Trading Metrics Glossary

Guidance last reviewed 2026-08-19

Every measure a trading journal is likely to show you, defined in one place.

Click any term to expand it. Where a measure has more to it than a definition, there is a link through to the full page.

Start here

If you are working through these for the first time, the order that makes them build on each other:

R-Multiples and Position Sizing — the unit everything else reads better in
Trading Performance Metrics Explained — the numbers a journal reports, and what each hides
Trading Expectancy Explained — what one average trade is worth
How Many Trades Before the Numbers Mean Anything — the caveat attached to all of them
Maximum Drawdown Explained — what a method costs to hold
MAE and MFE — Reading Your Exits — the only metrics that judge your exits
System Quality Number (SQN) and What is the K-Ratio? — single-score summaries
Risk-Adjusted Return Ratios Compared — Sharpe and its relatives

Trade-level results

Net P&LThe sum of every trade's result. The number everyone looks at first and the least informative one available — it cannot tell you whether it came from one trade or a hundred.

The sum of every trade's result. The number everyone looks at first and the least informative one available — it cannot tell you whether it came from one trade or a hundred. More

Win rateWinning trades divided by total trades. Says nothing about size: a 90% win rate can lose money and a 35% win rate can be excellent. Only interpretable alongside the payoff ratio.

Winning trades divided by total trades. Says nothing about size: a 90% win rate can lose money and a 35% win rate can be excellent. Only interpretable alongside the payoff ratio. More

Payoff ratio (win/loss ratio)Average win divided by average loss. The other half of the win-rate picture — together they give expectancy.

Average win divided by average loss. The other half of the win-rate picture — together they give expectancy. More

Profit factorGross profit divided by gross loss. Above 1.0 is profitable by definition. ⭐ Remove your best trade and recalculate — if it drops below 1.0, the record describes one good trade rather than a process.

Gross profit divided by gross loss. Above 1.0 is profitable by definition. ⭐ Remove your best trade and recalculate — if it drops below 1.0, the record describes one good trade rather than a process. More

ExpectancyWhat one average trade is worth: (win rate × average win) − (loss rate × average loss). More useful than win rate or payoff ratio alone, because it combines both.

What one average trade is worth: (win rate × average win) − (loss rate × average loss). More useful than win rate or payoff ratio alone, because it combines both. Full page

Average win / average lossThe mean result of each group, kept separate. ⚠️ Compare against the median — when the two diverge sharply, the average is describing a few outliers rather than a typical trade.

The mean result of each group, kept separate. ⚠️ Compare against the median — when the two diverge sharply, the average is describing a few outliers rather than a typical trade. More

Largest win / largest lossThe single best and worst results. Worth knowing mainly as a check: if the largest win is several times the next largest, most summary statistics are being carried by it.

The single best and worst results. Worth knowing mainly as a check: if the largest win is several times the next largest, most summary statistics are being carried by it.

Trade expectancy in RExpectancy measured in units of risk rather than currency. Survives changes in account size, which dollar expectancy does not.

Expectancy measured in units of risk rather than currency. Survives changes in account size, which dollar expectancy does not. More

Risk and position sizing

RThe money put at risk on a trade — the distance from entry to stop, multiplied by size. 🔴 Must be decided before entering; reconstructing it afterwards describes the outcome rather than the risk.

The money put at risk on a trade — the distance from entry to stop, multiplied by size. 🔴 Must be decided before entering; reconstructing it afterwards describes the outcome rather than the risk. Full page

R-multipleA trade's result divided by its R. A trade making twice what it risked is +2R; one stopped as planned is −1R. ⭐ Losses beyond −1R mean the stop moved, which dollar P&L cannot show.

A trade's result divided by its R. A trade making twice what it risked is +2R; one stopped as planned is −1R. ⭐ Losses beyond −1R mean the stop moved, which dollar P&L cannot show. More

Risk per tradeThe share of the account risked on any one position, usually 0.5%–2% by convention. Small because losses compound against you faster than gains compound for you.

The share of the account risked on any one position, usually 0.5%–2% by convention. Small because losses compound against you faster than gains compound for you. More

Fixed fractional sizingRisking a constant percentage of the account on every trade, so position size falls out of stop distance rather than being chosen.

