Quantis Trade
← Back to all metrics

Metrics / Information Ratio

Information Ratio

3 min read · Metric

Sharpe's benchmark-relative cousin. Measures the return you earned per unit of how-much-you-deviated-from-the-benchmark.

What it is

The Information Ratio (IR) is active return divided by tracking error. Active return is your strategy's return minus the benchmark return; tracking error is the standard deviation of that difference. The IR tells you whether you earned the right amount of outperformance for the active risk you took.

Where Sharpe asks “return per unit of total risk vs cash,” IR asks “return per unit of risk vs the benchmark.” The two answer different questions and shouldn't be conflated.

Formula

active_return  = strategy_returns - benchmark_returns
tracking_error = std(active_return) * sqrt(periods_per_year)
mean_active    = mean(active_return) * periods_per_year

information_ratio = mean_active / tracking_error

Typical ranges

  • Active mutual funds (long-run): most are 0–0.3, a meaningful chunk are negative (after fees).
  • Decent active strategy: 0.3–0.7.
  • Strong: 0.7–1.0 sustained over multi-year samples.
  • Above 1.5 on long samples: rare, and likely subject to capacity constraints (the strategy can't scale).

Common mistakes

  • Confusing IR with Sharpe. A long-only equity strategy with Sharpe 0.8 and IR 0.4 is doing fine in absolute terms but only modestly beating its benchmark. Both numbers matter; they don't substitute for each other.
  • Comparing across different benchmarks. IR vs SPY and IR vs the Russell 2000 are not the same number for the same strategy. Always state the benchmark.
  • Ignoring the sign of tracking error. Tracking error is a magnitude — your outperformance and underperformance both inflate it. A strategy with 10% tracking error and 5% active return is taking big benchmark-relative risk for modest reward.

What the platform flags

Quantis computes IR against the user's chosen benchmark (default SPY) and shows it alongside Sharpe and beta. A strategy with high Sharpe but low IR is probably just delivering market beta well — fine, but not active edge. A strategy with high IR and low Sharpe (plus low beta) is delivering uncorrelated alpha — the rarer and more valuable profile.

Further reading