Working on a strategy on a 30min time frame trading the NQ emini.

Entering position as a market order

After entry I am using SetPercentTrailing TS function to preserve 15% of the profits.

Issue I'm having is it looks great when back testing. 85% of the orders are closed by the SetPercentTrailing function. However, when testing with live capital the SetPercentTrailing function seems to close the position early.

For instance today I used it on the NQ. Entered short at 4111. Covered at 4108 using the SetPercentTrailing function. However, when I turned automation off and then turned it back on, it said it entered at 4114 and covered at 4078.

So my question is why are back testing results showing more of a profit then live? I assume it has something to do with the SetPercentTrailing function. I understand why it shows a different entry price due to slippage. However, I can't confidently say 30 point difference on the cover was due to slippage. Therefore, I have to assume it's something with the SetPercentTrailing function.

Any thoughts or suggestions on this?

FuturesTrading,

the problem might be found in how you do your backtests and how the system is coded. For example when you use the build in exit functionality like targets or trailing stops you should always use the look inside bar backtesting with the highest resolution possible. Otherwise the platform can't detect how a bar evolved from open to close and make assumptions that not necessarily reflect the situation that was present realtime.

Regards,

ABCTG

I have had the same problem. In backtesting "SetPercentTrailing" exits at the close of the next bar. I turned on interbar generation ([intrabarordergeneration =true]), but "SetPercentTrailing" won't execute in that environment. My solution is to write my own code for a custom built "SetPercentTrailing" instead of using EL. It will be worth a try.

As @ABCTG said you need to enable "LIBB" or "look inside bar backtesting" otherwise the platform can't detect how a bar evolved from open to close and make assumptions that not necessarily reflect the situation that was present realtime.

Also a word of warning. Be very careful with percentage, multiplication and division of prices on continuous contracts. Every time there is a contract roll, the history changes, and hence your 'percentage, multiplication or division' changes as well. This may not be pronounced with equity index contracts but in some of the energy and metal contracts the effect can be significant.