Skip to main content
Trading Edges

Quantifying Your Edge

Pomegra Learn

What Numbers Actually Prove You Have an Edge?

You've tested a trading idea. You made 100 trades. You won 56 of them. Is that an edge? Maybe. It could also be luck. You need to measure not just how often you win, but by how much you win versus lose, how consistent the results are, and whether the results are statistically significant. Quantifying an edge means converting a trading idea into hard numbers: win rate, profit factor, expectancy, and Sharpe ratio. These metrics answer the question: "Over time, does this strategy make money?" Without these numbers, you're trading on faith. With them, you're trading with evidence.

Quick definition: Quantifying an edge is the process of calculating key performance metrics (win rate, profit factor, expectancy, Sharpe ratio) that prove a strategy has positive expected value and that the results are not due to luck.

Key takeaways​

  • Win rate alone is misleading. A 40% win rate strategy can outperform a 60% win rate strategy if average wins are large and average losses are small.
  • Profit factor (gross profit / gross loss) ≥ 1.5 is a reasonable threshold for a tradable edge; <1.2 is marginal.
  • Expectancy (average profit per trade) is the foundation: Expectancy = (Win % × Avg Win) − (Loss % × Avg Loss). Positive expectancy is required for an edge.
  • Sharpe ratio (≥0.8–1.0) measures consistency and risk-adjusted returns. A 2% edge with high volatility is less valuable than a 1% edge with low volatility.
  • Statistical significance requires at least 30 trades; at least 100 is better. Fewer than 30 trades and your edge could be luck.
  • Run forward tests and monitor metrics quarterly to confirm the edge persists in live trading.

The key metrics​

Win rate​

Win rate is the percentage of profitable trades. If you make 100 trades and 56 are profitable, your win rate is 56%.

Why it's misleading: A 40% win rate strategy can be profitable if average wins are large and average losses are small. A 65% win rate strategy can be unprofitable if average wins are small and average losses are large. Win rate alone tells you nothing about expectancy.

Baseline: A 50% win rate is breakeven before costs. To overcome spread, slippage, and commission, you need at least 51–52% win rate if average win and loss are equal. If your average win is larger than your average loss, a 45% win rate might be sufficient.

Average win and average loss​

This is critical. Calculate the arithmetic mean of all winning trades and all losing trades. Then compare them.

Example:

  • 100 trades total
  • 56 wins: $300 + $280 + $250 + $200 + ... (average: $240)
  • 44 losses: -$150 - $140 - $130 - $120 - ... (average: -$110)

Your average win is $240. Your average loss is $110. The ratio is 2.18:1. This is a healthy edge indicator.

Baseline: If average win and loss are roughly equal, you need a 55%+ win rate to be profitable. If average win is 2x average loss, you need 40%+ win rate. Aim for an average win / average loss ratio of at least 1.5:1, ideally 2:1.

Expectancy​

Expectancy (or expected value) is the average profit or loss per trade, accounting for both win rate and size of wins/losses.

Formula:

Expectancy = (Win% × AvgWin) − (Loss% × AvgLoss)

Example:

Expectancy = (0.56 × $240) − (0.44 × $110)
          = $134.4 − $48.4
          = $86 per trade

If you make 100 trades per year, your expected profit is $8,600 (before transaction costs).

Interpreting expectancy:

  • Positive expectancy = edge exists.
  • Negative expectancy = you're gambling; stop.
  • Small positive expectancy (<0.5% of trade size) = edge is marginal; might not survive costs.
  • Large positive expectancy (>1–2% of trade size) = strong edge.

Test expectancy against transaction costs. If your expectancy is $86 per trade and you're trading a $10,000 position, that's 0.86% per trade. If round-trip transaction costs are 1%, you're breakeven or underwater. Your edge must exceed your costs.

Profit factor​

Profit factor is the ratio of gross profit to gross loss.

Formula:

Profit Factor = Gross Profit / Gross Loss

Example:

Gross Profit (all winning trades): $13,440
Gross Loss (all losing trades): -$4,840
Profit Factor = $13,440 / $4,840 = 2.77

Interpreting profit factor:

  • >2.0 = excellent edge.
  • 1.5–2.0 = good edge.
  • 1.2–1.5 = marginal, worth trading if low drawdown.
  • <1.2 = barely profitable; risky.
  • <1.0 = losing strategy.

Profit factor of 1.5 means for every dollar lost, you make $1.50. That's reasonable after considering slippage and costs will eat into it.

