Here is a scenario that plays out constantly: a trader finishes the month up 4%. They feel good. They keep trading the same way. Three months later they're down 18% and have no idea what happened.
The problem was visible in the data from month one. Win rate was declining. Average losers were growing. Two instruments they were trading had negative expectancy the entire time. But all they were watching was the P&L number — which, for a while, masked everything.
P&L is a lagging indicator. The metrics in this guide are leading indicators. They tell you what your P&L will do before your P&L does it.
1. Why P&L Lies to You
A trader can have positive P&L and still be on a path to blowing up. This sounds paradoxical, but it happens constantly — and it's one of the most dangerous traps in trading.
The mechanism is straightforward. If you run a strategy with negative expectancy but high variance, you can have good months through luck alone. The expected value per trade is negative, but the distribution is wide enough that random runs of winners are common. You feel like the strategy is working. You may even scale up. Then the distribution catches up with you.
Lucky Streak vs Real Edge
How do you tell the difference? Time and sample size.
A lucky streak tends to look like this: win rate is higher than your historical average, but you can't point to anything you changed. Your average trade size stayed the same. Your setups look the same. The market just happened to cooperate. Alternatively, you had a few large outlier wins that inflated your P&L while your median trade was actually a loser.
Real edge looks different. Win rate is stable across rolling 50-trade windows. Expectancy is consistent. Average win and average loss don't vary wildly month to month. Your best months look like better versions of your normal months, not statistical flukes.
The Minimum Sample Size Problem
This is where most traders make their biggest analytical mistake. They look at 20 or 30 trades and try to conclude something about their strategy.
Twenty trades is not a sample size. It's a coin flip with extra steps. With 20 trades, you can have a 60% win rate by pure chance with a strategy that has 45% true win rate. You can have a 35% win rate with a strategy that's genuinely 55% — and incorrectly conclude it's broken.
The rule of thumb used by professional systematic traders: you need a minimum of 100 trades to begin drawing conclusions, and 200-300 trades to analyze sub-categories reliably (by session, instrument, day of week, etc.).
If you don't have 100 trades yet, your job is not to analyze — it's to execute your strategy consistently and collect the data. Analysis before sufficient data is just storytelling.
What You Actually Need to Measure
P&L answers one question: did I make money? That's the least useful question you can ask your trading data. The useful questions are:
- What is my mathematical edge per trade (expectancy)?
- Am I executing my planned setups consistently (planned vs actual RR)?
- Where is my edge strongest and where am I giving it back (session/instrument breakdown)?
- Is my risk consistent, or do I size up emotionally (consistency score)?
- If I hit a bad streak, how bad can it get (max losing streak)?
Those questions require the 10 metrics below. Let's go through each one.
2. The 10 Metrics That Matter
Win Rate
Wins ÷ Total Trades × 100
Win rate is the metric everyone tracks and the metric that tells you the least in isolation. A 40% win rate with a 2:1 risk:reward is significantly more profitable than a 60% win rate with a 0.8:1 risk:reward. The numbers are straightforward:
- 40% win rate, 2:1 RR: Expected P&L per 10 trades = (4 × 2R) - (6 × 1R) = +2R
- 60% win rate, 0.8:1 RR: Expected P&L per 10 trades = (6 × 0.8R) - (4 × 1R) = +0.8R
The 40% win rate strategy produces 2.5x more profit per trade. Win rate only becomes meaningful in context of what you make when you win versus what you lose when you lose.
Day traders: 50–60%
Swing traders: 40–55%
Average Risk:Reward — Planned vs Actual
The gap between what you plan and what you execute
Actual RR = Actual Profit / Actual Loss per trade
This is the most diagnostic single metric in your toolkit. Most traders calculate their planned RR when they enter a trade. Very few track what their actual RR turned out to be. The gap between these two numbers tells you everything about your execution psychology.
If your planned RR is consistently 2:1 but your actual RR is 1.1:1, you have a specific problem: you are not letting your winners run to target. You're exiting early — either out of fear, or because you're taking partial profits too aggressively, or because you're moving take profits down after entry.
The reverse is also informative. If your actual RR is lower than planned because your losses are bigger than planned, you're letting losers run. You're moving stop losses, averaging down, or not accepting stops when they're hit.
