Difference between Strategy Tester and Live Market

[Andromat]

Hello,
I have found a significant discrepancy between back testing in Strategy Tester and actual auto trading the same strategy on live market on S&P 500 future. At about the same time span of roughly 23 hours, the strategy traded six times on live market with a negative Acc. Profit, but on back testing it only traded once with a large positive profit (see the highlighted cells in the attached file). I have checked all parameters and they are exactly the same. I have seen similar problem with CL and EUR/USD as well. What can be causing this problem and how can I prevent it, because it has created some doubts on the whole back testing performance of NinjaTrader.
Thanks

[NinjaTrader_Bertrand]

Welcome to our forums Andromat - our NinjaScript trainee will review and get back to you.

[NinjaTrader_CalH]

Hi Andromat,

Thanks for posting.

There are, expected, to be discrepancies when doing a Back-test Vs. Real-time live trading.

There many possibilities when you see discrepancies. They could due to when your order gets filled in live trade, if you get filled at the right price, running Calculate on Bar Close as True in the Parameters for the Stratrgy.

Please note that with FX spot currencies like EURUSD, that the spread is per se not figured into backtesting results. This can account for part of the discrepancies seen as well.

These are just some of the possibilities that you may run into when doing a back-test vs. Live Trade.

Please read our help guide section on Discrepancies: Real Time vs. Back-test
http://www.ninjatrader.com/support/h...ime_vs_bac.htm

Let me know if I can be of further assistance.