Risking a constant percentage of the account on every trade, so position size falls out of stop distance rather than being chosen. More

SlippageThe difference between the price you wanted and the one you got. Shows up as results worse than −1R and belongs in the record rather than being rounded back to plan.

The difference between the price you wanted and the one you got. Shows up as results worse than −1R and belongs in the record rather than being rounded back to plan.

The account curve

DrawdownHow far the account currently sits below its previous high-water mark. A new peak resets it to zero.

How far the account currently sits below its previous high-water mark. A new peak resets it to zero. Full page

Maximum drawdownThe largest peak-to-trough fall in a record. ⚠️ It is a lower bound on what to plan for, not an upper one — it grows with the length of the record even when nothing changes.

The largest peak-to-trough fall in a record. ⚠️ It is a lower bound on what to plan for, not an upper one — it grows with the length of the record even when nothing changes. More

Time underwaterHow long the account spent below a previous peak. Rarely reported and often what decides whether someone keeps trading a method.

How long the account spent below a previous peak. Rarely reported and often what decides whether someone keeps trading a method. More

High-water markThe highest account value reached so far. Drawdown is measured from it.

The highest account value reached so far. Drawdown is measured from it.

Recovery factorNet profit divided by maximum drawdown. A rough read on how much the account produced relative to the worst stretch it went through.

Net profit divided by maximum drawdown. A rough read on how much the account produced relative to the worst stretch it went through.

Exits

MAE — Maximum Adverse ExcursionThe furthest a trade moved against you while open. Comparing MAE across winners and losers is how you find out whether a stop is wider than it needs to be.

The furthest a trade moved against you while open. Comparing MAE across winners and losers is how you find out whether a stop is wider than it needs to be. Full page

MFE — Maximum Favorable ExcursionThe furthest a trade moved in your favor while open. Large MFE with small results points at an exit rule rather than at discipline.

The furthest a trade moved in your favor while open. Large MFE with small results points at an exit rule rather than at discipline. More

Exit efficiencyResult divided by MFE — what share of the available move you kept. ⚠️ 100% is not the target; it would mean every exit was the high tick.

Result divided by MFE — what share of the available move you kept. ⚠️ 100% is not the target; it would mean every exit was the high tick. More

Single-score summaries

SQN — System Quality NumberVan Tharp's score combining edge, consistency and sample size: (average R ÷ standard deviation of R) × √N. 🔴 N is capped at 100 in his definition, which many descriptions leave out.

Van Tharp's score combining edge, consistency and sample size: (average R ÷ standard deviation of R) × √N. 🔴 N is capped at 100 in his definition, which many descriptions leave out. Full page

K-ratioMeasures whether an equity curve rose steadily or in lucky lumps, by regressing the curve and comparing its slope to the scatter around it. ⚠️ Three versions exist and they scale differently.

Measures whether an equity curve rose steadily or in lucky lumps, by regressing the curve and comparing its slope to the scatter around it. ⚠️ Three versions exist and they scale differently. Full page

Sharpe ratioExcess return divided by the standard deviation of returns. Universally understood, and it penalizes large winning periods exactly as it penalizes losses.

Excess return divided by the standard deviation of returns. Universally understood, and it penalizes large winning periods exactly as it penalizes losses. More

Sortino ratioLike Sharpe but counting only downside deviation, so upside volatility stops being a penalty. ⚠️ Noisier, because it discards most of the sample.

Like Sharpe but counting only downside deviation, so upside volatility stops being a penalty. ⚠️ Noisier, because it discards most of the sample. More

Calmar and MAR ratiosAnnualized return divided by maximum drawdown — Calmar over 36 months, MAR since inception. The closest of these to what a person actually experiences.

Annualized return divided by maximum drawdown — Calmar over 36 months, MAR since inception. The closest of these to what a person actually experiences. More

Sterling ratioReturn over the average of several large drawdowns rather than the single worst. ⚠️ No agreed definition, so figures from different platforms are not comparable.

