Trading Performance Metrics Beyond Win Rate

Win rate measures the percentage of closed trades that made money. It does not show how large winners are relative to losers, what one trade is worth on average, how deep the drawdowns become or whether the trader followed the process that produced the sample.

A useful trading scorecard needs three layers: edge, risk and behavior. Start with six metrics rather than collecting every number your platform can display.

Why win rate is incomplete

Two systems can both win 60% of their trades and have opposite results. If the first earns an average of 1.5R on winners and loses 1R on losers, its expectancy is positive. If the second earns 0.4R and loses 1R, its expectancy is negative despite the same accuracy.

Win rate becomes useful when read with average win, average loss and the distribution of results. On its own, it can reward strategies that collect many small wins while hiding occasional large losses.

1. Expectancy: the average value of a trade

Expectancy combines win frequency with win and loss size:

Expectancy = (win rate x average win) - (loss rate x average loss)

Use absolute loss size in the formula. Expressing results in R, where 1R is the planned risk for a trade, makes samples with different position sizes easier to compare. Positive historical expectancy describes the observed sample; it does not guarantee the next trade or prove the edge will persist.

2. Payoff ratio: how much winners earn relative to losers

Payoff ratio is average winning trade divided by the absolute average losing trade. It explains how much pressure the strategy puts on win rate. A lower-frequency strategy may remain viable with a larger payoff, while a high-frequency winner may require tight control of occasional losses.

Read payoff with the largest losses and the result distribution. One extreme loser can be hidden inside an otherwise reasonable average.

3. Profit factor: gross profit relative to gross loss

Profit factor equals gross profit divided by the absolute value of gross loss. Above 1 means the observed sample produced more gross profit than gross loss; below 1 means the opposite. It is compact, but it does not show when losses occurred or how painful the path was.

Do not turn generic internet thresholds into universal pass marks. Compare like-for-like periods, include fees and inspect how dependent the ratio is on a few outlier winners.

4. Maximum drawdown: the path, not only the destination

Maximum drawdown is the largest decline from an equity peak to the following trough during the selected period. It makes two samples with similar net results easier to distinguish: one may have reached the result smoothly, while the other required surviving a decline the trader or account could not realistically tolerate.

Report drawdown in the unit that controls your decisions, such as currency, percentage or R. Also inspect duration and recovery; the same depth can create a very different operating problem if it lasts three sessions instead of three months.

5. Review coverage: how much of the sample is actually known

Review coverage equals completed trade reviews divided by total trades. It is a data-quality metric. If only selected winners or the most memorable losses were reviewed, behavioral conclusions inherit that selection bias.

Show the count next to the percentage: 44 of 50 trades communicates more than 88% alone. The daily trading journal routine is designed to close this gap before the weekly analysis begins.

6. Plan adherence: whether the strategy was actually executed

Plan adherence equals followed-plan trades divided by followed plus violated trades. It helps separate two different diagnoses: a plan may have weak results even when followed, or a potentially useful plan may be obscured by inconsistent execution.

Keep unknown reviews visible and exclude not-applicable decisions from rule-level denominators. Calculate your current sample with the free trading plan adherence calculator, then use the full adherence guide to investigate individual rules.

How to read the metrics together

• Positive expectancy, weak adherence: the observed result may be difficult to attribute to the written plan.
• Negative expectancy, strong adherence: execution may be consistent while the setup, costs or management logic need investigation.
• Good profit factor, severe drawdown: aggregate edge may hide a path that violates account or personal risk limits.
• Strong adherence, low coverage: the rate may reflect a selectively reviewed subset rather than the whole period.
• Stable P&L, rising violations: profitable rule breaks may be masking process drift.

TDLab's Analytics keeps performance, risk and discipline breakdowns available under the same account and date filters. The Discipline Score combines review coverage, plan adherence, execution quality, mistake control, journal consistency and active-rule adherence into a broader behavioral view.

Segment before changing the strategy

Account-wide averages can hide the condition that creates the result. Break the scorecard down by setup, symbol, session, time block, weekday, trade number and previous trade result. Change one dimension at a time and keep the metric definition stable between periods.

When a behavioral segment looks weak, rank the repeated issues with the trading mistakes by cost workflow before choosing one corrective rule.

A practical weekly metric review

1. Fix the account scope, date range and trade count.
2. Confirm import completeness and review coverage.
3. Read expectancy, payoff and profit factor together.
4. Inspect maximum drawdown, large losses and recovery.
5. Compare followed-plan and violated-plan trades.
6. Open the segment behind the largest meaningful change.
7. Choose one question or rule for the next period.

The weekly trading review checklist provides the full sequence for turning the scorecard into one next action.

Common questions

What is the most important trading performance metric?

No single metric answers every decision. Expectancy is a strong edge summary, but it still needs sample size, drawdown and process context. Use the smallest group of metrics that can distinguish edge, risk and execution.

How many trades are needed before metrics are useful?

There is no universal count because trade frequency and result distributions differ. Always show the sample, avoid strong conclusions from a few trades and compare stable definitions across increasing windows.

Should trading metrics be measured in money or R?

Use both when possible. Money shows account impact; R normalizes the result by planned risk and makes trades with different size easier to compare. Neither replaces checking whether planned risk was recorded consistently.