hi,
How come the ninjatrader playback engine gives me different results?
I will come up with a strategy that has a 10 Profit Factor on playback but live its not profitable.
I ask chatgpt about it and it says the strategy results are too optomistic
This is what Ninjatrader says:
“realtime” and “backtest” are two different concepts. A backtest uses historical data. You can use a 1 tick series if you want in a backtest but that is still historical data. In manually coding a 1 tick secondary bar would let you execute code in OnBarUpdate at 1 tick intervals but you are still using the backtest engine which may show differences from realtime. If you want to test in specifically realtime you need to use a realtime mode like the playback connection, a live brokerage connection while using a live/demo/sim account.
Only 10 profit factor in backtesting? I’ve created strategies that will lose maybe once in 10,000+ trades in backtesting. 
You should read this. https://ninjatrader.com/support/helpguides/nt8/NT%20HelpGuide%20English.html?discrepancies_real-time_vs_bac.htm
There are lots of ways this can go wrong, but I would say the #1 greatest hit is using Renko bars for your strategy. Generally speaking, unless they’re extremely carefully designed, they test spectacularly in back-testing and then completely don’t work in realtime. So, if your strategy runs on Renko bars, that’s where you should start.
1 Like