Return over the average of several large drawdowns rather than the single worst. ⚠️ No agreed definition, so figures from different platforms are not comparable. More

Ulcer IndexA drawdown measure that counts duration as well as depth — the only common one that does.

A drawdown measure that counts duration as well as depth — the only common one that does. More

Omega ratioUses the whole return distribution rather than summarizing it, so skew and fat tails are captured. ⚠️ Requires a threshold, and is uninterpretable without it.

Uses the whole return distribution rather than summarizing it, so skew and fat tails are captured. ⚠️ Requires a threshold, and is uninterpretable without it. More

Treynor ratioExcess return divided by beta. Built for judging one holding inside a diversified portfolio, and usually the wrong tool for a single trading account.

Excess return divided by beta. Built for judging one holding inside a diversified portfolio, and usually the wrong tool for a single trading account. More

Statistical terms

Sample sizeThe number of trades every other statistic is estimated from. ⭐ The one that governs all the rest — twenty trades produce confident-looking numbers that are not measurements.

The number of trades every other statistic is estimated from. ⭐ The one that governs all the rest — twenty trades produce confident-looking numbers that are not measurements. Full page

Standard deviationHow widely results scatter around their average. Appears in the denominator of Sharpe, SQN and most consistency measures.

How widely results scatter around their average. Appears in the denominator of Sharpe, SQN and most consistency measures.

Downside deviationStandard deviation counting only results below a target. The denominator of the Sortino ratio.

Standard deviation counting only results below a target. The denominator of the Sortino ratio.

SkewWhether results lean to one side. Positive skew — many small losses and rare large wins — is normal for trend following and is penalized by Sharpe and SQN alike.

Whether results lean to one side. Positive skew — many small losses and rare large wins — is normal for trend following and is penalized by Sharpe and SQN alike.

Fat tailsExtreme results occurring more often than a normal distribution would predict. Common in trading returns, and the reason several of these measures were invented.

Extreme results occurring more often than a normal distribution would predict. Common in trading returns, and the reason several of these measures were invented.

Confidence intervalThe plausible range around an estimate. A 60% win rate over 50 trades is consistent with anywhere from about 46% to 73% — a range wide enough to describe two different businesses.

The plausible range around an estimate. A 60% win rate over 50 trades is consistent with anywhere from about 46% to 73% — a range wide enough to describe two different businesses. More

Selection effectTesting several ideas and keeping the best-looking one. ⚠️ Over small samples the winner is usually the luckiest rather than the best.

Testing several ideas and keeping the best-looking one. ⚠️ Over small samples the winner is usually the luckiest rather than the best. More

ReshufflingRandomizing the order of your trades and redrawing the equity curve, repeatedly, to see what sequence alone can produce from a fixed set of results.

Randomizing the order of your trades and redrawing the equity curve, repeatedly, to see what sequence alone can produce from a fixed set of results. More

Market timing

Market holidays and early closesThe days US stock markets are closed or shut early, the rule behind each, and why futures and bond hours differ.

The days US stock markets are closed or shut early, the rule behind each, and why futures and bond hours differ. Full page

Expirations, rolls and key datesMonthly and quarterly expiration, triple witching, VIX expiry, futures rolls, index rebalances and period ends — all calculable from published rules.

Monthly and quarterly expiration, triple witching, VIX expiry, futures rolls, index rebalances and period ends — all calculable from published rules. Full page

The London–New York overlapThe busiest stretch of the trading day, and the weeks each year when it shifts because US and European clocks disagree.

The busiest stretch of the trading day, and the weeks each year when it shifts because US and European clocks disagree. Full page

Where a threshold is described as a rule of thumb it is a convention rather than a finding, and platforms differ in how they calculate several of these figures.

Chart Ding is a market clock with customizable alarms for traders — candle-close alerts, world market sessions, and holiday warnings. Open the app.

Nothing on this website is trading or investment advice. Chart Ding is a market clock and alarm tool — it does not recommend trades, evaluate strategies, or take account of anyone's circumstances. Nothing here should be construed as a recommendation or relied on as the basis for a trading decision. Consult a licensed professional before placing real money at risk. Trading involves risk of loss.