Guide
What is a good Sharpe ratio? (And why the number can lie to you)
Sharpe ratio is the number every trader reaches for first, and the number most often used to make a mediocre strategy look impressive. It's a genuinely useful statistic — but only when you know what it's quietly assuming, and what it can't tell you at all.
Quick answer
A Sharpe ratio above 1 is generally considered acceptable, above 2 is strong, and above 3 is excellent — for a real, out-of-sample, cost-adjusted result. The number is easy to inflate by testing many parameter combinations, ignoring costs, or reporting only the in-sample period. A high Sharpe on an untested strategy tells you almost nothing.
What Sharpe ratio actually measures
Sharpe ratio measures return per unit of risk: how much reward you got for the volatility you had to sit through, expressed as a single number. A higher Sharpe means a smoother, more consistent path to your returns. It's a genuinely useful way to compare two strategies with different volatility profiles — a smooth 8% a year can be a "better" result than a wild 15% a year, and Sharpe captures that.
Rough benchmarks (with real caveats)
These are the numbers people commonly cite, and they're a reasonable starting intuition — but treat them as a first filter, not a verdict.
- Below 1: weak. The volatility you're taking on isn't being well compensated.
- 1 to 2: acceptable to good. Many solid, real strategies live here.
- 2 to 3: strong. Worth real scrutiny of how it was measured — this is also the range where inflated results start to look suspiciously good.
- Above 3: excellent if genuine, but increasingly rare and increasingly worth double-checking. Many published "Sharpe 4+" backtests do not survive out-of-sample testing.
Why a high Sharpe ratio can still be meaningless
The formula for Sharpe doesn't know how the number was produced. It can't tell the difference between a genuinely repeatable edge and a number manufactured by one of a few common mistakes.
It was the best of many attempts
Test a hundred parameter combinations and report only the best Sharpe, and you've selected for luck as much as skill. This is the multiple-testing problem, and it's why a deflated Sharpe ratio — one discounted for how many combinations were searched — is a far more honest number than the raw best result.
It excludes trading costs
A Sharpe ratio computed on gross returns, before commission and slippage, can look completely different from the same strategy's Sharpe on net returns. High-turnover strategies are hit hardest here — the more often you trade, the more the gap between gross and net Sharpe tends to widen.
It's measured in-sample
A Sharpe ratio from the period a strategy was tuned on tells you how well it fits the past. Only an out-of-sample Sharpe — measured on data the strategy never saw during development — tells you anything about the future. The gap between the two is one of the most informative numbers in the whole process.
It says nothing about sample size
A Sharpe ratio from twelve trades and one from twelve hundred trades are not equally trustworthy, even if the number is identical. Small samples produce noisy, unstable Sharpe estimates that can look excellent purely by chance.
What to ask before you trust a Sharpe ratio
Whenever you see a Sharpe ratio — your own or someone else's — four questions separate a real number from a manufactured one: Is it out-of-sample? Does it include realistic costs? How many parameter combinations were tried to get here? And how many trades is it actually built on? A Sharpe of 1.2 that clears all four is worth more than a Sharpe of 3 that clears none of them.
Or see the honest version automatically
The Honest Backtest Engine reports Sharpe both gross and net of real trading costs, deflates it for the number of parameter combinations you searched, and always headlines the out-of-sample figure — so the number you see is one you can actually trust.
See how it worksFrequently asked questions
What is a good Sharpe ratio for a trading strategy?
As a rough guide, above 1 is acceptable, above 2 is strong, and above 3 is excellent, for an out-of-sample, cost-adjusted result. The same number computed in-sample or before costs is much less meaningful.
Why can a high Sharpe ratio be misleading?
It can result from testing many parameter combinations and reporting only the best, excluding trading costs, measuring only the in-sample period, or being based on too few trades. Each of these inflates the number without reflecting a real, repeatable edge.
What is a deflated Sharpe ratio?
A deflated Sharpe ratio discounts the raw Sharpe for how many parameter combinations were searched to find it, correcting for the likelihood that the best of many tries looks good purely by chance.
Should I compare Sharpe ratios between different strategies?
Only if they were measured the same way, on the same type of period (both out-of-sample, both net of costs), and with comparable sample sizes. Comparing an in-sample Sharpe to an out-of-sample Sharpe is comparing two different things.