The 7 Trading Metrics That Actually Predict Profitability

Win rate, total P&L, and trade count don't predict whether you stay profitable. The trading metrics that actually predict profitability are expectancy, profit factor, R-multiple distribution, MFE/MAE ratio, plan adherence rate, time-of-day P&L, and drawdown recovery time. Each one tells you something your win rate hides.

Why is win rate misleading?

A 70% win rate can lose money and a 35% win rate can print. Win rate ignores what each trade pays versus what it costs, so it tells you nothing about profitability on its own. Two traders prove it.

Trader A wins far more often and bleeds out. Trader B wins less than half the time and clears $4,000. Trend-following systems run 30-40% win rates and stay profitable. Mean-reversion scalpers run 75-85% and barely break even. Win rate can't tell you which one you are. These seven can.

What are the 7 metrics that predict profitability?

1. Expectancy

Expectancy is the average amount you make per trade given your win rate and average win/loss size. It's the cleanest single read on whether your system works.

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

Win 45% of the time, average winner $250, average loser $120:

(0.45 × $250) − (0.55 × $120) = $112.50 − $66 = $46.50 per trade

Multiply by your trade frequency and you get your expected income. Negative expectancy means trading harder just loses faster. Healthy discretionary traders run 0.2-0.5R. Anything under 0.1R leaves you at the mercy of slippage and commissions.

2. Profit factor

Profit factor is gross profit divided by gross loss. It answers one question: for every dollar your losers cost, how many do your winners earn?

Profit factor = Gross profit / Gross loss

$12,000 in winners against $7,500 in losers gives a profit factor of 1.6. Here's the map:

• Below 1.0 — losing money.
• 1.0 to 1.2 — break-even after costs. Not survivable.
• 1.2 to 1.5 — fragile. One bad week erases a month.
• 1.5 to 2.0 — the realistic target for serious discretionary traders.
• Above 3.0 — usually a small sample or a curve-fit backtest. Be skeptical.

A profit factor that climbs as you tag and drop bad setups is proof you're learning. One that falls means your edge, your discipline, or the regime changed.

3. R-multiple distribution

R expresses each trade as a multiple of the risk you took. Risked 1%, made 2.4%? That's +2.4R. The shape of all your R-values matters more than the average.

Same average, different futures. Trader D's entire edge lives in rare big winners — cut one +8R trade and the year goes flat. Plot a histogram of your R-multiples and check two things: is the right tail fat enough to hold +3R trades, and do losers cluster at −1R or leak into −2R and −3R from violated stops? Short right tail plus outlier losers means you're cutting winners and letting losers run.

4. MFE / MAE ratio

MFE is how far a trade went in your favor before it closed. MAE is how far it went against you. Together they grade your exits, independent of when you actually got out.

Two leaks show up fast. If your winners reach 3.2R of MFE on average but close at 1.1R, you're capturing a third of the move — your target's too tight or you're panic-exiting. If your winners show MAE near −0.9R, you're entering at bad prices and getting bailed out. Fix the entry. Trend-followers run healthy at 0.4-0.6 captured; mean-reversion sits at 0.7-0.9. Below 0.3 and your exits are leaking.

5. Plan adherence rate

Plan adherence is the percentage of trades that matched your stated plan — right setup, right size, right session, rules followed. Across thousands of journals, it tracks profit factor more reliably than any chart pattern.

Plan adherence = (Trades that matched plan) / (Total trades)

A trader who follows a mediocre plan 90% of the time usually beats one with a great plan they follow half the time. Off-plan trades get taken in worse mental states, at worse prices, with worse risk — they bleed out what the on-plan trades earn. Below 70% means the system you think you're running isn't the one you're actually running. Above 85% means if P&L is bad, the plan is the problem, not you. Most journals miss this because it needs one honest tag per trade. It's worth more than every other metric here.

6. Time-of-day P&L

Time-of-day P&L groups your results by session — Asia, London open, the NY overlap, NY afternoon. Most traders earn in two sessions and give it back in a third without knowing which is which.

This trader earned $3,750 in two strong sessions and handed back $1,070 in two weak ones. Stop trading Asia and the NY afternoon and they keep the full $3,750 while trading 40% less. Most "I need more discipline" problems are really "I trade a session where I have no edge."

7. Drawdown recovery time

Drawdown recovery time is the trading days it takes to climb from a drawdown back to your prior equity peak. Drawdown is normal. Your recovery speed reveals how you behave under pressure.

A disciplined trader recovers a 5% drawdown in 5-15 days by trading the same plan, often slightly smaller. An undisciplined one takes 30-60 days or never recovers, because the drawdown triggers oversizing, revenge trades, and strategy-hopping. Recovery under 10 days is excellent. Past 25 days, or drawdowns that compound, means the problem isn't the drawdown — it's your response to it.

A 20-minute weekly review

Sunday, in order: Is expectancy positive and trending up? Profit factor above 1.5? Any new −2R outliers in the histogram? Are winners leaving profit on the table? Adherence above 80%? Any session bleeding money? In a drawdown — how's the recovery curve? Write down one change for next week, based on the worst of the seven. Just one.

Where the AI coach fits

Plan adherence and drawdown recovery are the two metrics most traders skip, because they need honest reads in the moment, not math at week's end. The AI coach inside Trader+AI reads your full history — trades, journal notes, mood, adherence flags — and surfaces both before they cost you. Ask it "what's my adherence on London opens this month?" and it computes the answer from your actual data.

You don't need AI for any of this. A spreadsheet with these seven formulas does the job. But the AI catches the pattern the moment it forms, not three weeks later when the damage is done. Pick one metric this week, track it honestly, and let it tell you the truth your win rate has been hiding.