A Practical Metrics Guide: Sharpe, Drawdown, Win Rate—What to Trust

Your backtest shows:

• Total return: 80%

• Sharpe ratio: 3.5

• Win rate: 75%

Your reaction: "This is amazing!"

Reality: These numbers might be completely misleading.

Why? Because metrics can lie—or at minimum, tell incomplete stories.

In this guide, we'll break down the most common performance metrics:

• What each one actually measures

• What it reveals (and what it hides)

• Which ones to trust

• How to spot red flags

The metrics that matter

1. Total return

What it measures: How much your account grew (or shrank) during the test period.

Formula:

`` Total Return = (Ending Balance - Starting Balance) / Starting Balance × 100% `

Example:

• Starting balance: $10,000

• Ending balance: $15,000

• Total return: ($15,000 - $10,000) / $10,000 = 50%

What it reveals:

• Whether the strategy made money (positive return) or lost money (negative return)

What it hides:

• Time period: 50% in 1 year is great. 50% in 10 years is terrible.

• Risk: Did you make 50% with smooth, steady gains? Or wild swings?

• Survivability: Did you ever drop -40% before recovering to +50%?

Red flags:

• ❌ Returns too good to be true (100%+ annual on a low-risk strategy)

• ❌ Returns only positive in bull markets (test it on 2022 data)

What to trust:

• ✅ Annualized return (adjusts for time period)

• ✅ Risk-adjusted return (Sharpe ratio, covered below)

Bottom line: Total return alone means almost nothing. Always pair it with risk metrics.

2. Sharpe ratio

What it measures: Return per unit of risk (volatility).

Formula:

Sharpe Ratio = (Strategy Return - Risk-Free Rate) / Standard Deviation of Returns

Simplified:

Sharpe = Avg Return / Volatility

Example:

• Avg annual return: 20%

• Risk-free rate: 4% (Treasury bonds)

• Standard deviation (volatility): 10%

• Sharpe ratio: (20% - 4%) / 10% = 1.6

What it reveals:

• How much return you're getting per unit of risk

• Whether the strategy is efficient (high return with low volatility = good)

What it hides:

• Tail risk: Sharpe assumes normal distribution. Black swan events (crashes) aren't captured.

• Drawdown severity: A strategy can have great Sharpe but still drop -30% in one month.

• Win distribution: 10 small wins + 1 massive loss can still show decent Sharpe.

Red flags:

• ❌ Sharpe > 3.0 on a backtest (probably curve-fit)

• ❌ Sharpe varies wildly across different test periods (unstable strategy)

What to trust:

• ✅ Out-of-sample Sharpe (from walk-forward testing)

• ✅ Live trading Sharpe (after 6-12 months)

Bottom line: Sharpe ratio is one of the best metrics, but don't use it alone. Pair with max drawdown.

3. Max drawdown

What it measures: The largest peak-to-trough decline in your account.

Formula:

Max Drawdown = (Peak Balance - Trough Balance) / Peak Balance × 100%

Example:

• Account peaks at $15,000

• Drops to $12,000

• Max drawdown: ($15,000 - $12,000) / $15,000 = 20%

What it reveals:

• The worst-case scenario you experienced during the test

• How much pain you'd endure before the strategy recovers

What it hides:

• Duration: Did the drawdown last 1 week or 6 months?

• Frequency: Was this the only drawdown, or did you hit -15% five times?

• Future drawdowns: The worst drawdown is usually still ahead of you.

Red flags:

• ❌ Max drawdown > 30% (you'll quit before the strategy recovers)

• ❌ Multiple drawdowns > 20% (strategy is too volatile)

• ❌ Recovery time > 6 months (psychologically unbearable)

What to trust:

• ✅ Max drawdown + recovery time (how long did it take to get back to break-even?)

• ✅ Drawdown-to-return ratio (max DD should be < 50% of total return)

Example:

• Total return: 40%

• Max drawdown: 25%

Ratio: 25% / 40% = 62.5%

Interpretation: You risked 62.5% of your gains to achieve them. That's high. Ideally, this ratio should be < 50%.

Bottom line: Max drawdown tells you how much pain you'll endure. If you can't handle it, the strategy won't work for you—even if it's profitable.

4. Win rate

What it measures: Percentage of trades that were profitable.

Formula:

Win Rate = (Number of Winning Trades / Total Trades) × 100%

Example:

• Total trades: 100

• Winning trades: 60

• Win rate: 60 / 100 = 60%

What it reveals:

• How often you're "right"

What it hides:

• Profit size: Winning 90% of the time means nothing if your losses are huge.

• Strategy type: Mean-reversion strategies often have high win rates (70-80%). Trend-following strategies often have low win rates (30-40%).

The win rate trap:

Most beginners chase high win rates. This is a mistake.

Example:

Strategy A (high win rate, low reward/risk):

• Win rate: 80%

• Avg win: $50

• Avg loss: $200

• Expected value: (0.8 × $50) + (0.2 × -$200) = $40 - $40 = $0

Strategy B (low win rate, high reward/risk):

• Win rate: 40%

• Avg win: $300

• Avg loss: $100

• Expected value: (0.4 × $300) + (0.6 × -$100) = $120 - $60 = $60

Strategy B is more profitable, despite a lower win rate.

Red flags:

• ❌ Win rate > 85% (probably cherry-picked or curve-fit)

• ❌ Win rate < 30% (unless avg win >> avg loss)

What to trust:

• ✅ Win rate + profit factor (covered below)

• ✅ Win rate + avg win/loss ratio

Bottom line: Win rate is psychologically important (feels good to be right), but mathematically, it's less important than reward/risk ratio.