Sharpe ratio​

Sharpe ratio measures risk-adjusted returns. It's the ratio of average return to volatility of returns.

Formula (approximate):

Sharpe Ratio = (Average Return − Risk-Free Rate) / Std Dev of Returns

For trading, use:

Sharpe Ratio = (Average Daily/Monthly Return) / (Std Dev of Daily/Monthly Returns) × √(periods per year)

If your average monthly return is 1% and standard deviation of monthly returns is 2%, your Sharpe is roughly (1% / 2%) = 0.5.

Interpreting Sharpe ratio:

  • >1.5 = excellent consistency; returns exceed volatility significantly.
  • 1.0–1.5 = good consistency.
  • 0.5–1.0 = acceptable but volatile.
  • <0.5 = very volatile; unlikely to persist.

Why Sharpe matters: Two strategies might have the same expectancy but different Sharpe ratios. A Sharpe of 1.5 is more valuable than a Sharpe of 0.5 because returns are more consistent and drawdowns are smaller.

Maximum drawdown and drawdown duration​

Maximum drawdown is the largest peak-to-trough decline from your highest cumulative profit.

Example: Your strategy peaks at +$50,000. Then a losing streak occurs, and equity drops to +$20,000. Maximum drawdown: $30,000, or 60% of peak.

Interpreting MDD:

  • <15% = excellent edge with low risk.
  • 15–30% = acceptable.
  • 30–50% = risky but possible if expectancy is high.
  • >50% = dangerous; stop trading.

Drawdown duration: How long does it take to recover to new equity highs? A $30,000 drawdown that recovers in 2 weeks is manageable. The same drawdown taking 6 months is psychologically brutal and suggests high volatility.

Number of trades and statistical significance​

A 60% win rate on 10 trades is meaningless. The same win rate on 500 trades is evidence of an edge.

Minimum trades: 30 for basic validity, 100 for confidence, 500+ for robust evidence.

Statistical significance test (simplified):

For a strategy with win rate p, the standard error is:

SE = sqrt(p × (1 − p) / n)

Your result is statistically significant if the observed win rate differs from 50% by at least 1.96 × SE (95% confidence).

Example: 100 trades, 56 wins (56% win rate).

SE = sqrt(0.5 × 0.5 / 100) = sqrt(0.0025) = 0.05
Difference from 50%: 0.06
1.96 × SE = 0.098
Is 0.06 > 0.098? No.
Result is NOT statistically significant.

Example: 500 trades, 275 wins (55% win rate).

SE = sqrt(0.5 × 0.5 / 500) = sqrt(0.0005) = 0.022
Difference from 50%: 0.05
1.96 × SE = 0.043
Is 0.05 > 0.043? Yes.
Result IS statistically significant.

This is why you need 100+ trades to have confidence in an edge.

Decision tree​

Real-world examples​

Low win rate, high expectancy strategy.

  • 80 trades, 32 wins (40% win rate)
  • Average win: $600
  • Average loss: $200
  • Expectancy: (0.40 × $600) − (0.60 × $200) = $240 − $120 = $120 per trade
  • Profit Factor: (32 × $600) / (48 × $200) = $19,200 / $9,600 = 2.0
  • This is profitable. Win rate is low but average win is 3x average loss. Sharpe depends on consistency, but the metrics are solid.

High win rate, low expectancy strategy.

  • 100 trades, 68 wins (68% win rate)
  • Average win: $150
  • Average loss: $400
  • Expectancy: (0.68 × $150) − (0.32 × $400) = $102 − $128 = −$26 per trade
  • This is a losing strategy despite the 68% win rate. Average losses are larger than average wins. Don't trade it.

Medium metrics, good Sharpe strategy.

  • 200 trades, 106 wins (53% win rate)
  • Average win: $300
  • Average loss: $250
  • Expectancy: (0.53 × $300) − (0.47 × $250) = $159 − $117.5 = $41.5 per trade
  • Profit Factor: (106 × $300) / (94 × $250) = $31,800 / $23,500 = 1.35
  • Monthly return: 2% (assuming ~$8,300 profit monthly on $400k capital)
  • Monthly volatility: 1.2% (based on rolling monthly returns)
  • Sharpe: (2% / 1.2%) × sqrt(12) = 1.67 × 3.46 ≈ 5.8 (annualized)
  • Strong edge. Consistent returns, low volatility, positive expectancy. This is tradable.

