"I win more often than I lose, so my method works." It is the most common sentence among traders, and one of the most misleading. A method that wins 70% of the time can lose money. A method that wins 35% of the time can make money.
To know where you stand, you need a few figures, calculated correctly and read together. Here they are, what they measure, and the traps in each.
The win rate
win rate = winning trades ÷ total number of tradesIt is the most watched figure and the least useful on its own. It says nothing about the size of the wins and losses. A method that takes small gains and lets losses run shows a flattering win rate until the day one loss wipes out twenty gains.
It mostly serves one purpose: knowing what to expect psychologically. With a 40% win rate, streaks of six or seven losses in a row are normal. Better to know it before they happen.
The average win to loss ratio
ratio = average win ÷ average lossThe average win is the mean of the winning trades, the average loss that of the losing trades (as a positive value). A ratio of 2 means your wins are on average twice as big as your losses.
Win rate and ratio go together. The minimum win rate to avoid losing money, before costs, is:
minimum win rate = 1 ÷ (1 + ratio)| Win to loss ratio | Minimum win rate |
|---|---|
| 0.5 | 66.7% |
| 1 | 50% |
| 2 | 33.3% |
| 3 | 25% |
A trader who wins 70% of the time with a ratio of 0.4 loses money. A trader who wins 35% of the time with a ratio of 2.5 makes money.
Expectancy per trade
expectancy = net profit ÷ number of trades
= (win rate × average win) − ((1 − win rate) × average loss)This is the figure that sums up the previous two: what one trade of your method earns, on average. Example: 40% win rate, average win of $150, average loss of $75.
- Expectancy: 0.4 × 150 − 0.6 × 75 = 60 − 45 = $15 per trade.
A positive expectancy is the basic condition. Also compare it with the cost of a trade (spread, commission, swap): an expectancy of $2 with $3 of costs per trade is a losing method.
The profit factor
profit factor = sum of wins ÷ sum of lossesWith the previous example, over 100 trades: 40 wins of $150 ($6,000) and 60 losses of $75 ($4,500). Profit factor: 6,000 ÷ 4,500 = 1.33.
- Below 1: the method loses money.
- Between 1 and 1.3: it makes money, but a small deterioration (costs, slippage, an unfavourable period) can tip it over.
- Above 2 over a large number of trades: it is rare, and deserves to be checked. A very high profit factor over 15 trades proves nothing.
The maximum drawdown
Drawdown is the drop between a peak of the balance and the low that follows. The maximum drawdown is the largest of these drops over the period.
It is the figure that tells you what the method put you through. Two methods with the same final profit can have drawdowns of 8% and 45%. The second one requires watching almost half the account disappear before it comes back, which few traders actually manage.
Look at it as a percentage, to compare periods, and as an amount, to feel it. And remember: it takes +25% to erase −20%, and +100% to erase −50%.
Percentage gain, without deposits
An account that goes from $1,000 to $2,000 has not necessarily gained 100%: it may have received an $800 deposit. An honest return calculation neutralises deposits and withdrawals, by measuring performance between each money movement and then chaining the periods. It is the method fund managers use, and the one to use to compare two months or two methods.
How many trades before concluding
This is the most neglected point. Over ten trades, chance dominates: a mediocre method can show ten wins, a good method can show six losses.
- Fewer than 30 trades: no conclusion. Keep applying the rules without changing them.
- 30 to 100 trades: a trend appears, to be confirmed.
- More than 100 trades in varied market conditions: the figures start describing your method, not luck.
Changing method after five losses makes it impossible to know whether the method was bad or the streak was normal.
Breaking down to find what costs you
Once the overall figures are known, breakdowns reveal what works and what costs:
- By opening hour: many traders discover that their losses concentrate on one session.
- By day of the week: some days go badly for you, with no obvious reason at first.
- By symbol: the method that works on gold may lose on indices.
- By period: compare months with each other, and the last month with the rest.
A simple rule often comes out of these breakdowns: stop trading the time slot or the symbol that costs you. It is the cheapest improvement there is.
Streaks
Note the longest winning streak and the longest losing streak. The losing streak is there to check that your position size can take it: at 2% risk, eight losses in a row make about −15%. If that figure keeps you awake, it is your risk per trade that needs lowering, not your method that needs changing.
Reading the figures together
No figure is enough on its own. A complete reading, in one sentence, looks like this: "Over 140 trades in four months, a 42% win rate, a ratio of 1.9, an expectancy of $11 per trade, a profit factor of 1.38, a maximum drawdown of 9%, and trades taken after 4 pm lose money."
That sentence says everything: the method is profitable, modestly, with a bearable risk, and it has an identified weakness that can be fixed.
The takeaway
Keep your journal without exception, including for the trades you are not proud of. Wait for at least thirty trades before judging, and a hundred before concluding. Look at expectancy and drawdown before the win rate. And when a breakdown shows that a time slot costs money, believe the figures.