2 diff machines 2 different backtest results

[neb1998]

How is it possible if your using the same data provider, the same strategy, indicators, fills, etc that 2 identical systems tested on different machines produce different results?

The # of trades is only off by 2 total trades over 2 years but its worrying me as to why this can occur.

Any help would be appreciated.....Remember just a backtest to test settings, not an optimization.

[NinjaTrader_RyanM]

Hi neb,

There's a lot to line up between two NT installations to make the same. You likely have most covered but also need to look at things like computer time and time zone, NT version.

One way to help line up the NinjaTrader settings is to use backup / restore and you should generally expect similar results if everything matches up the same.

[neb1998]

Computer time i will have to check on, i have the rest synced. my computer time syncs up with my broker time though, so i cant imagine its different...could be though.

I will try the backup restore, its just that i develop on one machine and trade on another, and my first check is to make sure the backtest results agree on both machines before i modify the code to work in real time. One thing i can now say i do love about ninja, the trades will be generally identical in market replay (variance is less than 5% typically)...

Oh i do miss the days of MC and expecting a 20% drop in performance from backtesting to real time.