Are You Measuring Performance—or Only P&L?

P&L tells you what changed in the account. It does not tell you why.

A positive week can contain poor decisions. A negative week can contain correct execution of a strategy with normal variation. Useful measurement keeps outcomes, risk and behaviour connected without confusing them.

Use three layers of measurement

Outcome metrics

What happened financially: net result, average result, drawdown and costs.

Strategy metrics

How the defined setup behaved: win rate, average win, average loss, expectancy, profit factor and results by condition.

Process metrics

How well the operator followed the plan: eligible setups taken, invalid trades, missed valid trades, size errors and management violations.

Normalise results with R

One R is the planned risk from entry to initial invalidation for that trade.

If the planned loss is one R and the realised result is +1.5R, the gain was one and a half times the planned risk. If the result is −1R, the planned loss was realised. Slippage and costs can make a loss worse than −1R.

R allows trades with different monetary size to be compared, but only when the initial risk is recorded honestly before the outcome.

Know what each metric answers

| Metric | Basic definition | Useful question | Limitation | |---|---|---|---| | Win rate | Winning trades ÷ total classified trades | How frequently did the strategy win? | Says nothing about win and loss size. | | Average win | Total positive results ÷ winning trades | How large was a typical win? | Can be distorted by a few outliers. | | Average loss | Absolute total negative results ÷ losing trades | How large was a typical loss? | Must include costs and execution differences. | | Expectancy | (Win rate × average win) − (loss rate × average loss) | What was the average result per trade in the sample? | A historical estimate, not a promise. | | Profit factor | Gross profit ÷ absolute gross loss | How much gross gain occurred per unit of gross loss? | Unstable in small samples; undefined with no gross loss. | | Maximum drawdown | Largest observed peak-to-trough decline | What decline occurred in this path? | A future drawdown may be larger. | | Longest losing streak | Greatest consecutive-loss count | What sequence occurred in the sample? | Future sequences can differ. | | Rule adherence | Fully compliant trades ÷ reviewed trades | Was the strategy executed as written? | Requires honest, consistent grading. |

Example: win rate is not the whole story

Consider two hypothetical strategies before costs:

• Strategy A wins 70% of trades, averages +0.4R per win and −1R per loss.
• Strategy B wins 40% of trades, averages +2R per win and −1R per loss.

Their estimated sample expectancies are:

• A: (0.70 × 0.4R) − (0.30 × 1R) = −0.02R.
• B: (0.40 × 2R) − (0.60 × 1R) = +0.20R.

The lower-win-rate example has the higher estimated expectancy in this invented sample. Costs, execution and future variation could change both results.

Add excursion and execution metrics

Maximum favourable excursion (MFE)

The greatest unrealised movement in the trade’s favour while it was open.

Maximum adverse excursion (MAE)

The greatest unrealised movement against the position while it was open.

Together, MFE and MAE can help test whether exits, targets or invalidation rules behave as expected. They should be calculated consistently from appropriate data.

Also record:

• Planned versus realised entry.
• Planned versus realised exit.
• Spread, commission, swap or other relevant cost.
• Order rejection, partial fill or execution delay.
• Time in trade.

These fields help distinguish a strategy problem from an execution problem.

Measure opportunity and execution

A journal containing only trades cannot show everything the strategy offered.

Track:

• Eligible opportunity count: Every setup that met the written rules.
• Execution rate: Eligible opportunities taken ÷ eligible opportunities observed.
• Missed valid trades: Eligible setups not executed.
• Invalid-trade rate: Executed trades that failed eligibility.
• Management adherence: Trades managed according to the chosen rule.

A low execution rate can bias the live sample if the trader selectively takes setups after seeing part of the outcome.

Consistency is not forced daily profit

In a performance review, consistency means that the same decision process is applied across comparable opportunities.

It does not mean:

• Every day must be profitable.
• Every trade must have the same result.
• Activity must be created on quiet days.
• Risk should be increased to smooth a target path.

If the current programme has a formally defined consistency condition, measure that exact condition according to the current official rule. Do not replace it with the general meaning used in this lesson.

Avoid the small-sample trap

Five trades can feel important and still say little about a strategy’s long-run behaviour.

Before drawing a conclusion, ask:

• Was the review window chosen before the result?
• Is the sample large enough to contain different market conditions?
• Are all trades from the same strategy version?
• Were invalid trades separated?
• Are costs included?
• Does one outlier dominate the total?
• Is the comparison like-for-like?

Do not keep changing the review window until the preferred conclusion appears.

Use a review hierarchy

After each trade

Capture facts and screenshots. Do not redesign the strategy.

After each session

Reconcile account values, classify process and note operational issues.

Weekly or at the prewritten interval

Review adherence, opportunity capture, costs and condition fit.

At the strategy sample threshold

Evaluate expectancy, drawdown, streaks and potential rule changes. Version and retest any change.

Knowledge check

Continue to Lesson 24: Complete Evaluation Operating Plan

Important educational notice

This lesson is provided for general educational purposes only. Historical and hypothetical performance metrics cannot guarantee future results, evaluation success or payouts. All calculations depend on the completeness and accuracy of the supplied data. Current programme rules and official account information always take precedence.