Energy-Folklore Rules Are Failing the Numbers Test
The oil market is a breeding ground for heuristics. Higher volatility means we should avoid the ETF. Relative strength should mean reversion. And a beaten-down oil stock is a bargain if crude is oversold. These rules sound reasonable—until you put them in front of the data. The latest run of platform research on oil-linked instruments suggests that a lot of this common wisdom is not just unhelpful; it is often pointed in the wrong direction.
The Contango Roll-Cost Reflex
Start with the most sacred cow: USO versus Brent. The standard story is that when Brent volatility spikes, the contango structure erodes the ETF's returns, creating a negative gap between USO and Brent over the next ten days. The data says otherwise. Across 712 trading days, the 178 days with top-quartile Brent realized volatility produced an average 10-day gap of +1.27%, compared with +0.41% on all other days. That is a high-minus-low difference of +0.86%—the exact opposite of what the roll-cost and contango press would have you expect. This is a small edge, to be sure, but it is a direct contradiction of a pervasive narrative. The market is not simply a machine for punishing leveraged commodity exposure in volatile times.
The Relative Strength Trap
Another well-worn idea: if an energy stock has been persistently strong or weak relative to the market, the next move should fade it. That idea fails as well. Over roughly three years and 694 overlapping daily observations, XLE's 10-day relative strength versus SPY did not meaningfully predict the next 10-day return. When Brent traded below its 50-day moving average, the slope was negative at -0.216 with a p-value of about 0.078. That directionally matches the mean-reversion hypothesis, but it is weak and fragile—and in the 409 observations in that regime, it clearly does not rise to the level of a dependable signal. The story repeats in the volatility space: HAL's 20-day realized volatility relative to Brent's was not a precursor to above-baseline forward returns. Top-quintile days averaged +1.26% over the next 10 days versus +2.14% for the bottom 80%. The signal pointed the wrong way, and with only 31 top-quintile signals, confidence is thin.
When the Strategy Backtests Poorly
Then there are the simple rule-based strategies. Buying OXY when Brent's 14-day RSI is below 40 and OXY's own 14-day RSI is also low sounded like a classic oversold bounce. It delivered -10.02% over seven closed trades with a 29% win rate. Over the same window, SPY buy-and-hold returned +68.30%, meaning the strategy underperformed by 78.32 percentage points. A different idea—buying the oil refiner PSX when Brent rises more than 1% and PSX closes down on the same day—did better, returning +54.83% across 87 trades with a 56% win rate. But it still trailed SPY by 13.47 points. Two very different rulebooks, same verdict: the market is not giving up easy edge.
The Curve Regime That Never Comes
Finally, we should flag the structural difficulty in testing these heuristics. A study on COP's beta to Brent across Treasury yield-curve regimes had 729 classifiable days—but only three of those days had the yield spread above its trailing 12-month average. That is not a test; it is a footnote. COP's daily beta to Brent was 0.3516 on ordinary below-average days, essentially the full-sample beta of 0.3520. The regime distinction collapses under the weight of its own non-occurrence. When a market spends almost no time in the state you are trying to study, the assumption that the state matters becomes unprovable.
The broader takeaway from this batch of research is that the energy complex is not friendly to simple narratives. The usual reflexes—fade strength, buy weakness, expect contango drag, rely on regime flips—are either absent or inverted in the recent data. That does not mean the market is random. It means the rules we carry around are too coarse for the structure actually at work. The next time someone recites an oil-market proverb, the first question should be: what does the data say?