Last updated:
The Sharpe ratio of active management. How consistent your edge is.
Information ratio is excess return over a benchmark divided by the tracking error — the volatility of that excess return. It measures the consistency of outperformance: an IR of 0.5 is good for an active equity manager, 1.0+ is exceptional and very rare over long periods. Unlike Sharpe, IR compares to a benchmark rather than the risk-free rate.
IR = (Rp − Rb) / σ(Rp − Rb)IR = (Rp − Rb) / σ(Rp − Rb)Same shape as Sharpe but with benchmark substituted for risk-free rate. The numerator is the size of your active bet; the denominator is how lumpy that bet is.
IR is the right metric for evaluating an active manager. A manager who reliably beats the index by 1% per year with 2% tracking error has IR of 0.5 — a respectable, sustainable edge. A manager who happened to beat by 5% in one year with 15% tracking error has IR of 0.33 and almost certainly will not repeat.
Information ratios above 1.0 over 10+ years are extremely rare. Renaissance Medallion is famously above 2.0, but most "great" managers settle in around 0.5–0.7.
Your portfolio returned 13.0% vs. benchmark 11.5%. The monthly active return series has annualized standard deviation of 3.0%.
IR = (13.0% − 11.5%) / 3.0% = 1.5% / 3.0% = 0.50
That is solid — same range as a competent active equity manager.
Industry guidelines for sustained 10+ year information ratios:
| IR | Verdict |
|---|---|
| < 0 | Underperformed benchmark net of cost. Closet indexer at best. |
| 0 – 0.25 | Weak. Not statistically distinguishable from luck. |
| 0.25 – 0.5 | Decent active manager. |
| 0.5 – 0.75 | Strong. Top quartile of long-term active managers. |
| 0.75 – 1.0 | Exceptional. Top decile. |
| > 1.0 | Rarely sustained over decades. Audit before believing. |
Active Return · Tracking Error · Sharpe Ratio · Alpha (Jensen's)
Sharpe uses the risk-free rate as the baseline. IR uses your benchmark. IR is appropriate when evaluating a strategy that aims to beat a specific index, not absolute return.
At least 3 years for a stable estimate. 5+ years is preferable. IR is noisy at short windows.
In theory if tracking error is zero, but that would mean the portfolio is identical to the benchmark — IR of zero/zero, undefined. In practice tracking error is always positive.
Yes — against multiple benchmarks (S&P 500, VT, equal-weight, factor portfolios) so you can see the IR against whichever benchmark is most apt.
Upload your brokerage CSV — Foliolytic computes Information Ratio plus 70+ other metrics using real historical prices, real Treasury yields, and real CPI data. Free, no signup, your data stays in your browser.
Analyze your portfolio free →