The 7 Metrics Every Trading Journal Should Track (and the 3 That Don't Matter)

Most "7 metrics" listicles are boilerplate. Here are the seven that actually predict trading performance, plus the three popular ones that are pure vanity.

Every trading blog has a "7 metrics every trader should track" article. They're mostly the same list. Win rate, P&L, profit factor, max drawdown, average win, average loss, total trades. The list isn't wrong. It's just incomplete on one side and bloated on the other.

Here's the version that actually correlates with trading improvement, plus the three popular metrics that are pure vanity and should be ignored.

The 3 metrics that don't matter (much)

1. Win rate (on its own)

Win rate is the most over-tracked number in retail trading and it predicts almost nothing on its own. A 70% win rate trader with -2R losers and +0.5R winners is bleeding capital. A 35% win rate trader with +3R winners and -1R losers is compounding. The number that matters is expectancy, which is win rate combined with R-multiple. Track expectancy. Glance at win rate occasionally to make sure your strategy isn't drifting into territory where the streaks become unmanageable.

2. Total trades count

"You took 247 trades this month." So what. The number of trades you took tells you nothing about whether they were good trades. High trade count usually correlates with overtrading, not with skill. Some of the best discretionary traders take 20 trades a month. Some scalpers take 200 a day. The number isn't predictive of anything diagnostic.

The one place trade count is useful: as a sample size context for your other metrics. Expectancy across 200 trades is more trustworthy than expectancy across 20. But that's trade count as a footnote, not as a primary metric.

3. Longest winning streak

This is the metric most likely to show up on a journal's dashboard and contribute zero to your improvement. A 7-trade winning streak doesn't mean your strategy works. A 7-trade losing streak doesn't mean it's broken. Streaks are statistical artifacts of your win rate and sample size. A trader with a 50% win rate has a 50% chance of seeing a 7-trade streak in any given 100-trade window, just by random distribution. It's noise.

The reason it shows up on dashboards is that it makes traders feel something. Pride on a long winning streak, fear on a long losing streak. Both feelings interfere with execution. Take it off the dashboard.

The 7 that actually matter

1. Expectancy in R, with sample size

The single most important number on your dashboard. Expectancy = (win rate × average win in R) - (loss rate × average loss in R). Tells you how much you make per trade on average, in units that compound across position sizes. Anything under 50 trades is too noisy to trust. Anything over 200 is solid.

If your expectancy is +0.2R or higher across 200+ trades, your strategy is real. If it's negative across the same sample, no amount of position sizing or psychology work will fix it, the strategy itself is the problem.

2. R-multiple distribution (the histogram, not the average)

The shape of your R distribution matters more than the average. Two traders can both have an average of +0.3R per trade but completely different distributions. One has a tight bell curve around 0.5R wins and 0.5R losses. The other has lots of small losses and occasional 4R outliers.

The second trader will survive longer, even with the same expectancy, because outlier wins compound faster and pad through losing streaks. The first trader needs the math to work out perfectly. Look at the histogram.

3. Profit factor

Total winning R divided by total losing R. A profit factor above 1.5 means you're making 1.5 times as much on winners as you're giving back on losers. Above 2.0 is genuinely strong. Below 1.2 is fragile.

This is similar to expectancy but framed differently. The reason to track both is that profit factor is more sensitive to outliers, which catches problems expectancy can hide. A strategy with one massive +20R lottery win and 50 small losses will look fine on expectancy. Profit factor will show you that without that single outlier, the strategy is negative. That's an important thing to know.

4. Max drawdown and recovery time

Max drawdown is the worst peak-to-trough loss in your equity curve. Recovery time is how long it took (in trades or in days) to make it back. Both numbers tell you whether your strategy is psychologically survivable.

A strategy with +50% expected annual return and 35% max drawdown is mathematically fine. Most traders quit during the 35% drawdown. So the drawdown number plus recovery time tells you whether you can actually trade your own strategy. If the answer is no, you need to position-size smaller until the answer becomes yes.

5. Exit efficiency: MFE-capture ratio and MAE-distance

MFE-capture: how much of the maximum favourable excursion did your average winner actually capture? If your winners exit at 40% of MFE, you're leaving more than half the available move on the table. That's a fixable problem, usually a too-close target or a fear-of-giveback exit habit.

MAE-distance: how close did your losers get to your stop before stopping out? If your losers stop at 95% of MAE on average, your stops are positioned exactly where the market wants to flip. Move them further and you'll cut the stop-out rate.

This pair of metrics is what separates traders who improve their execution from traders who plateau. The data is there in your fills, most journals just don't surface it well.

6. Rule-followed rate (and rule-followed PnL vs rule-broken PnL)

For every trade, did you follow your defined rules? Yes, no, or partial. Aggregate over your sample.

This is the one metric that diagnoses whether your execution matches your strategy. A trader with a positive-expectancy strategy but a 60% rule-followed rate is, in practice, trading a different strategy than the one they're testing. The 40% of trades that broke rules are noise that drags the actual strategy's results down.

Look at PnL on rule-followed trades vs rule-broken trades. If rule-followed trades are clearly profitable and rule-broken trades are clearly negative: tighten discipline, no other changes needed. If both are profitable: your rules might be too restrictive and you can loosen them. If both are negative: the strategy is the problem, not the discipline.

7. Regime-segmented PnL

Tag every trade with the market regime that was present (trending, choppy, low-vol, high-vol, news-driven, whatever schema fits your trading). Look at PnL by regime.

Almost every strategy has a regime where it shines and a regime where it bleeds. If you can't filter by regime, you can't see this. So you'll keep trading the strategy in regimes where it doesn't work and blame your psychology when it loses money.

This is where having a fundamentals layer in your journal pays off. TradeSave+ has the Risk Sentiment dashboard , COT, currency strength, news, and calendar all tied to your trade history, so the regime tag goes on every trade automatically rather than something you have to define and maintain by hand.

How to actually use this

Replace whatever vanity metrics are on your dashboard with the seven that matter. The point isn't to memorise the list, it's to set up the dashboard once so you can look at the right numbers in 30 seconds whenever you need to.

Then look at the numbers weekly. Not to feel something, the dashboard isn't there to make you feel motivated. To answer one specific question each week: "is my expectancy positive on the rule-followed trades, and is my exit efficiency improving on the winners?" If both are yes, keep going. If either is no, you've got a specific thing to work on.

The traders who improve are the ones who treat the dashboard as a diagnostic tool. The traders who don't improve treat it as a scoreboard.

Why diagnostic dashboards are hard to build

Most journals show win rate and PnL because they're easy to compute and they make the user feel something on every login. The metrics that actually predict improvement (expectancy with confidence interval, MFE-capture, rule-followed PnL, regime-segmented breakdowns) require structured data, defined setups, defined rules, defined regime tags, and not every journal's data model supports that.

TradeSave+ was built around exactly this list. Custom fields auto-aggregate into stats. The Edges and Leaks analyser ranks every (factor, value) bucket by PnL impact with statistical confidence chips, so the answer to "do my A+ continuations work better in trending or choppy regimes" is a chart, not a manual pivot table.

7-day free trial if you want the dashboard pre-built around the seven metrics that matter.

One final thing

The 7-metrics framing is useful but the real test is whether your journal lets you answer questions you actually have. "Did my A+ continuation setups in trending regimes work better in London session or NY session" is the kind of question that actually moves the needle. If your journal makes that easy, you'll improve. If it doesn't, you'll keep tracking expectancy on a generic dashboard and wondering why the number is moving slowly.