Best Performing System – How to Analyse??

• 12/07/2018 at 4:54 PM #86532
  GraHal

  It’s been quite a while since I dragged and dropped data into excel from PRT and attempted a meaningful analysis using Pivot Tables / Charts. I vowed I would never do it again … mainly due to frustrations resulting from 100 row limit of drag and drop (or is that 1,000 row limit?).

  I am at the stage where I want to know which of my Auto-Systems (if any! 🙂 ) are worth taking forward into Live Trading.

  Clearly there has to be a common baseline, re Lot Size, Date Started etc in order to make a meaningful comparison.  All these things can be easily achieved in excel but not much re meaningful analysis can be achieved by just using the PRT Platform on it’s own? Or am I missing a trick here?

  How do you guys do your analysis to know which is your best performing System when comparing many Systems all with each other?

  Thank You for Feedback

  

  GraHal

  Participant

  Topics: 203

  Replies: 8388

  Been thanked: 2034 times
  12/07/2018 at 5:49 PM #86535
  Vonasi

  GraHal – I hope you don’t mind but I have moved your post to the General Discussion section as it is not really a platform issue that can be resolved  -or not resolved!

  In answer to your question from my side only time can ultimately tell us what we want to know about a strategies performance and robustness. The ultimate OOS! I also tend to use my margin calculating snippet to see if a strategy is affordable to trade and that along with draw down eliminates an awful lot of very profitable strategies – which possibly says more about my bank balance and risk aversion than it does about a strategy!

  A lot of my strategies are based around a similar structure and so I generally know with a glance at the code how ‘risky’ a strategy is just by which bits of code for exit etc are in any particular strategy.

  At the end of the day forward testing gives us more confidence than any Excel spreadsheet comparison ever could.

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1810 times
  12/07/2018 at 5:55 PM #86536
  Inertia

  Hi Grahal,

  Good question,

  Of course the more money it makes the better it is however:

  To go from demo to live it should be first tested on 100+ trades mini.

  1. Best trades must be better then the worst trades
  2. Avg wining must be better the avg losses
  3. Drawdown must be reasonable… max 25% of the max run up…
  4. Consecutive wins better then the losses one…

  Then off you go! And just watch it once a week (if you can…)

  Cheers, Happy hour now !

  1 user thanked author for this post.

  GraHal

  

  Inertia

  Participant

  Topics: 40

  Replies: 223

  Been thanked: 52 times
  12/07/2018 at 6:31 PM #86538
  GraHal

  Vonasi wrote:

  forward testing gives us more confidence

  I agree and what I am trying to do objectively in a structured way  is to decide which few out of many systems making profit (on / after Forward Test) are the best.

  I’m fed up clicking on View Performance and trying to remember results to compare when I click View Performance for the next System which is also profitable.

  Excel Pivot Tables would facilitate System ranking after baseline (normalise?) of the performance criteria etc.

  It would be sooo good if we could sort the AutoTrading Running Systems by % of Winning Trades and Gain to Loss Ratio etc. I hope version 11 has biiggg improvements in the Auto-Trading window!?

  (I started this in this Forum and then as I thought about the above shortcomings I copy and pasted into the Platform Forum, but yeah no probs.  I probably need to (again) make the above suggestions in the PRT Platform Improvements Topic).    

  

  GraHal

  Participant

  Topics: 203

  Replies: 8388

  Been thanked: 2034 times
  12/07/2018 at 6:52 PM #86539
  Vonasi

  GraHal wrote:

  It would be sooo good if we could sort the AutoTrading Running Systems by % of Winning Trades and Gain to Loss Ratio etc

  I think everyone is with you on this one. It is probably one of the most information limited windows available in PRT. Along with the limited amount of results available in an Optimisation results window and the limited trade history in the performance graph window…. someone stop me or we will be here all day.

  Maybe Nicolas can ask PRT next time he speaks with them whether PRT have planned any improvements to the ProOrder Autotrading window in v11. Knowing something is coming even if we don’t know when is always better than knowing nothing at all!

  1 user thanked author for this post.

  GraHal

  

  Vonasi

  Moderator

  Topics: 192

  Replies: 5088

  Been thanked: 1810 times
• Author
  Posts