Expectancy
The single most important number in your data
Expectancy is the average amount you expect to make (or lose) per trade, calculated across your full sample. It is the single most important metric in your performance data because it directly answers the question: does my strategy have mathematical edge?
A concrete example: 50% win rate, average win of $200, average loss of $100.
That $50 means that over a large enough sample, each trade you take contributes $50 in expected profit. With 100 trades per month, that's $5,000 in expected monthly profit — regardless of the specific sequence of wins and losses.
The critical implication: if your expectancy is negative, you cannot fix it with position sizing, discipline, or mindset work. You must fix the strategy itself. A negative expectancy strategy run with perfect discipline is just a slower path to losing money.
Profit Factor
Gross Profit ÷ Gross Loss
Profit factor is more robust than win rate because it captures the magnitude of wins and losses, not just their frequency. A profit factor of 1.5 means for every $1 you lose, you make $1.50. At 2.0, you make $2.00 per $1 lost.
What makes profit factor particularly useful is its relationship to win rate. If you know your win rate and profit factor, you can derive your average win/loss ratio. It also degrades gracefully as sample size shrinks — it remains meaningful with 50 trades, where more complex metrics become unreliable.
One specific use case: comparing profit factor across instruments or sessions. If your EURUSD profit factor is 1.8 and your XAUUSD profit factor is 0.9, you now have an objective reason to reduce or eliminate XAUUSD trades.
>1.5: good
>2.0: excellent
>3.0: exceptional (often indicates curve fitting)
1.0–1.2: barely profitable — transaction costs may eliminate your edge in live trading.
Maximum Drawdown
Not just the number — the shape matters
Recovery Factor = Net Profit / Max Drawdown
Most traders know their max drawdown percentage. Fewer understand why the shape of the drawdown is equally important.
A drawdown that goes from $100,000 to $90,000 in a single trade is a different problem than a drawdown that grinds from $100,000 to $90,000 over 60 trades. The first is a position sizing error or a black swan event — identifiable and fixable. The second is a systematic issue: your edge may be eroding, you may be overtrading, or market conditions may have shifted against your strategy.
Recovery factor is your primary benchmark for drawdown quality. It answers: how much profit do you generate per unit of drawdown risk? A recovery factor above 2 means you're generating more than twice your max drawdown in net profit — you're being compensated adequately for the risk you take.
For prop firm traders specifically: track your drawdown relative to your limit, not just as an absolute number. If your FTMO max drawdown limit is 10% and you're at 7.5%, you have effectively 2.5% remaining — at your current volatility, that's potentially only 2-3 bad trades away from a violation.
Average Trade Duration
Are you holding long enough — or too long?
Average trade duration by itself is not predictive. What you want is to correlate trade duration with P&L outcome. Sort your trades into buckets by hold time — 0-30 minutes, 30 minutes to 2 hours, 2-8 hours, overnight — and calculate your profit factor and expectancy within each bucket.
This analysis reveals specific behavioral patterns that are invisible in aggregate stats. Some traders discover that their best trades consistently run for 4-6 hours, but they have a habit of closing everything in under 2 hours. They are cutting their edge short. The data shows it plainly: their trades closed in the 4-6 hour window have a profit factor of 2.4, while trades closed under 2 hours have a profit factor of 0.9.
Duration analysis also enables time-of-day breakdown. If you know your average winning trade lasts 3 hours and you tend to enter between 8-10am London, you know most of your winning trades resolve before 1pm London. Trades entered after 11am London may get caught in the dead zone before the New York open — which might explain why your afternoon trades consistently underperform.
Performance by Session and Time of Day
London, New York, Asian — which one is actually paying you?
Session analysis is one of the highest-leverage performance improvements available to most traders, because the fix — when you find a bad session — is simply to stop trading in it. No strategy change required. No psychological work. Just stop.
When you break down P&L by trading session, you will almost always find significant asymmetry. Most traders have one session where the majority of their edge lives, and one or two sessions where they give a meaningful portion of it back.
A typical example from real trader data: London session profit factor 2.1, New York open profit factor 1.7, New York afternoon profit factor 0.6, Asian session profit factor 0.4. The trader is making 90% of their money in two sessions and spending the rest of the day slowly leaking it in two others. The fix takes five minutes: stop trading in the afternoon NY session and Asian session.
