Strange behavior when forward testing a strategy

[BoostTraderDE]

Hello everyone,

I am currently testing the strategy “EagleEyePeakHoursTradingNQ”, which shows a great performance in the backtest. The strategy also performs well in backtest in the YT videos of various testers.

When I activate and test (Demo-Account) the strategy on the NQ 15-minute chart with exactly the same settings, it opens countless trades in a row per candle during testing and closes them again very quickly.
In the backtest it opens only about 2-6 trades per day.

I also noticed that it aims for a take profit of 6 ticks, whereas the strategy is set to 24 ticks.

What could be the reason for these errors?

Details can be found in the attached pictures.
I am grateful for any help.

Best regards,
BoostTraderDE

[NinjaTrader_ChelseaB]

Hello BoostTraderDE,

It's possible the strategy is not implementing 1-tick intra-bar granularity and TickReplay and causing different behavior in real-time or backtest.

To understand what specifically is different it is necessary to write out the condition, bar, and order information to file to compare.

Below is a link to a support article with directions and a video.

https://support.ninjatrader.com/s/article/Developer-Guide-Comparing-Strategy-Results

And a link to the example script linked from the article.

https://support.ninjatrader.com/s/article/Real-time-Replay-Historical-Comparisons-Test

Please provide the debugging output from real-time and backtest and I can assist with analyzing to see what is causing differences.

[BoostTraderDE]

Hi Chelsea,

I have implemented the Strategie now with Calculation = On BarClose, as that is exactly, what the Backtest is doing also and it works.
Thank you for your answer.

Best regards,
Tim