Hype Tested

Claims tested

Does the RSI oversold strategy work on crypto?

Short answer: No edge: the difference could easily be luck.

"Buy when RSI is oversold" is one of the first rules people learn. We used the textbook settings (14 days, buy under 30, sell over 70) on Bitcoin and compared it with holding.

The rule we tested: Buy Bitcoin when its 14-day RSI drops below 30 (oversold), sell when it rises above 70.

No edge

Across 62 results the average difference was -0.73%, inside the noise band of ±2.96%. We can't tell it apart from luck: no measurable edge.

-0.73%average difference over 62 results. Anything within ±2.96% could be luck.
$0.5$0.75$1$1.5$2$320222023202420252026+102%RSI 14 (30/70)+109%Holding Bitcoin
Growth of $1, log scale. Dashed line = break-even.
RSI 14 (30/70) compared with Holding Bitcoin, 2021-08-01 to 2026-09-28
RSI 14 (30/70)Holding Bitcoin
Total return+102.5%+109.3%
Per year+14.7%+15.4%
Worst drop-50.5%-76.7%
Volatility+39.1%+52.0%

How we tested it

RSI (Relative Strength Index) compares recent gains with recent losses on a 0 to 100 scale, using Wilder's smoothing over 14 days. We buy when it closes below 30, stay in until it closes above 70, and trade at the next day's close, so the test never uses information it didn't have. Out of the market the money sits in cash at 0%. We compare it month by month with simply holding Bitcoin.

Period 2021-08-01 → 2026-09-28. Costs: 0.10% per round trip. Data: Uniswap v3 pool state on Ethereum mainnet, computed by Hype Tested, up to 2026-09-28. Engine 0.1.0.

Trades

BoughtSoldReturn
2021-08-032021-10-07+40.4%
2022-01-082022-03-29+13.0%
2022-05-102023-01-12-39.3%
2023-03-102023-03-18+32.3%
2023-08-182023-10-21+14.7%
2024-06-252024-10-30+16.6%
2025-02-272025-05-09+21.9%
2025-11-182026-01-15+2.6%
2026-02-012026-08-20-4.9%

What this doesn't tell you

More claims we tested