A profitable track record feels like proof — but luck alone produces plenty of them. This tool compares your results with thousands of no-edge systems, then does the part that really matters: it allows for how many variations you tried to find this one. That's where a lot of promising-looking edges gently come apart — and that's completely normal, not a failing.
Each chart below is the running profit of a system with no edge at all. It uses the same win and loss sizes as yours, but the wins land at exactly the break-even rate — so on average it goes nowhere. We ran twelve of these pure-luck systems over your 40 trades and sorted them best to worst.
None of these has a real edge — yet the green ones look just like a winning strategy, and from the inside you couldn't tell them apart. That's the useful takeaway: a rising curve, on its own, can't tell you whether it's skill or luck. It's why the test above weighs your result against chance rather than taking the curve at face value.
Every time you change something and re-check whether the backtest looks better, that's one variation. They all count — including the ideas you tried and threw away.
Try enough variations and one will look great — through luck alone, even when none of them has a real edge. The more you try, the better the luckiest one looks. It's not a sign of skill; it's a lot of coin-flips where the best one gets remembered and the rest forgotten.
It's easy to tune and judge on the same data without realising. A real edge is found on one slice of history and still holds up on data you never touched (out-of-sample). If none was held back, every variation was fitted to the same noise — so the best one is bound to look good.
That's why this tool asks. A result can look like 1-in-500 — but if it took 30 tries to find, luck alone would manage it far more often. The deflated number is the honest one.
The method, honestly: we model your trades as wins of your average size and losses of 1R, then simulate thousands of systems with no edge at all (win rate set to exact break-even). "Luck probability" is how often a no-edge system matches or beats you. Then we deflate: if you tried N variations and kept the best, noise gets N chances to look good — so the bar to beat rises. A real edge clears the deflated bar; a mined one only clears the raw one. This is a teaching model on summary stats, not a full audit — but it's the same logic that separates signal from story. Not financial advice.