Backtest: When Brent crude closes up more than 1% and PSX closes down on the same day,...
The setup is easy to root for: crude spikes, Phillips 66 drops, and the thesis says the market has overdone the input-cost squeeze — buy the refiner at the close, hold five sessions, let crack spreads re-price. On its face, the trade worked: the strategy returned 54.83% on $100,000. But a passive SPY holder made 68.30% over the same window. It made money; it just didn't beat doing nothing.
That gap is the real story. The study below walks through the exact entry rule — Brent up more than 1%, PSX down on the same session, exit after five trading days — and then breaks down what worked, what didn't, and where the edge actually showed up.
The full evidence is in the analysis: trade-by-trade results, benchmark comparison, best and worst outcomes, and the limits of the simulation.
When Brent crude closes up more than 1% and PSX closes down on the same day, buy PSX at the close; exit after 5 trading days. A refiner falling on an oil-up day signals an overdone input-cost squeeze, and crack spreads re-price with a lag, so PSX tends to rebound.
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: When Brent crude closes up more than 1% and PSX closes down on the same day, buy PSX at the close; exit after 5 trading days. A refiner falling on an oil-up day signals an overdone input-cost squeeze, and crack spreads re-price with a lag, so PSX tends to rebound.
The key numbers
The charts
The takeaway
The strategy returned +54.83% on $100,000 starting capital across 87 closed trades with a 56% win rate. Over the same window SPY buy-and-hold returned +68.30%, so the strategy finished trailing the benchmark by 13.47 points. Best single trade +17.37%, worst -11.72%.
The fine print
- Simulated results on historical data — fills, slippage and costs are idealized.
- Past performance does not predict future results.