Backtest: Buy VLO at the close when Brent crude falls more than 1% on the day and VLO c...
The setup sounds reasonable: when crude sells off hard and Valero holds its ground, the market is pricing in fatter refining margins. Betting on a few more sessions of that decoupling has a certain logic to it. The backtest does not agree with that logic.
Across 67 closed trades, the strategy returned -26.24% on $100,000 in starting capital, winning only 26.9% of the time and trailing a simple SPY buy-and-hold by 94.54 percentage points. The best trade gained 12.11%, but the worst lost 6.76% — and the losing trades came far more often. The full breakdown is in the analysis below, including the equity curve, trade-by-trade results, and what actually went wrong.
Buy VLO at the close when Brent crude falls more than 1% on the day and VLO closes flat or higher; exit after 4 trading days or on a 2% stop-loss, whichever comes first. A refiner holding gains on a down-crude tape signals widening crack margins and oil-to-equity decoupling that tends to carry over for several sessions.
How this was measured
This is a simulated backtest generated from the plain-English strategy below, executed bar-by-bar on historical market data using the price + news data mode with $100,000 starting capital. Strategy: Buy VLO at the close when Brent crude falls more than 1% on the day and VLO closes flat or higher; exit after 4 trading days or on a 2% stop-loss, whichever comes first. A refiner holding gains on a down-crude tape signals widening crack margins and oil-to-equity decoupling that tends to carry over for several sessions.
The key numbers
The charts
The takeaway
The strategy returned -26.24% on $100,000 starting capital across 67 closed trades with a 27% win rate. Over the same window SPY buy-and-hold returned +68.30%, so the strategy finished trailing the benchmark by 94.54 points. Best single trade +12.11%, worst -6.76%.
The fine print
- Simulated results on historical data — fills, slippage and costs are idealized.
- Past performance does not predict future results.