How to calculate and track these metrics​

In a spreadsheet:

DateEntryExitP/LWin?CumulativeMax Equity
1/2100105+5Yes+5+5
1/3105103-2No+3+5
1/4103108+5Yes+8+8
.....................

Then calculate:

  • Total Wins: Count "Yes" entries
  • Total Trades: Count all rows
  • Win Rate: Wins / Total
  • Avg Win: Average of positive P/L
  • Avg Loss: Average of negative P/L
  • Expectancy: (Win Rate × Avg Win) − (Loss Rate × Avg Loss)
  • Profit Factor: (Wins × Avg Win) / (Losses × Avg Loss)
  • Max Drawdown: Lowest (Cumulative − Max Equity to date)

Using backtesting software: Most platforms (TradeStation, NinjaTrader, Python libraries like Backtrader) calculate these automatically. Verify the numbers make sense; some platforms calculate metrics differently.

Common mistakes in quantifying edges​

Only looking at win rate. You're tempted by a strategy with 70% win rate but don't check expectancy. After calculating, average win is $100 and average loss is $800. It's a losing strategy. Check all metrics.

Using in-sample data to calculate metrics. The metrics are inflated because you optimized the strategy on that data. Always calculate metrics on out-of-sample or forward-test data.

Not accounting for compounding and position sizing. Your backtest assumes fixed position size on every trade. If you size up as equity grows, returns might be different. Use consistent position sizing in calculations.

Ignoring transaction costs in expectancy. Your backtest expectancy is $50 per trade. Your actual costs are $40 per trade. Real expectancy is $10 per trade, not $50. Always subtract costs.

Confusing profit factor with win rate. A profit factor of 1.5 is reasonable. A win rate of 1.5% is terrible. Keep the metrics separate.

Using too short a sample to calculate Sharpe ratio. Sharpe ratio calculated on 10 trades is noise. Use 50+ trades to get a meaningful estimate.

FAQ​

What's the minimum win rate I need for an edge?​

It depends on the ratio of average win to average loss. If they're equal, you need 51–52% win rate (to overcome costs). If average win is 2x average loss, you need 33%+ win rate. If average win is 0.5x average loss, you need 67%+ win rate. Calculate expectancy to know for sure.

Is a profit factor of 1.3 tradable?​

Marginally. After accounting for slippage and commissions, you're left with very little edge. If Sharpe is high (consistent results), maybe. If Sharpe is low (volatile), it's risky. Aim for 1.5+.

How many trades should I accumulate before I trust my metrics?​

At least 100; ideally 300+. At 100 trades, you have reasonable confidence. At 300+, you have strong confidence. Below 100, results could be luck.

What's a good Sharpe ratio for a trading strategy?​

0.8+ is acceptable, 1.0+ is good, 1.5+ is excellent. A Sharpe of 2.0+ is rare and often indicates overfitting or a very niche inefficiency.

Should I aim for high win rate or high profit factor?​

High profit factor matters more. A 40% win rate with a 2.0 profit factor (average win is 3x average loss) beats a 65% win rate with a 1.1 profit factor. Profit factor reflects true edge strength.

Can expectancy and Sharpe ratio both be high?​

Yes, and that's ideal. High expectancy means good average return per trade. High Sharpe means those returns are consistent with low volatility. Both indicate a robust, tradable edge.

My backtest has great metrics but live trading is worse. What went wrong?​

Slippage, commissions, or overfitting. Backtest metrics often assume perfect fills. Real trading has wider spreads and larger slippage. Your backtest might be overfitted to historical data. Compare metrics on out-of-sample data to backtest data.

Summary​

Quantifying an edge means calculating five key metrics: win rate, average win/loss, expectancy, profit factor, and Sharpe ratio. Win rate alone is misleading; a 40% win rate can outperform 65% win rate if average wins are larger. Expectancy (average profit per trade) is the foundation: it must exceed your transaction costs to be tradable. Profit factor ≥1.5 is a reasonable threshold; <1.2 is marginal. Sharpe ratio ≥0.8 indicates consistent returns. You need at least 100 trades to calculate metrics with confidence; fewer than 30 and results could be luck. Calculate metrics on out-of-sample data, not the data used to discover the strategy. Track metrics quarterly to monitor edge persistence. These numbers are your evidence that an edge is real, not hope or gut feel.

Next​

Pure Price Action Edge