1
Expectancy
Average result per trade after wins, losses, and costs.
Lesson · Backtesting and Trading Systems
Performance Metrics and Expectancy
Calculate win rate, average win, average loss, expectancy, profit factor, return, volatility, trade frequency, and sample size without relying on one metric.
Core concepts
What to understand
Keep each definition operational so it can be implemented and tested.
2
Payoff and frequency
Win rate, average size, and opportunity count work together.
3
Uncertainty
Small samples can show unstable metrics that change with a few trades.
Interactive lab
Performance Metrics Lab
Adjust win rate, payoff, costs, and sample size. Observe expectancy, profit factor, equity, and result uncertainty.
Performance Metrics Lab
Deterministic synthetic educational data · no live market connection
BACKTEST MODEL
Process
Use a metric set
No single number describes return, risk, and reliability.
Measure edge
Calculate expectancy and gross versus net payoff.
Measure path
Calculate volatility, drawdown, and losing clusters.
Measure evidence
Check trade count, regime coverage, and metric stability.
Knowledge check
Check your understanding
Select one answer for each question. Feedback appears immediately.
1. Positive expectancy requires:
Win probability and payoff must combine to exceed losses and costs.
2. Profit factor is gross profit divided by:
It compares total winning dollars with total losing dollars.
3. Why is trade count important?
Metrics based on few observations can move sharply with new trades.
4. 42% of trades win 2.1R, 58% lose 1R, and each trade costs 0.08R. Expectancy per trade is:
0.42 × 2.1 − 0.58 × 1.0 − 0.08 = 0.882 − 0.580 − 0.08 = 0.222R, and the breakeven win rate is (1 + 0.08) ÷ (2.1 + 1) = 34.8%.
Practical rules
- Report multiple complementary metrics.
- Show trade count and tested dates beside every result.
- Use net results after costs.
Common mistakes
- Optimizing only win rate.
- Comparing systems with different exposure using total return alone.
- Treating a small sample estimate as stable.
References
Official reading
The labs are simplified simulations. These sources cover historical performance, data, costs, or execution limitations.
FINRA · Smart Beta—What You Need to Know
Explains that historical backtesting is hypothetical and does not predict future performance.
FINRA Rule 2210 · Communications with the Public
Includes restrictions on performance projections and misleading performance claims.