Backtest: Buy OXY at the close when its daily news-sentiment score is in the top quinti...
A burst of positive OXY headlines that the stock sells into looks like a mispricing — the kind of sentiment reset that reverses once energy flows reprice the news. Tested bar-by-bar on historical price and news data, that premise does not hold up.
The rule buys at the close when OXY's news-sentiment score sits in the top quintile of its prior 20 sessions but the stock finished lower, exiting after five trading days or on a 2% stop. Across 82 closed trades it returned -38.77% on $100,000, winning 23.2% of the time. Over the same window SPY buy-and-hold returned +76.34%, a gap of 115 points.
Winners did exist — the best single trade gained 14.43% — but they arrived too rarely to carry the book. The full trade-level evidence, charts and statistics follow below.
Buy OXY at the close when its daily news-sentiment score is in the top quintile of its prior 20 sessions but the stock closes lower; exit after 5 trading days or on a 2% stop-loss, whichever comes first. Positive OXY-specific headlines being sold into usually mark a sentiment reset that gets reversed once energy flows reprice the news.
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 OXY at the close when its daily news-sentiment score is in the top quintile of its prior 20 sessions but the stock closes lower; exit after 5 trading days or on a 2% stop-loss, whichever comes first. Positive OXY-specific headlines being sold into usually mark a sentiment reset that gets reversed once energy flows reprice the news.
The key numbers
The charts
The takeaway
The strategy returned -38.77% on $100,000 starting capital across 82 closed trades with a 23% win rate. Over the same window SPY buy-and-hold returned +76.34%, so the strategy finished trailing the benchmark by 115.11 points. Best single trade +14.43%, worst -3.42%.
The fine print
- Simulated results on historical data — fills, slippage and costs are idealized.
- Past performance does not predict future results.