Backtest evidence

Which timeframe keeps Stoch RSI's four parameters profitable?

Stoch RSI Crossovers by Timeframe: 36 Settings Tested

Finding

Only the four-hour chart. On four-hour bars 30 of 36 settings were profitable in 2025 and all 36 in 2024, so 30 came through both years, against 3 of 36 hourly and none on 15-minute. The hourly version trades about 1,230 times a year and the 15-minute about 4,850, and the more it trades the more the spread decides. The four-hour default 14/14/3/3 returned +1,437.3 pips in 2025 and +2,430.4 in 2024. Hourly win rates in 2025 ran from 38.17% to 41.84% with a median of 39.8%, but the net result did not follow and only 3 of 36 were profitable in both years.

Key results

Profitable in both years
30 of 36 four-hour, 3 of 36 hourly, 0 of 36 on 15-minute
Four-hour default 14/14/3/3
+1,437.3 pips in 2025, +2,430.4 in 2024
Trades a year
about 1,230 hourly, about 4,850 on 15-minute

Scope

  • USDJPY, 2025-01-01 to 2025-12-31, with 2024 run identically for comparison
  • 15-minute, hourly and four-hour bars; 6,226 hourly bars in 2025, 24,903 on 15-minute, 1,610 on four-hour
  • 3 RSI periods (7, 14, 21) x 3 stochastic periods (7, 14, 21) x 2 %K (3, 5) x 2 %D (3, 5) = 36 combinations
  • 0.3 pip spread, 0.1 lots, zero slippage

Method

Measured with Formiq's backtester. A long fills at the close of the bar where %K crosses above %D and a short where it crosses below; the only exit is the opposite crossover. No stop, no target, no time exit. Level-based signals such as 20/80 were not tested.

Limitations

  • One pair, two years.
  • Stoch RSI implementations differ in how RSI is smoothed and how %K is taken. These figures come from Formiq's.
  • Signals are %K/%D crossovers only; level thresholds such as 20/80 were not tested and would reduce the trade count.
  • Entries and exits both fill at bar closes, and the spread is fixed at 0.3 pips — an assumption that matters a great deal for an indicator that trades this often.
  • Four-hour settings trade 230 to 397 times a year.