Trading Styles & Strategy Development · Intermediate · 4 min
Metrics: win rate, risk-reward, profit factor, drawdown
You can calculate win rate, average risk-reward, profit factor, expectancy and maximum drawdown from a list of trades, and say what each one hides.
From trade list to answer
A finished backtest is a list of trades: dates, prices, results. Scrolling through it, it is hard to say whether the rules are any good.
A handful of numbers turn the list into answers. Every backtesting tool reports them, and every strategy seller quotes the flattering ones. Knowing what each number measures, and what it hides, protects you from both your own optimism and other people's marketing.
How often, how much, together
Take the 30 fictional trades from the expectancy lesson: 12 wins of 2R and 18 losses of 1R.
- The win rate is the share of winning trades: 12 ÷ 30 = 40%. It says how often, not how much.
- The risk-reward in results is the average win divided by the average loss: 2R ÷ 1R = 2.
- The profit factor is gross profit divided by gross loss: 24R ÷ 18R = 1.33. Below 1, the rules lost money.
- Expectancy combines them: +0.2R per trade, before costs.
Two more numbers complete the picture:
- Maximum drawdown: the largest fall from a peak in the account, in percent.
- Longest losing streak: the most losses in a row.
The pain on the way
At 1% risk per trade, the account grows from 10,000 to about 10,584. The maximum drawdown is about 3%, during the losing runs, and the longest losing streak is three trades.
Each number hides something:
- Win rate hides the size of wins and losses.
- Profit factor hides the order of trades and how long the bad stretches last.
- Drawdown in a backtest shows only one historical order of trades. The same trades in a different order can produce longer losing runs and deeper drawdowns.
- All of them depend on the sample. Thirty trades are far too few to trust any of these numbers.
Read them together, compare them with the number of trades, and plan for a worse drawdown live than in the test.
Common mistakes
- Quoting the win rate alone. It is the easiest number to make look good.
- Ignoring drawdown. A rule set you abandon during its normal drawdown never delivers its expectancy.
- Trusting metrics from few trades. Small samples make all metrics unstable.
- Comparing before-cost and after-cost results. Always check whether costs were included.
The takeaway
Win rate, risk-reward, profit factor and drawdown each show one side of a rule set. Read them together, and over many trades.
Check yourself
Quick quiz
4 questions. Instant feedback, nothing is saved.
-
1Tap the largest drop from a peak.
Tap the chart.
The maximum drawdown here is about 3%, during a run of three losses in a row.
-
2Winners add up to 24R and losers to 18R. What is the profit factor?
Profit factor = gross profit ÷ gross loss = 24 ÷ 18 = 1.33. Above 1 means more was won than lost, before considering how reliable that is.
-
3Why is the win rate alone misleading?
A 70% win rate with small wins and large losses can lose money, a 35% win rate with large wins can make it.
-
4Why should you expect a deeper drawdown live than in the backtest?
The historical sequence is one of many possible ones. The lesson on Monte Carlo shows how much worse it can get.
of 4 correct.