Is your backtest real?

Upload a trade log. We check whether those prices actually traded, what the strategy looks like after real costs, and whether the numbers survive a hard look. Free, and we do not keep your file.

Be honest about the version count. Trying twenty things and keeping the best one is the most common way a backtest lies, and this is how we correct for it.

What we look for

Did those trades really happen?

For US stocks and ETFs with timestamps, we compare each fill to what actually traded that minute. A fill at a price that never existed is invented profit.

Broken test harness

Trades that never went against you by even a cent, exits before entries, win rates no real strategy reaches. These are bugs, not edges.

What it costs in real life

We rerun your results with real spreads and fees. Most retail edges are smaller than the spread, and this is where they disappear.

Are the numbers just luck?

We correct for how many versions you tried, then run a thousand simulations to see how much of the result survives.

A good score means we found nothing fake. It does not mean the strategy will make money. The only real test after this is trading it small and live.

Back to the risk checks →   How this works →