Trading Journal Metrics: The Numbers That Actually Matter

A trading dashboard can contain dozens of statistics and still tell you almost nothing useful.

The question is not how many metrics you can calculate. The question is whether a metric helps you make a better decision.

A useful metric should help identify at least one of the following:

whether the strategy has positive expectancy;
how uncertain that estimate is;
where drawdown comes from;
whether execution matches the tested process;
how management changes the payoff distribution;
whether performance depends on specific environments;
whether recent behavior has drifted from the plan.

No single number defines an edge. Metrics should be read as a system.

1. Trade count

Trade count changes the meaning of every other number.

An expectancy of +0.50R over twelve trades and +0.15R over five hundred trades are not equally informative.

A large sample can still be weak when:

trades come from one regime;
rules changed;
trades are highly correlated;
losing trades are missing;
too many variables were tested.

Trade count is a starting point, not proof.

2. Total R

Total R is the sum of outcomes normalized by initial risk.

It shows the cumulative result, but it grows with sample size. A strategy earning +20R across 500 trades may be weaker per opportunity than one earning +10R across 50 trades.

Read total R with average R.

3. Average R per trade

Average R = Total R ÷ Number of trades

If 200 trades produce +30R:

30R ÷ 200 = +0.15R per trade

Average R is a practical expression of historical expectancy.

R is useful because it normalizes outcomes relative to initial risk. A $500 win means something different when risk was $100 versus $1,000.

4. Median R

The median is the middle result after trades are sorted.

It is less sensitive to extreme winners than the average.

A strategy may have positive average R but negative median R because a few large winners carry the system. That may be legitimate, but it creates a different psychological and risk profile.

Compare:

average R;
median R;
largest winner;
percentage of total profit from top trades.

5. Win rate

Win rate = Winning trades ÷ Total trades

It tells you how often the strategy wins. It does not tell you whether it makes money.

Win rate should be read with:

average winner;
average loser;
expectancy;
losing streak;
payoff distribution.

6. Average winner

Average winner shows the mean size of profitable trades.

A falling average winner may indicate:

early exits;
faster break-even moves;
smaller targets;
lower-volatility conditions;
failure to hold the plan;
a changed strategy mix.

Segment it by setup, environment, management and strategy version.

7. Average loser

Average loser may differ from −1R because of:

partial losses;
early invalidation exits;
slippage;
oversized losses;
stop movement.

A smaller average loser is not automatically better. Tighter stops may reduce average loss while damaging win rate and average winner.

8. Payoff ratio

Payoff ratio = Average winner ÷ Average loss magnitude

If the average winner is +2R and average loser is −1R, the ratio is 2.0.

A larger payoff ratio allows a lower win rate, but it may be distorted by one exceptional winner.

9. Expectancy

A common formula is:

Expectancy = (Win rate × Average win) − (Loss rate × Average loss)

Example:

win rate: 42%;
average winner: +1.9R;
loss rate: 58%;
average loser: −1R.

(0.42 × 1.9R) − (0.58 × 1R) = +0.218R

Expectancy estimates the average historical value per trade.

Calculate gross and net expectancy when costs matter. A low-timeframe strategy may be positive before spread, fees and slippage and negative afterward.

10. Profit factor

Profit factor = Gross winning R ÷ Gross losing R

If winners produce 90R and losers cost 60R, profit factor is 1.5.

It does not reveal trade count, drawdown, concentration of profit or stability.

11. Maximum drawdown

Drawdown measures the decline from an equity peak to a later trough.

It helps evaluate:

how much pain the strategy produced;
whether position sizing was survivable;
whether the system is psychologically executable;
whether current drawdown is unusual.

Maximum drawdown describes one historical path. A different sequence of the same trades can create a different drawdown.

Review depth, duration and time to recovery.

12. Maximum losing streak

The longest consecutive losing sequence helps set realistic expectations.

The historical maximum is not a fixed future limit. Future streaks can be longer.

Use it to inform position sizing, risk limits and psychological expectations.

13. Maximum Favorable Excursion

MFE is the furthest a trade moved in your favor while open.

If a trade reached +2.5R before closing at +0.8R, MFE was +2.5R.

It can help study:

targets;
early exits;
trailing;
break-even;
capture efficiency.

MFE is not “missed profit.” It is known only after the move.

14. Maximum Adverse Excursion

MAE is the furthest a trade moved against you while open.

It can help evaluate:

stop placement;
entry timing;
normal adverse movement;
whether winners require excessive heat.

Do not tighten stops solely because historical winners had small MAE. A tighter stop changes the distribution.

15. Capture percentage

One simple version is:

Realized positive R ÷ MFE

If a trade reached +3R and closed at +1.5R, capture was 50%.

A low percentage may be reasonable for a system pursuing rare large moves. A high percentage may reflect an efficient fixed-target method—or targets that are too close.

Segment by management, setup and environment.

16. Planned versus realized R

The gap may reveal:

early exits;
late entries;
target changes;
slippage;
management inconsistency.

It does not mean every planned target should have been reached. It measures process deviation.

17. Rule-adherence rate

Measure adherence by category:

setup;
entry;
risk;
management;
complete plan.

Then compare expectancy of compliant and non-compliant trades.

When rule-breaking trades outperform, do not immediately abandon the rules. The result may come from small samples, selective exceptions or lucky outliers.