Energy Dip-Buying Keeps Trailing the Index in the Platform's Backtests

When crude gets jittery, the reflex in energy coverage is to reach for the integrateds — the idea being that Exxon and Chevron offer a tempered, dividend-cushioned way to play a barrel. The platform's research runs this week keep landing on the same uncomfortable answer: in the actual data, that tempering doesn't show up where the thesis says it should, and the tactical version of the trade has been expensive.

The asymmetry that isn't there

Two findings out of the platform this week poke at the same assumption from different angles, and both come back negative. The first is XOM's five-day realized beta to Brent: the hypothesis was that Exxon gets more oil-sensitive when crude sells off, which would make it a worse hedge than it looks. What the data showed was the opposite of dramatic — mean beta of 0.279 when Brent's trailing five-day return was negative versus 0.300 when it was positive, a gap of -0.021 running in the wrong direction. The medians tell the same story (0.260 down versus 0.294 up), so this isn't a handful of outliers dragging the average around.

The second, published a day later, looked at Chevron around earnings. The intuition was that company-specific news would dilute oil sensitivity in earnings week. The average five-day event beta came in at 0.187 against 0.248 over the prior 20 sessions — a paired difference of -0.060 that looks directionally right, but the finding is explicit that the average is doing all the work. That is not a foundation to build a view on.

A headline tape with no catalyst

If you went looking for a crude catalyst in this week's news flow, you'd come up empty. The headline set tagged to XOM is a Form 4 filing for Airbnb Class A with the shares up 2.35%, Liberty Global's Class C stock touching a 52-week low at $8.93 after a 23.72% decline over the past year with the RSI reading oversold, and the SEC clearing a Tesla-crafted plan that automatically opts investors into management-aligned proxy votes. All neutral in sentiment, none of it about oil.

That matters because the geopolitical-premium trade has its own research problem. Studying five-day forward spreads after Iran/Hormuz headline-intensity spikes, Brent averaged -0.35% while XOM lost only -0.08%; on quiet days XOM actually gained 0.33% against Brent's -0.10%. The Brent-minus-XOM spread was -0.26% after spikes versus -0.42% on quiet days. The premium people expect to collect in the barrel doesn't reliably show up — and the raw numbers lean away from it, not toward it.

Three backtests, one verdict

The tactical layer is where the evidence gets blunt. Buying DVN at the close after a 2%-plus down day returned -14.26% on $100,000 across 29 closed trades — with a 59% win rate — while SPY buy-and-hold returned +85.28% over the same window, a 99.53-point gap. Buying XOM after a 2%-plus Brent down day: -12.52% across 32 trades, a 44% win rate, trailing by 97.80 points. Buying CVX after a 1%-plus Brent up day: -5.15% across 49 trades, a 59% win rate, trailing by 90.42 points.

The win rates are the interesting part. Two of the three strategies won more often than they lost and still lost money, because the loss tail is fat: worst trades of -15.30%, -13.38% and -5.54% against bests of +6.97%, +5.46% and +4.18%. That is the signature of a setup that harvests small, frequent gains and occasionally gets run over — a payoff shape that looks fine right up until it doesn't.

None of this argues the energy majors are broken businesses, and none of it is a forecast. It argues that the shopworn tactical playbook — buy the integrateds on a crude dip, lean into the Hormuz headline — has been a reliably expensive way to express an oil view, and this week's batch of findings is the latest place that keeps surfacing. The data leans one way. Worth remembering the next time a geopolitical headline hits the tape and the reflex kicks in.