Break this down further by day of week. Friday performance is often notably weaker due to position squaring before the weekend. Monday mornings can be erratic due to gap openings. If your data shows Friday is consistently your worst day, you have an objective reason to size down or skip Fridays entirely.
Performance by Instrument
Which pairs actually make you money?
The same principle as session analysis applies to instruments. Most traders trade 4-6 pairs or instruments. When you break down P&L by instrument, you typically find that 2-3 of them are profitable and 2-3 are not — sometimes significantly not.
The problem is that traders diversify into instruments they don't fully understand. They trade EURUSD profitably because they've traded it for years and internalized its behavior. Then they add XAUUSD because it's popular, or GBPJPY because the volatility looks appealing, or some commodity because they read an article. They haven't studied those instruments the same way, and their performance reflects it.
Gold (XAUUSD) in particular has specific behavioral characteristics — tendency to spike and reverse on news, correlation with DXY and yields, specific session behavior — that differ meaningfully from major FX pairs. A strategy that works cleanly on EURUSD may perform erratically on gold without modification.
Calculate your profit factor and expectancy by instrument. Focus on your top 2-3. Remove the rest until your performance in your core instruments is stable and well-understood. You can always add instruments back later — but traders rarely regret simplifying.
Consecutive Losses (Maximum Losing Streak)
Most traders massively underestimate this number
where N = total number of trades
Most traders dramatically underestimate how long a losing streak they should expect — even with a genuinely profitable strategy. This causes two problems: they break their rules during normal variance, and their position sizing doesn't account for the drawdown that streak will produce.
A worked example: you have a 55% win rate and you've taken 200 trades. The expected maximum losing streak is:
So with 200 trades and a 55% win rate, you should expect to see a losing streak of roughly 7 consecutive losses at some point. With 500 trades, that number grows to around 8-9. This is not a sign your strategy is broken — it's normal statistical variance.
Why this matters for position sizing: if you risk 2% per trade and hit a 7-trade losing streak, you're down 14%. If you're on a prop firm account with a 10% max drawdown, that streak blows your account — even if your strategy has genuine positive expectancy.
The fix: either size down (0.5-1% per trade instead of 2%) or accept that a streak this long signals a potential edge breakdown and reduce size dramatically until performance recovers. Track your actual losing streak against your expected maximum. If you exceed the expected maximum significantly, investigate — market conditions may have changed.
Consistency Score (Return Standard Deviation)
High variance + positive expectancy can still mean gambling
(Higher = more consistent relative to average)
Standard deviation of returns is the closest thing trading has to a Sharpe ratio at the trade level. It measures how variable your results are — not just whether they're positive on average.
Here's why it matters even with positive expectancy: a strategy with positive expectancy but very high variance can still be psychologically and financially dangerous. If your average trade is +$50 but your individual trades range from -$2,000 to +$3,000, you have a volatile strategy that will produce terrifying drawdowns even while being "profitable" on paper. The real-world probability of hitting a drawdown limit before realizing the positive expected value is high.
High consistency — low standard deviation relative to mean return — is the signature of a real edge. It means you're not relying on occasional large wins to offset frequent losses, or vice versa. Each trade contributes a relatively predictable amount. This is what institutional trading desks aim for. It's what passes FTMO challenges. It's what scales.
If your wins are $50, $200, $1,800, $30, $450 and your losses are -$100, -$90, -$80, -$110, -$95, your wins are gambling and your losses are systematic. The losses suggest a disciplined stop, but the wins suggest no consistent take-profit discipline. That standard deviation mismatch is a major flag.
3. How to Read Your Metrics Together
No metric exists in isolation. Each one is a clue. When you read them together, they tell a complete diagnostic story about your trading that P&L alone can never tell.
