and more clearly in code.  I also built some great processes for executing the strategy on the 4H timeframe from one day to another.  And I had good discipline in carrying out all these processes.

All that said, it still seems a struggle – and it might be good to put the strategy aside for a little while.  Hence I will wrap up all I have done on this, tie up any loose ends, so that I can pick it up again further down the line and know where I stopped (hopefully together with TraderDi).

Lesson learned & next trading focus

There’s a great deal of stuff I progressed with from working on this strategy the past several months (mostly on a part-time basis):

  • Coding skills improved
  • Ability to articulate trading ideas into code
  • Building processes
  • Heavier usage of TradeStation platform
  • Discipline
  • Simplicity in operations

All of these will benefit in other areas of trading that I will research and develop.

Whilst I have been doing that, I have also diligently been part of the SmartForexLearning trading room, in which Felix guides us in the art of trading reversals in various markets on various timeframes.  Here also I managed to build, and am still building, processes to make the trading more efficient and routine-based.

Given that Felix gives a lot of his time in the room day in and day out, I think my time might be better spent focusing on taking the trades there and applying all my coding and other trading skills in that environment for now.

And that’s today wrap.  Happy weekend everyone!

 

This entry was posted in TF007 Strategy, Trend-Following Strategy. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.