Learn how to test trading strategies across multiple stocks simultaneously to validate robustness and identify best-performing symbols.
Time required: 15 minutes
Why this matters: A strategy that works on one stock might fail on others. Multi-symbol testing proves robustness.
Result: You'll have 5 separate backtest runs, one per symbol. Go to Backtesting → Results tab to compare them side-by-side.
Once all backtests complete, analyze which symbols work best:
| Symbol | Return | Sharpe | Max DD | Trades |
|---|---|---|---|---|
| AAPL | +12.5% | 1.4 | -15% | 18 |
| MSFT | +15.8% | 1.7 | -12% | 22 |
| GOOGL | +8.2% | 1.1 | -18% | 15 |
| NVDA | +22.3% | 1.6 | -25% | 20 |
| AMD | -3.2% | -0.2 | -28% | 25 |
Analysis: Strategy works well on MSFT, AAPL, NVDA. Fails on AMD. Avoid AMD or investigate why (high volatility? Different price action?).
Let's optimize RSI oversold threshold for MSFT:
| RSI Oversold | Return | Sharpe | Trades | Win Rate |
|---|---|---|---|---|
| 20 | +11.2% | 1.3 | 8 | 62% |
| 25 | +14.5% | 1.6 | 12 | 58% |
| 30 | +15.8% | 1.7 | 15 | 55% |
| 35 | +14.1% | 1.5 | 18 | 52% |
| 40 | +12.3% | 1.3 | 22 | 49% |
Optimal value: 30 - Best Sharpe ratio (1.7), high return, reasonable trade count. Use this value going forward.
Always optimize for risk-adjusted returns (Sharpe ratio), not raw returns. High return with huge drawdown is bad.
If you test 1,000 combinations, one will look perfect by chance. That's not edge, that's luck.
Skipping validation on unseen data. Always test optimal parameters on different time period or symbols.