The goal is not to look at all 10 numbers independently — it's to identify patterns that point to specific fixable problems. Here is the diagnostic matrix experienced traders use:
| Pattern | What It Means | The Fix |
|---|---|---|
| High win rate + low profit factor | Winners are too small. You're cutting profits early while absorbing full losses. | Stop moving take profit targets down. Let trades hit original target. |
| Low win rate + positive expectancy | Strategy works mathematically but feels painful. Each individual loss hurts. High risk of abandonment during normal variance. | Accept lower win rate consciously. Build discipline to hold through losses. Never close a winner early to protect win rate. |
| Positive expectancy + high return variance | Edge exists but inconsistent execution. Emotional trade sizing or inconsistent entry criteria. | Standardize position size. Tighten entry rules. Check if variance tracks with emotion tags. |
| Good overall stats + specific session/pair bleeding | You have a real edge in certain conditions, but you're giving it back in others. | Stop trading in sessions or instruments with profit factor below 1.2 until overall performance stabilizes. |
| Actual RR far below planned RR | Execution failure. Strategy may be sound but you're not running it correctly. | Strict rule: take profit stays where you set it. Reduce trade frequency if needed to maintain discipline. |
| Max losing streak exceeding expectation | Either bad luck (normal) or edge degradation (serious). Need more data to tell the difference. | Reduce size to 25% of normal until 20 trades pass. If results recover, resume. If not, revisit strategy. |
A Real Example Diagnosis
Imagine a trader with these numbers after 150 trades:
- Win rate: 52%
- Planned RR: 2:1
- Actual RR: 1.1:1
- Expectancy: +$18/trade (low)
- Profit factor: 1.18 (borderline)
- London session profit factor: 1.9
- NY afternoon profit factor: 0.7
- EURUSD profit factor: 2.1
- GBPUSD profit factor: 0.8
The diagnosis is clear and actionable, even though overall profitability looks acceptable. The actual RR is barely over half the planned RR — the trader is systematically cutting winners. If they held to target, their expectancy would be close to $50/trade and profit factor would be around 1.7-1.8 instead of 1.18.
Additionally, they should immediately stop trading GBPUSD (profit factor 0.8) and NY afternoon session (profit factor 0.7). These two changes, combined with holding winners to target, would likely double their effective profit factor without changing a single thing about their entry criteria.
That is the power of metric-driven analysis. The strategy didn't need to change. The execution and session selection did.
4. The Problem: Doing This Analysis Manually Takes Hours
If you're running all of the above from a spreadsheet or trading platform export, here's what the actual workflow looks like:
Export your trade history to CSV. Clean the data — broker exports are often messy, inconsistent date formats, lots of irrelevant rows. Build your metrics table. Calculate win rate (easy). Calculate expectancy (moderate — requires average win and loss columns). Build a pivot table for session breakdown — this requires parsing timestamps into session buckets (London: 8am-4pm GMT, NY: 1pm-9pm GMT, etc.), which is non-trivial.
Calculate profit factor by instrument — requires another pivot table filtered by instrument. Calculate max drawdown — requires running a peak-trough calculation across your equity curve, which is a multi-step formula in Excel. Build a drawdown visualization. Calculate standard deviation of returns. Cross-reference actual vs planned RR — which means you need to have been recording your planned RR for each trade at the time of entry (most traders don't).
That workflow, done correctly, takes 3-4 hours for a clean dataset. Most traders don't have a clean dataset. They have screenshots, rough notes, and a broker statement that doesn't include half the information they need.
The result: most traders look at P&L, maybe calculate win rate, and stop there. They never see "I lose money on Fridays." They never see "GBPUSD is destroying my account." They never see "my actual RR is 1.1 when I'm targeting 2.0." These insights exist in the data — they're just too buried to surface without significant analytical work.
Connect your MT4 or MT5 account and every metric in this guide is calculated instantly. No formulas. No spreadsheets. No CSV exports. See your real analytics in minutes →
5. How TSB Pro Gives You All of This Automatically
TSB Pro was built specifically because serious traders need this data and the manual process is too expensive in time and expertise. Every metric in this guide is calculated automatically from your connected MT4 or MT5 trade history.
30+ Analytics Charts
All 10 metrics above — win rate, actual vs planned RR, expectancy, profit factor, max drawdown, trade duration, session performance, instrument performance, losing streak, and consistency score — are calculated and visualized automatically. You don't set up formulas. You connect your account and the data is there.
