Average Result Per Trade (Expectancy), Explained
The number that tells you whether the wins pay for the losses, and how to read it on every Botsfolio setup.
Average result per trade, also called expectancy, is what a strategy makes or loses on an average trade after fees. We measure it in R, where 1R is the amount a trade loses if it hits its stop. A setup averaging +0.20R makes a fifth of its risk per trade over time. A win rate can't tell you this: a strategy can win 70% of its trades and still lose money if its losses are bigger than its wins. On Botsfolio, every setup's track record shows its average per trade, net of a 0.12% round-trip fee.
What R means
R is the risk on a trade: what you lose if price hits your stop. If a trade risks $100, then 1R is $100. A win of $250 is +2.5R. A loss at the stop is -1R. A trade closed early at a small loss might be -0.4R.
Measuring in R instead of dollars makes trades comparable. A $50 win on a trade that risked $25 (+2R) and a $500 win on a trade that risked $1,000 (+0.5R) are very different results, even though the second one looks bigger.
How to calculate it
The simplest way: add up every trade's result in R and divide by the number of trades.
If you know the win rate and the average win and loss, you can also use:
Average result per trade = (win rate × average win) - (loss rate × average loss)
Say a setup wins 40% of its trades. Its average win is +2.5R and its average loss is -1R:
(0.40 × 2.5) - (0.60 × 1) = 1.00 - 0.60 = +0.40R per trade
Over 100 trades, that's about +40R. If each trade risked $100, the record would be up about $4,000, before compounding and with plenty of losing streaks along the way.
Why a win rate isn't enough
A win rate only counts how often a trade closed in profit. It says nothing about how big the wins and losses were.
Strategy A feels better to trade. It wins seven times out of ten. It also loses money, because each of its three losers costs three times what a winner makes. Strategy B loses more often than it wins, and it's the one that makes money.
The same thing shows up in our own backtests. The first table lists setups that win at least half their trades and still lose money. The second lists the best averages per trade:
Of the 224 pattern, coin and timeframe combinations with at least 30 backtested trades, 82 win half their trades or more, and 34 of those still average a loss per trade after fees.
| Setup | Win rate | Avg per trade |
|---|---|---|
| Break and Retest SOL 1H 1,061 trades | 50% | -0.12R |
| Fair Value Gap ZEC 1H 957 trades | 53% | -0.01R |
| Fair Value Gap ETH 1H 924 trades | 52% | -0.06R |
| Liquidity Sweep SOL 1H 629 trades | 50% | -0.09R |
| Liquidity Sweep ZEC 1H 591 trades | 51% | -0.03R |
| Setup | Win rate | Avg per trade |
|---|---|---|
| Momentum Divergence SOL 8H 54 trades | 63% | +0.26R |
| Fair Value Gap ZEC 1D 37 trades | 65% | +0.26R |
| Wave Momentum HYPE 4H 43 trades | 58% | +0.21R |
| Wave Momentum ZEC 8H 46 trades | 63% | +0.19R |
| Wave Momentum SOL 12H 33 trades | 58% | +0.19R |
A win rate near 50% tells you almost nothing on its own. After fees, plenty of setups that win about half the time still lose money on average. The average per trade is the number that tells you whether the wins pay for the losses.
Turning R into money
If you risk 1% of your account on each trade, an average of +0.20R works out to about +0.2% of the account per trade. Over 100 trades, that's roughly +20% before compounding.
The path won't be smooth. Even a strategy with a solid positive average will string together five, eight, sometimes more losers in a row. How deep those stretches go is what drawdown measures, and it grows with the risk you take per trade.
How to read it on Botsfolio
Every setup track record shows its average per trade (on the setup card it's labeled avg edge per trade). A few things are already built in:
- Fees are included. Every result is net of a 0.12% round-trip fee, which is on the high side for most exchanges.
- Outliers are capped. Winners count as at most +5R, so one lucky trade can't carry the average.
- Sample size is shown. Fewer than 20 trades is marked provisional, and full confidence takes 100 or more.
- No look-ahead. A setup only enters on the candle where it actually formed, so the backtest never uses prices it couldn't have known.
You'll see the same number in the pattern and coin tables on the track record, and on every individual track-record page. The full method is on our setup performance methodology page.
Setup cards also show the typical dip first: how far price usually moved against a trade, in R, before it worked out. A setup with a good average but a typical dip close to -1R often comes within a whisker of its stop. That's worth knowing before you size a trade.
How many trades do you need?
An average from 10 trades is mostly noise. Two lucky wins can make a bad setup look great. As the count grows, the average settles toward what the setup actually does.
That's why every number on Botsfolio shows its trade count next to it, and why anything under 20 trades is marked provisional. When two setups have similar averages, the one with 300 trades behind it is far more trustworthy than the one with 25.
FAQ
Educational analysis, not financial advice. Backtests are tests on past prices. Past performance does not predict future results.