Backtest will be a disaster because (at least US Tech 100 cash 1€ and US 500 cash 1€) contain corrupted data at Sept 12 and Sept 16. The huge drop I showed yesterday also won’t help. Stop your backtest before Sept12 and I think it is fine.
My backtests as performed until Aug 7 in the past, just show as they did on that date.
The $1 variants also have the gap at Sept. 16 but the anomaly on Sept. 12 I don’t readily see.
The $20 and $100 variants I can not find any more (in Futures versions, yes, but I don’t think we talk about those – anyway the gap on Sept 16 is present for the (Sept 2025 and Dec 2025 (!)) Futures as well. Thus, a clear case of corrupted data throughout.).
Give it a try and let us know whether you now can deal with that. Side note : you shouldn’t be backtesting all In-Sample anyway, and you can just as well cut off early. For example, my own backtests (on all instruments) are performed until Aug. 15. This gives me a nice OOS hint to one month into the “future” (which for IG is not allowed to trespass Sept. 12, as it now appears).
.
.
