Backtest evidence

Can period 14 with a level of 25 be recommended for DMI/ADX, and what does the level actually filter?

Are DMI/ADX DI Crossovers Profitable? 1,344 Settings Tested

Finding

It cannot be recommended across timeframes. At period 14 and level 25 the DI cross was profitable in both 2024 and 2025 on the 15-minute chart, but the hourly chart returned 101 trades and −68.1 pips in 2025, and the four-hour chart was profitable in both years on only 19 trades in 2025. Of the 56 hourly settings, 9 DI-cross settings and 16 DI-comparison settings were profitable in both years. ADX stood at 25 or above on 40.51% of 15-minute bars, 47.24% of hourly bars and 37.58% of four-hour bars at period 14 in 2025, and requiring level 25 left 339 of 2,184 crosses, 101 of 534 and 19 of 141. The level keeps roughly half the market while discarding more than three quarters of the crosses.

Key results

Period 14, level 25 on the hourly chart
2025: DI cross 101 trades at −68.1 pips, DI comparison 181 at −388.5. 2024: +137.5 and +72.3 pips
Share of bars with ADX at or above 25
40.51% on 15-minute, 47.24% hourly, 37.58% four-hour
DI crosses that passed level 25
339 of 2,184 on 15-minute, 101 of 534 hourly, 19 of 141 four-hour

Scope

  • USDJPY, 2025-01-01 to 2025-12-31, with 2024 run identically for comparison
  • 15-minute, hourly and four-hour bars
  • 2 readings (DI cross / DI comparison) x 8 periods (5 to 50) x 7 minimum ADX levels (0 to 40) = 112 combinations; 3 timeframes x 4 windows = 1,344 runs
  • 0.3 pip spread fixed, 0.1 lots, filled at the close

Method

Measured with Formiq's backtester. The DI cross enters only on the bar where +DI and −DI change places; the DI comparison enters in the direction of the leading line whenever ADX is at or above the level. The only exit is the opposite DI signal, with entry and exit sharing one rule set. Fixed stops and targets are measured separately.

Limitations

  • One pair, two years.
  • Fills use the close, with a fixed 0.3 pip spread and 0.1 lots.
  • ADX(14) needs 28 bars before its first value, so warmup is loaded before each measured window.
  • The main sweep exits on the opposite DI signal; fixed stops and targets are measured separately.
  • ADX smooths the absolute difference between +DI and −DI, so it does not distinguish up from down. The level narrows the distance between the lines, not the direction.