Win rate is a vanity metric. Use these instead.
Win rate is the first number every trader quotes and the least useful one they own.
Consider two traders. The first wins 80% of the time, making ₹1,000 on winners and losing ₹5,000 on losers. Over 100 trades: 80 × 1,000 − 20 × 5,000 = −₹20,000.
The second wins 35% of the time, making ₹6,000 on winners and losing ₹1,500 on losers. Over 100 trades: 35 × 6,000 − 65 × 1,500 = +₹112,500.
The first trader has the better-sounding statistic and a shrinking account. Win rate tells you how often you're right. It says nothing about whether being right pays you more than being wrong costs you.
Expectancy — what one trade is worth
The foundation. Expressed in R (multiples of your planned risk), it's simply the average R across all your trades.
An expectancy of +0.35R means every trade you take is worth about a third of your risk, on average — winners and losers together. Risk ₹5,000 per trade and each one is worth ₹1,750 in expectation. Your job then becomes taking more of the same trades, not finding better ones.
Negative expectancy means the opposite: every trade you take costs you money on average, and trading more frequently just loses faster.
System Quality Number — is the edge real?
Expectancy alone can mislead. A system with +0.5R expectancy driven entirely by one lucky 12R trade is not the same as one that grinds out +0.5R consistently.
Van Tharp's System Quality Number accounts for this:
SQN = (mean R ÷ standard deviation of R) × √NThree things at once:
- Mean R — is there an edge?
- ÷ standard deviation — is it consistent, or one outlier carrying everything?
- × √N — how much evidence do you have? A great result over 12 trades is weaker evidence than a good result over 200.
| SQN | Reading |
|---|---|
| Below 1.6 | Below average — hard to trade profitably |
| 1.6 – 2.0 | Average |
| 2.0 – 2.5 | Good |
| 2.5 – 3.0 | Excellent |
| Above 3.0 | Superb |
Trader Blueprint shows a warning beside SQN until you have 30+ closed trades. Below that the √N term flatters small samples and the number simply isn't trustworthy yet. Better an honest "too early to tell" than a confident wrong answer.
Payoff ratio
Average win divided by average loss. Paired with win rate it tells you whether the system can work at all: at a 40% win rate you need a payoff above 1.5 just to break even.
Max drawdown and recovery factor
Max drawdown is the largest peak-to-trough fall your equity took. It's the number that decides whether you can actually trade the system — a strategy with a 40% drawdown is mathematically fine and psychologically unbearable for most people.
Trader Blueprint measures it against the equity peak it actually fell from, not against your opening balance, so it doesn't quietly shrink as the account grows.
Recovery factor is net profit ÷ max drawdown. It asks: how much did you make per unit of worst pain? Below 2 means the drawdown consumed a large share of everything you earned.
A word on the Kelly criterion
Kelly computes the theoretically optimal fraction of capital to risk:
K% = W − (1 − W) ÷ payoff ratioTrader Blueprint displays it, at half-Kelly, and then deliberately refuses to call it a recommendation — because the number is frequently absurd.
Run it on a good 45-trade sample and it can suggest risking 17% of your account per trade. Kelly assumes your edge is known exactly and stable forever. On a few dozen trades it is heavily overfit to a sample that includes luck, and following it would ruin a real account during the first normal losing streak.
So the app shows the figure, labels it a theoretical ceiling, and notes that experienced traders cap real risk near 1–2%. A statistic worth seeing is not automatically a statistic worth obeying.
Reading them together
No single number is the answer. The Analytics tab puts six side by side with a plain-English verdict on each, plus a sentence synthesising them.
The pattern worth learning: expectancy tells you if there's an edge, SQN tells you if you can trust it, drawdown tells you if you can survive it. A system needs all three.