Performance Breakdowns Built In
Session performance, day-of-week performance, and instrument performance are built-in views, not things you build manually. Filter by any time period. Compare month-over-month. See if your London session performance has improved since you added a rule two months ago. This is the kind of iterative analysis that turns data into decisions.
Drawdown Visualization — Shape, Not Just Number
TSB Pro shows you the drawdown curve, not just the maximum drawdown percentage. You can see whether a drawdown was a single sharp event or a slow grind — which changes your diagnosis completely. The equity curve with drawdown overlay gives you context that a single percentage number cannot.
Emotion Correlation
Tag your trades with emotional states — stressed, confident, hesitant, impulsive, calm. Over time, TSB Pro correlates emotion tags with performance data. You get specific, data-backed insight: trades tagged "stressed" have a profit factor of 0.7, while trades tagged "calm" have a profit factor of 2.1. Now you have an objective reason to implement a pre-trade checklist that filters for mental state, not just setup quality.
AI Coach
The TSB AI Coach analyzes your patterns and surfaces specific, actionable insights. Not generic advice — insights specific to your data. It identifies patterns that are hard to see even when you're doing manual analysis: "Your worst trading day is Friday after 2pm London. Trades taken in that window have negative expectancy. Here's the data." That level of specific, personalized analysis is what separates traders who improve systematically from those who learn by gut feel.
Prop Firm Compliance Tracking
For traders on funded accounts, TSB Pro tracks your daily loss in real time relative to your limit, shows your current drawdown position, and alerts you when you're approaching dangerous thresholds. You know exactly how many bad trades you can absorb before hitting a violation — not after the fact, but in real time while the trading day is still going.
Connect MT4 or MT5, see your real metrics. All 10 analytics from this guide, plus 20+ more, calculated automatically. Start your free trial →
6. Building Your Analysis Habit
Having the data is not enough. You need a systematic process for turning data into decisions. Here is the review cadence that works across trading styles and time frames.
Weekly Review: 20 Minutes, 5 Questions
Do this at the same time every week — Friday after the close or Sunday before the open. Keep it short. The goal is not a comprehensive audit; it's to maintain ongoing awareness of your key metrics.
- What was my expectancy this week? Is it higher or lower than my 4-week rolling average? If lower, why?
- Did I follow my rules? Not whether I was profitable — whether I took setups that matched my criteria, held stops, and managed trades according to plan.
- What was my worst trade and why? Worst by process, not by outcome. Did I take a trade that violated my rules? Did I move a stop? Did I close a winner too early? One specific learning per week compounds.
- What session and instrument performed best? Any shifts from recent weeks? If your London session underperformed this week, is that noise or a pattern emerging?
- What will I do differently next week? One specific change. Not "trade better." Something specific: "I will not open new trades after 3pm London." Small, testable adjustments over time.
Monthly Review: Look at Rolling 3-Month Data
Monthly P&L is too noisy for strategic decisions. A single bad week can make a profitable month look like a bad one. A single big win can make a mediocre month look great. Use 3-month rolling windows for your monthly review.
Check rolling expectancy, profit factor, and drawdown. Compare to the prior 3-month window. Are things improving, stable, or deteriorating? Identify your top-performing and worst-performing sessions and instruments for the period. Check whether your actual RR is trending closer to or further from your planned RR — this is a leading indicator of execution discipline.
Quarterly: Strategic Assessment
Once per quarter, ask the bigger question: is this strategy still working? Markets change. Volatility regimes shift. Correlations break down. A strategy that worked well in a trending environment may underperform in choppy conditions. A strategy that worked in one volatility environment may need adjustment as conditions change.
Look at 6 months of rolling data and ask: if a new trader showed me these stats, would I say this strategy has edge? If the answer is "no" or "I'm not sure," that's your signal to either adapt the strategy or suspend it and move to paper trading while you investigate.
The traders who last in this business are not the ones who never have losing periods. They're the ones who identify performance deterioration early — through metrics, not feelings — and make rational adjustments before the deterioration becomes catastrophic.
Frequently Asked Questions
See Your 10 Metrics, Automatically
Stop doing this in spreadsheets. TSB Pro calculates every metric in this guide from your MT4 or MT5 trade history — automatically, in real time, with 30+ charts and AI-powered insights.
Start Free Trial