5. Profit factor

What it measures: Total profit divided by total loss.

Formula:

Profit Factor = Total Profit from Wins / Total Loss from Losses

Example:

• Total profit: $10,000

• Total loss: $4,000

• Profit factor: $10,000 / $4,000 = 2.5

What it reveals:

• For every $1 you lose, how much do you make?

What it hides:

• Trade distribution: Are your wins consistent, or is one massive win inflating the number?

• Risk of ruin: Even a profit factor of 2.0 can blow up if you hit a bad streak.

Red flags:

• ❌ Profit factor > 3.0 (check for overfitting)

• ❌ Profit factor relies on 1-2 huge wins (not repeatable)

What to trust:

• ✅ Profit factor > 1.5 with consistent wins

• ✅ Profit factor stable across multiple test periods

Bottom line: Profit factor is a great summary metric. If it's < 1.5, your strategy needs work.

6. Risk of ruin

What it measures: Probability of losing your entire account.

Formula (simplified):

Risk of Ruin ≈ ((1 - Win Rate) / (1 + Win Rate))^(Account / Risk per Trade)

Example:

• Win rate: 50%

• Risk per trade: 2%

• Account: $10,000

Risk of ruin: Very low (because risk per trade is small relative to account).

What it reveals:

• How likely you are to go broke, even with a profitable strategy

What it hides:

• Assumes consistent risk per trade (if you double down after losses, risk spikes)

• Assumes independent trades (if your strategy has correlated losses, risk is higher)

Red flags:

• ❌ Risk of ruin > 5% (too dangerous)

• ❌ Risking > 3% per trade (courting disaster)

What to trust:

• ✅ Risk 1-2% per trade max (keeps risk of ruin near zero)

Bottom line: Risk of ruin is the most important metric for survival. Ignore it, and you'll blow up—even with a profitable strategy.

The metrics you should ignore (or use carefully)

1. Number of trades

Why it's misleading:

• 100 trades in 1 year sounds impressive

• But if 90 are tiny winners and 10 are massive losses, you're not profitable

When it matters:

• If you have < 30 trades, your backtest isn't statistically significant

• If you have > 1,000 trades, you might be over-trading (slippage/commissions will kill you)

2. Avg win vs avg loss

Why it's misleading:

• "Avg win = $200, Avg loss = $100" sounds great

• But if you only win 30% of the time, you're losing money

When it matters:

• Pair with win rate to calculate expected value:

` Expected Value = (Win Rate × Avg Win) - (Loss Rate × Avg Loss) ``

3. Total number of consecutive wins

Why it's misleading:

• "10 consecutive wins!" sounds amazing

• But it's probably luck, not skill

When it matters:

• If your strategy requires consecutive wins to be profitable, it's fragile

How to interpret metrics together

Don't look at one metric in isolation. Use them as a system:

The "good strategy" checklist:

• ✅ Sharpe ratio > 1.0 (preferably > 1.5)

• ✅ Max drawdown < 20% (preferably < 15%)

• ✅ Profit factor > 1.5 (preferably > 2.0)

• ✅ Risk of ruin < 5% (achieved with 1-2% risk per trade)

• ✅ Win rate matches strategy type (mean reversion: 60-80%, trend: 30-50%)

• ✅ Drawdown-to-return ratio < 50%

Example:

Strategy A:

• Total return: 60%

• Sharpe: 1.8

• Max DD: 12%

• Profit factor: 2.2

• Win rate: 55%

Verdict: ✅ Good strategy. Proceed to walk-forward testing.

Strategy B:

• Total return: 120%

• Sharpe: 0.9

• Max DD: 35%

• Profit factor: 1.3

• Win rate: 48%

Verdict: ❌ Too risky. High return is offset by massive drawdowns and low Sharpe. Redesign.

How FlyTradr helps you analyze metrics

FlyTradr's Backtesting Lab calculates all key metrics automatically:

Performance dashboard:

• Returns: Total, annualized, monthly breakdown

• Risk metrics: Sharpe ratio, max drawdown, volatility

• Trade metrics: Win rate, profit factor, avg win/loss

• Equity curve: Visual representation of growth + drawdowns

Red flag detection:

FlyTradr highlights:

• ⚠️ Sharpe > 3.0 (possible overfitting)

• ⚠️ Max DD > 30% (high risk)

• ⚠️ Win rate > 85% (suspiciously high)

• ⚠️ Profit factor < 1.2 (marginally profitable)

Comparative analysis:

• Compare metrics across:

  • Different time periods (bull vs bear markets)

  • Different instruments (SPY vs QQQ)

  • Different parameter sets (RSI 10 vs RSI 14)

The bottom line

The metrics that matter most:

1. Sharpe ratio (risk-adjusted returns)

2. Max drawdown (worst-case pain)

3. Profit factor (win/loss ratio)

4. Risk of ruin (survival probability)

The metrics that are misleading:

1. Total return (without risk context)

2. Win rate (without profit context)

3. Avg win/loss (without frequency context)

The rule:

Never trust a single metric. Always look at the full picture: returns, risk, consistency, and survivability.

Next steps

1. Backtest your strategy in FlyTradr's Backtesting Lab

2. Review all metrics (not just total return)

3. Check for red flags (Sharpe > 3.0, DD > 30%, etc.)

4. If metrics pass, proceed to walk-forward testing

5. If they don't, redesign and retest

Remember: A strategy with mediocre returns and great risk metrics beats a strategy with amazing returns and terrible risk metrics.

Survivability > profitability.

Ready to analyze your strategy properly? Start backtesting with full metrics at FlyTradr.