Does the MACD reading with the highest win rate make the most money?
MACD's Three Readings Tested: 2,052 Settings on USD/JPY
Finding
The reverse. On the 1-hour chart in 2025, the histogram turn won more than half its trades in all 57 settings (53.72 to 60.93%) and finished the year negative in 41 of them. The signal-line cross won 33.54 to 40.46% and finished positive in all 57. The turn is the only reading whose average loss is larger than its average win. The zero-line cross matched an EMA-versus-EMA crossover at the same periods in all 228 runs.
Key results
- 1-hour histogram turn, win rate
- All 57 settings above 50% (53.72 to 60.93%). 41 of them finished the year negative
- 1-hour signal-line cross, win rate
- 33.54 to 40.46%. All 57 finished 2024 and 2025 in profit
- Zero-line cross against an EMA crossover
- 228 of 228 runs matched on trades and pips
- 12/26/9 on the 1-hour chart
- Signal cross 451 trades, 38.58%, +2,223.6 pips. Histogram turn 373 trades, 56.84%, +761.1 pips
- 4-hour zero-line cross
- 0 of 57 settings finished both years in profit
Scope
- Instrument: USD/JPY
- Period: 2024-01-01 to 2024-12-31 and 2025-01-01 to 2025-12-31
- Timeframes: 15-minute, 1-hour, 4-hour
- Grid: 57 settings per reading per timeframe per year
Method
Run through Formiq's backtester at the close, with a fixed 0.3 pip spread and no slippage. Five fast periods x four slow periods (19 pairs with fast < slow) x three signal periods gives 57 settings, run under three readings across M15, H1 and H4 for 2024, 2025, and the two halves of 2025: 2,052 runs. The zero-cross comparison runs the same 19 period pairs through the moving-average rule with EMA averages. Every figure in the article is checked against the sweep output in research/macd-verify.run.ts.
Limitations
- One pair over two years. This does not establish a MACD setting that works everywhere.
- Fills at the close with a fixed spread and no slippage, so live results will differ.
- USD/JPY rose 1,632.5 pips over 2024, which favours trend-following rules.