“RSI(2) gives 90% profitable trades” — on Bitcoin (RSI(2) is a popular indicator strategy)
Source: a public post with an equity curve and the promise of a near-breakeven strategy.
What the stand showed
The claim
The author showed a pretty curve and promised that an RSI(2) strategy on Bitcoin wins 90% of trades. No fees, no slippage were in his math.
What we did
We took 700 daily BTC-USDT candles and ran the same strategy honestly — with fees and slippage (the gap between the expected and the real fill price). We checked for overfitting (tuned to fit the past, then fails on new data) and for cherry-picking the best variant (an SPA test against plain “buy and hold”).
What the stand showed
The share of profitable trades is 60%, not 90%. Profit factor 0.91: for every dollar won there is more than a dollar lost. Over 700 days the strategy lost 14%, and at its worst the deposit sank −36% (that drop is the drawdown). An independent test (SPA) gave a result of 0.51 — there is no real edge over the market, the numbers are explained by cherry-picking.
Verdict
Refuted. The claimed 90% did not hold up. The strategy is loss-making and has no statistical edge.
Seal · verify it yourself
SealedThe audit is sealed: SHA-256 fixes the content, the Bitcoin stamp fixes the date. Don’t take our word for it — verify right here.