Which Metrics Matter in a Trading Journal?
A trading journal can compute dozens of statistics from the same trade history, but a small core set does most of the analytical work: win rate, average win and loss, profit factor, expectancy, maximum drawdown and trade count. Each of these numbers answers one question well and quietly ignores another, which is why reading any of them in isolation is the most common way to misjudge a strategy. This guide defines the core set with a worked example, shows which metrics mislead alone, and explains how to pair them so they generate useful review questions.
Key takeaways
- Six metrics cover most of a journal review: win rate, average win/loss, profit factor, expectancy, max drawdown and trade count.
- No metric is meaningful alone — a 90% win rate loses money if the average loss is far larger than the average win.
- Expectancy combines win rate and payoff into one number: the average result per trade across the whole sample.
- Profit and drawdown belong together — two strategies with the same return can carry very different risk.
- Twenty trades prove little: in a small sample, two extra losses can flip a positive expectancy negative.
- Breakdowns by strategy, instrument, session and cost share turn account-level metrics into specific review questions.
Six numbers do most of the work
Every journal statistic is computed from the same trade history, so most of them overlap. Six form the practical core: win rate, average win and loss, profit factor, expectancy, max drawdown and trade count. Each answers one question cleanly — and each has a blind spot that one of the others covers.
Win rate
The share of trades closed in profit. Misses size completely: it says nothing about how large wins are relative to losses.
Average win / loss
Mean profit of winners versus mean loss of losers — their ratio is the payoff ratio. Misses frequency: a great ratio means little if wins are rare.
Profit factor
Gross profit divided by gross loss; above 1.0 the account made more than it lost. Misses sequence: it ignores how losses cluster.
Expectancy
The average result per trade, combining win rate and payoff. Misses variance: a positive average can hide brutal swings around it.
Max drawdown
The largest peak-to-trough fall in the account. Misses profitability: a shallow drawdown alone does not make a strategy worth trading.
Trade count
The sample size behind every other number. Not a quality measure itself — it decides how much the other five can be trusted.
The core set from a 20-trade sample
The fastest way to see how the metrics relate is to compute all of them from one small history. Take 20 closed trades on a single account:
Computing the core set
- Sample: 20 closed trades — 12 winners, 8 losers.
- Gross profit = $1,440 → average win = 1,440 ÷ 12 = $120.
- Gross loss = $1,200 → average loss = 1,200 ÷ 8 = $150.
- Win rate = 12 ÷ 20 = 60%. Payoff ratio = 120 ÷ 150 = 0.80.
- Profit factor = 1,440 ÷ 1,200 = 1.20.
- Expectancy = (0.60 × 120) − (0.40 × 150) = $12 per trade.
- Max drawdown: a four-loss streak mid-sample cost $560 from the equity peak.
Net result: +$240 over 20 trades. The account is profitable, but every metric describes the same thin edge from a different angle — wins are smaller than losses, and the 60% win rate is what keeps the expectancy positive. You can rerun this arithmetic on your own numbers with the Trading Expectancy Calculator.
What each metric hides on its own
The classic mistake is treating one strong number as a verdict. A 90% win rate sounds excellent — but if the average win is $20 and the average loss is $400, the expectancy is 0.9 × 20 − 0.1 × 400 = −$22 per trade, and the profit factor is 0.45. The mirror-image error is admiring net profit without asking what drawdown was endured to earn it, or whether one outsized trade produced most of it.
| Metric | What it tells you | What it hides |
|---|---|---|
| Win rate | How often trades close in profit | How much is won or lost per trade — meaningless without the payoff ratio |
| Average win / loss | The size of typical wins versus losses | How often each side occurs |
| Profit factor | Profit earned per unit of loss | Loss clustering, drawdown depth and sample size |
| Expectancy | The average per-trade edge in money | Variance around the average; a few outliers can carry it |
| Max drawdown | The worst decline endured so far | Whether the strategy earns anything — and future risk |
| Net profit | The bottom line to date | The risk taken to get it, and how concentrated it is in a few trades |
Read metrics in pairs
The blind spots disappear when metrics are paired. The first essential pair is win rate × payoff ratio: together they decide whether a strategy is viable at all, because each implies a minimum value for the other.
Breakeven win rate = 1 ÷ (1 + payoff ratio)
- payoff ratio
- average win ÷ average loss
- breakeven win rate
- the win rate at which expectancy is exactly zero, before costs
The sample above has a payoff ratio of 0.80, so it needs to win 1 ÷ 1.8 = 55.6% of trades just to break even. Its actual win rate is 60% — a real edge, but a margin of only 4.4 points. The Breakeven Win Rate Calculator maps this trade-off for any payoff ratio.
