The Test Bench › indicator

Does RSI (14) actually work?

DEAD

No edge left once tested on a fair footing.

The claim

RSI (14) signals when to be in or out of the market — J. Welles Wilder

What we did

Tested against Buy & hold SPY, SPY 1993-2026, thresholds swept both directions, OOS>=2014, null-floor controlled.

The result

Result: 0.01 dSharpe (vs buy-hold), OOS

OOS edge +0.01 vs buy-hold; data-mining floor +0.35 (below it = noise)

Verdict

As a mechanical buy/sell timing signal on the index, RSI (14) scored +0.01 out-of-sample -- under the +0.35 data-mining floor. Overbought/oversold is a story the chart tells after the move, not before.

What the test did and did not do

survivorship-free universe
net of real costs
tested out of sample
point-in-time data
checked for look-ahead

Almost nobody publishing a backtest says which of these they did. We list the ones we failed as well as the ones we passed.

Tested 2026-06-23. The same treatment, applied to every other system, is on the test bench.

← Price vs 50-day MA All findings Rate of Change (12) →

Written with AI assistance from our own test database. The numbers are computed, not composed.
General market commentary for education, not personalised investment advice. Backtested results are hypothetical. Trading carries a real risk of loss.