Ready to go further
Next: work out your own Sharpe ratio
The Sharpe ratio calculator takes a list of returns and shows the ratio and how uncertain it is.
What is included
- What it is, in plain words
- The formula or mechanism, with exact arithmetic
- How it hits grid, martingale and DCA bots
- A check you can run before you switch a bot on
Who it is for
- People who run or plan to run a trading bot
- Anyone reading a backtest or a bot sales page
- Readers who want the numbers, not the pitch
- Not advice, and not a signal service
Key facts
- The Sharpe ratio is the average return, less a cash rate, divided by the standard deviation of returns. It says how much return a bot earned for each unit of up and down movement.
- To annualise a monthly Sharpe, multiply by the square root of 12, about 3.46. This assumes returns do not depend on the previous month, which is often false.
- In the simulation, bots with no edge at all showed an annual Sharpe above 1 in 16.2% of 12-month records, 4.5% of 36-month records and 1.4% of 60-month records (calculated by us on 29 Oct 2026).
- Pick the best of 100 no-edge bots and its 12-month Sharpe averages 2.96. A shop window of bots will always have a star.
- A high Sharpe is not safety. Martingale, grid and option-selling bots can post a very high one until the day a large loss arrives.
The formula, with numbers
Annual Sharpe = monthly Sharpe x square root of 12
Take six made-up monthly returns: 2, -1, 3, 1.5, -2, 4 percent. The average is 1.25% and the standard deviation is 2.32%. With the cash rate set to 0 the monthly Sharpe is 1.25 / 2.32 = 0.54, and the annual Sharpe is 0.54 x 3.46 = 1.87. That looks good, and it comes from six months. The Sharpe ratio calculator does the same sums for your own returns.
What lucky records look like
To see how much a Sharpe ratio can mean on a short record, we simulated 5,000 bots at each record length. Each had a true average return of exactly 0, with monthly swings of 3%. Whatever Sharpe they showed was luck, because there was nothing to find.
| Months of record | Annual Sharpe above 1 | 5th percentile | 95th percentile | Best of 100 bots (average) |
|---|---|---|---|---|
| 12 | 16.2% | -1.84 | 1.78 | 2.96 |
| 36 | 4.5% | -1.00 | 0.97 | 1.54 |
| 60 | 1.4% | -0.76 | 0.74 | 1.13 |
With 12 months, 16.2% of bots with no edge had an annual Sharpe above 1, and the luckiest 5% were at 1.78 or higher. With 60 months, the range narrows to -0.76 to 0.74, but the best of 100 bots still averages 1.13. This is the same selection effect as in overfitting: test enough bots, or enough settings, and one will look strong.
What the ratio is good for
Used carefully, the Sharpe ratio does two jobs well. It lets you compare two bots on one scale when both ran over the same period with real costs, and it is not changed by leverage: doubling the size of every position doubles both the average return and the standard deviation, so the ratio stays the same. That second point is useful, because it separates how good a rule is from how hard it is being pushed. A bot with a modest ratio and a high leverage setting is taking a large risk for the return it earns.
What it cannot do is tell you how sure to be. The range in the table is the honest answer to that. A measured annual Sharpe of 1 from 12 months fits a bot with no edge and a bot with a real one, and the data cannot separate them. It takes years of real results, or a very large edge, before the number carries weight. Until then, treat it as a description of the past, and put more weight on the length of the record, the costs, and the worst month.
Where the Sharpe ratio misleads
- A short record. The table above. A 12-month Sharpe is a very rough estimate.
- Hidden tail risk. The ratio uses the standard deviation, which is small for a bot that wins a little almost every month and loses a lot rarely. See martingale strategy and the Sortino ratio, which counts only the downside.
- Smoothed or edited returns. Returns that are averaged, marked to a model or left out reduce the standard deviation and raise the ratio.
- Different periods. A daily, weekly and monthly Sharpe are not the same number, and annualising assumes independence from period to period.
- No costs. A record before spread and fees shows a Sharpe that real trading will not deliver. See slippage.
How to read one in a bot report
Ask for how many months it covers, whether it is after costs, whether the returns are real fills or a backtest, and how many versions were tried before this one. Then look at max drawdown and the worst month, and run the returns through the Monte Carlo simulator to see how much worse they could have looked.
What this page does not tell you
It does not say what Sharpe ratio is good. The returns are made up or simulated, the cash rate is set to 0, and this is education, not advice.
Keep Going
Related Tools and Guides
Keep exploring
FAQ
Frequently Asked Questions
What is the Sharpe ratio?
The average return, less a cash rate, divided by the standard deviation of returns. It measures return per unit of volatility.
What is a good Sharpe ratio?
There is no single good value. A Sharpe above 1 is often called good, but on a short record bots with no edge reach it by luck, so the length and quality of the record matter more.
How do you calculate the Sharpe ratio?
Take the average return per period, subtract the cash rate, divide by the standard deviation of the returns, and multiply a monthly figure by the square root of 12 to annualise it.
Can a bad bot have a high Sharpe ratio?
Yes. A bot that wins small amounts often and loses rarely and heavily can show a high Sharpe until the large loss arrives, and a short record can show one by luck.
What is the difference between Sharpe and Sortino?
Sharpe divides by the standard deviation of all returns. Sortino divides by the downside deviation, which counts only returns below a target.
Are the numbers on this page real?
Example numbers are made up to show how the maths works, and the page says so. A figure only counts as exact when the page says it was calculated on a stated date. Nothing you enter in a calculator leaves your browser.
Can I use this to decide whether to run a bot?
It can show whether a bot's numbers hold together and how much an account can lose. It cannot say what the market will do, and it is not advice. Use it to ask better questions, then read the terms of your own broker or exchange.
Is this financial advice?
No. It is general education about a calculation. The returns are made up or simulated and it does not consider your situation.
Know the Risk Before You Run It
Free calculators and guides, added every week. Leave your details and we will email you when new ones go up.
Trading carries a risk of loss