The second essential pair is return × drawdown. A strategy that made 30% in a year with a 35% max drawdown and one that made 18% with an 8% drawdown are not close — the second earned far more per unit of pain, and would survive position sizes the first could not.
Sample size: 20 trades prove little
Every metric above is an estimate from a sample, and small samples are noisy. In the worked example, just two additional $150 losses would turn the totals into $1,440 gross profit against $1,500 gross loss — expectancy flips to roughly −$2.70 per trade and the profit factor drops below 1.0. Nothing about the strategy changed; the sample was simply too small to pin the numbers down.
Breakdowns that locate the problem
Account-level metrics say what is happening; breakdowns say where. The same core set becomes far more useful when computed per slice of the history:
Per strategy (magic number)
EAs tag their orders with a magic number, so each strategy gets its own metric set. One losing EA can hide inside a profitable account.
Per instrument
Expectancy on EUR/USD can be positive while gold trades quietly drain it. Symbol-level stats show which markets actually pay.
Per session
Splitting by entry hour or session — Asian, London, New York — reveals whether the edge lives in one part of the day.
Cost share
Swap plus commission as a share of gross profit. A rising cost share points at holding time and trade frequency, not the entry logic.
Holding time
Average duration of winners versus losers. Losers held far longer than winners is a classic signature of hesitant exits.
Long vs short
The direction split. Some setups only work one way, and a journal makes that asymmetry visible instead of anecdotal.
Cost share deserves a number: if the sample’s $1,440 gross profit came with $70 of swap and $146 of commission, costs consumed $216 — 15% of the gross. That share is invisible in net profit alone, and it grows with every extra trade and every extra night a position is held.
Act on questions, not autopilot
Metrics describe; they do not instruct. A drop in profit factor is not an order to stop trading, and a strong month is not an order to double size. The productive use of a journal is turning each reading into a specific review question:
- Is expectancy still positive after costs, and over how many trades?
- Which strategy, instrument or session contributes most of the drawdown — and is its sizing appropriate?
- Did one outlier trade produce most of the net profit, or is the edge spread across the sample?
- Are losers held longer than winners — and if so, is that by design or by hesitation?
Answering those questions from your own MetaTrader history — rather than from memory — is the entire case for keeping a journal, covered in more depth in why a trading journal matters. The metrics are the starting point of the review, never the end of it.
Frequently asked
Which single metric is the most important in a trading journal?
None on its own. Expectancy comes closest to a one-number summary because it combines win rate and payoff, but it still says nothing about drawdown, variance or sample size. A useful review always reads at least two metrics together — typically expectancy alongside max drawdown and trade count.
Should I track metrics for the whole account or per strategy?
Per strategy, wherever possible. Account-level numbers blend everything together, so one strong strategy can hide a weak one. Grouping by magic number, symbol or setup tag turns the same metrics into comparisons you can act on — around 100 trades per group is a common working baseline before they start to stabilize.
Can a strategy with a high win rate still lose money?
Yes. Win rate ignores trade size. A system that wins 90% of the time with $20 average wins and $400 average losses has an expectancy of 0.9 × 20 − 0.1 × 400 = −$22 per trade. The payoff ratio has to be read next to the win rate before either number means anything.
What is the difference between profit factor and expectancy?
Profit factor is gross profit divided by gross loss — a ratio, where values above 1.0 mean the account made more than it lost. Expectancy is the average result per trade in money. They usually move together, but expectancy is easier to relate to position sizing and trade frequency.
Related guides
Profit Factor Explained
Gross profit ÷ gross loss — how to read the ratio, why identical values can hide very different strategies, and what it leaves out.
Trading Expectancy Explained
The average result per trade, R-multiples, and why win rate alone cannot tell you whether a strategy made money.
Why a Trading Journal Matters
Why consistent trade records beat memory: patterns, cost drag, rule adherence and a review loop that actually happens.
Related free tools
Free, no login required.
Related NuvoraSync features
Sources & further reading
- MQL5 Documentation — Statistics calculated in the tester — official definitions of profit factor, expected payoff, drawdown and related statistics.
- MetaTrader 5 Help — Strategy Tester report — how MetaTrader presents win rate, profit factor and drawdown in its reports.
Want to analyze your own MetaTrader account data automatically?
NuvoraSync is a read-only MetaTrader journal and analytics workspace. Connect MT4 or MT5 once and your trades, drawdown and performance update on their own — no manual entry, no signals, just your own data.
This article is for educational purposes only. It does not provide trading signals, investment advice, financial recommendations, broker recommendations or trade execution. Calculations are based on user inputs and are estimates only.