Quantis Trade
← Back to all metrics

Metrics / Hit Rate

Hit Rate

3 min read · Metric

Percentage of bars (or trades) with positive return. Useful, often misread, never sufficient on its own.

What it is

Hit rate is the fraction of observation periods where the strategy made money. The observation period matters: bar-level hit rate measures every bar in the backtest (e.g. every day); trade-level hit rate measures only the periods you held a position. They produce different numbers, and conflating them is the most common mistake.

50% is the coin-flip baseline. Above 50% means more bars positive than negative; below 50% means more bars negative. But 50% is meaningless on its own — what matters is the average size of the wins versus the losses (expectancy).

Formula

# Bar-level
bar_hit_rate = count(bars where return > 0) / total_bars

# Trade-level
trade_hit_rate = count(trades closed at profit) / total_trades

# Both reported as percentages 0-100

Typical ranges

  • SPY daily bars long-run: 53–54% positive (slight upward drift).
  • Mean-reversion strategies: 55–65% hit rate, small wins, occasional bigger losses.
  • Trend-following: 35–45% hit rate, big wins offsetting many small losses.
  • Cross-sectional momentum: 50–55% on monthly bars.
  • Above 65% on bars: usually indicates either a very low-vol strategy (lots of nothing-bars counted as tiny wins) or look-ahead bias.

Common mistakes

  • Confusing bar vs trade hit rate. A strategy that holds for 100 bars and exits flat counts 100 zero-bars in bar-level hit rate but 1 trade in trade-level. The two numbers can differ by 20+ percentage points for the same strategy.
  • Treating high hit rate as a quality signal. A strategy that wins 70% of trades but loses 5x on the 30% losers has negative expectancy. Hit rate without average win/loss is meaningless.
  • Comparing across timeframes. Daily bar hit rate is structurally lower than monthly bar hit rate (more noise in the daily measurement). Use one timeframe per comparison.

What the platform flags

Quantis reports trade-level hit rate alongside average win and average loss, so you can compute expectancy directly. A trade hit rate below 40% with a poor win/loss ratio is the trend-following pattern (many small losses, few big wins); a hit rate above 60% with the win/loss ratio inverted is the mean-reversion pattern (many small wins, occasional big loss). Both can be profitable strategies — the hit rate alone doesn't tell you which.

Further reading