Designing and Building a Profitable Automated Trading System

Discussion in 'Automated Trading' started by ScottD, Dec 9, 2008.

  1. On the YM it was a bit better. 11 trades, 131 points, 1.10 capture ratio, 75% in market.
     
    #201     Feb 9, 2009
  2. Suprisingly the model did ok today. Still using the same parameters since last Friday. For the ES, it made 12 trades for 9.25 points, but that only represented a capture ratio of 0.20 (not that good). It lost 8.75 points alone on the reversal on bar #20. One problem with an end of bar strategy, is that you have to wait for the end of the bar! LOL. Lost another 3.75 points on the chop after 3:00. Not sure what do do about conditions like that. I'm exploring some "adaptive" measures, but not sure if they will work. I'd like to hear from others on what they might be using. The key is to be able to recoginze periods of chop early enough so that you can exit until dominance is re-established.
     
    #202     Feb 10, 2009
  3. On the YM, the model didn't do so well but it did end positive. 12 trades for 27 points. Capture ratio was tiny (.07).
     
    #203     Feb 10, 2009
  4. For the NQ it was better. 16 trades for 30 points. Capture ratio was 0.47.
     
    #204     Feb 10, 2009
  5. The model did reasonably well today. 11 trades and 13.25 points. Capture ratio was nice: 0.79! A little chop around 14:00, but it made it up later.
     
    #205     Feb 11, 2009
  6. The model results for the NQ was:
    11 trades for 23.25 points
    0.71 capture ratio, 75% in market

    The model results for the YM was:
    11 trades for 88 points
    0.66 capture ratio, 78% in market.

    Given that I'm getting fairly consistent results across the three primary futures contracts (ES, NQ and YM), I believe the logic is reasonably sound.
     
    #206     Feb 11, 2009
  7. veggen

    veggen

    wow, thats some amazing results!!

    Allmost to amazing! Where is the downside??
     
    #207     Feb 11, 2009
  8. WOOO! Average profit of 2 points. Guess we can go buy that island! LOL!
     
    #208     Feb 11, 2009
  9. The system had about $10k per contract of downside. That's about downside of $10,000/$50/point=200 Points or 20%. Great, at least we're at a low....
     
    #209     Feb 11, 2009
  10. I have to admit that I'm rather suprised that such simple logic can produce this kind of results. But as you can see from my trade logs, I'm not hiding anything. You can see the indicator values, price, volume, etc for each and every bar. I can tell you though that a high volatility chop day will eat this alive. However, since one doesn't get those every day, I expect in the end it will average out, hopefully positive!
     
    #210     Feb 11, 2009