Backtest evidence

Is there an MFI setting that stays profitable from one year to the next?

MFI Oversold Entry Frequency: 240 Settings Tested

Finding

This test did not find one. On 15-minute bars the textbook long setup — period 14, buy below 20, close above 80 — took 145 trades for +1,379.1 pips in 2025 and 113 trades for −654.2 pips in 2024. Many settings barely fire: on four-hour bars 48 of 240 took no trades at all in a year and 137 took fewer than ten; on hourly bars 20 took none and 81 fewer than ten. Restricted to settings with at least 20 trades, the hourly long side in 2025 won 47.5% to 72.73% of trades, yet the median net result was −304 pips and only 13 of those 65 settings finished positive.

Key results

Textbook long setup on 15-minute bars
145 trades and +1,379.1 pips in 2025; 113 trades and −654.2 pips in 2024
Settings that did not fire on four-hour bars
48 of 240 took no trades; 137 took fewer than ten
Long side with 20+ trades, hourly 2025
47.5% to 72.73% win rates, median −304 pips, 13 of 65 positive

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
  • 6 periods (7 to 40) x 5 entry levels x 4 exit levels x 2 directions = 240 settings, per timeframe and per year
  • 0.3 pip spread, 0.1 lots, zero slippage

Method

Measured with Formiq's backtester. The long side buys when MFI drops below an entry level (10 to 30) and closes when it rises above an exit level (50 to 90); the short side mirrors it. No stop or target. MFI is RSI weighted by volume, and forex has no consolidated exchange volume to weight by.

Limitations

  • One pair, two years.
  • Forex volume is not a market-wide figure, and MFI's calculation rests on it.
  • Entries and exits both fill at bar closes.
  • Settings with very few trades are included in the aggregates; the figures excluding them are given separately in the article.