Backtest: Buy PSX at the close when its 10-day total return is more than 4 percentage p...
A refiner that falls more than four points behind an integrated major while crude is climbing is supposed to be a coiled spring: the product crack catches up, the gap closes, the trade pays. PSX against XOM, filtered by Brent's ten-day return, put that idea through 26 closed trades.
It finished higher 53.8% of the time — a coin flip with a slight edge. The trouble was what the wins were worth. The best single trade added 6.57%, the worst gave back 10.09%, and the whole run returned 5.85% on $100,000 starting capital. Over the same window, SPY buy-and-hold beat it by 70.48 points.
The setup triggers often enough to be worth understanding. Why the payoff never compounded is the more useful question, and the trade-by-trade evidence sits below.
Buy PSX at the close when its 10-day total return is more than 4 percentage points below XOM's while Brent crude's 10-day return is positive; exit when PSX closes above its 5-day simple moving average or after 10 trading days, whichever comes first. A refiner lagging an integrated major during a crude upswing is discounting a delayed product-crack catch-up, so the gap tends to close once the crude signal is absorbed.
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 PSX at the close when its 10-day total return is more than 4 percentage points below XOM's while Brent crude's 10-day return is positive; exit when PSX closes above its 5-day simple moving average or after 10 trading days, whichever comes first. A refiner lagging an integrated major during a crude upswing is discounting a delayed product-crack catch-up, so the gap tends to close once the crude signal is absorbed.
The key numbers
The charts
The takeaway
The strategy returned +5.85% on $100,000 starting capital across 26 closed trades with a 54% win rate. Over the same window SPY buy-and-hold returned +76.34%, so the strategy finished trailing the benchmark by 70.48 points. Best single trade +6.57%, worst -10.09%.
The fine print
- Simulated results on historical data — fills, slippage and costs are idealized.
- Past performance does not predict future results.