APA vs XOP 20d forward edge after APA EPS upward revision with flat/down Brent
An upward EPS revision while Brent is flat to down sounds like a clean stock-specific signal: the company is executing even as the commodity backdrop drags on the group. The data over the past three years says otherwise. Across APA’s quarterly estimate updates, only two episodes matched the exact condition—a positive one-month consensus revision for next-quarter EPS alongside flat-to-down Brent. In both cases, APA lagged XOP over the following 20 trading days, with an average excess return of -5.88% versus a -0.40% baseline. Zero of the two events were positive.
With a sample that small, the apparent edge is essentially anecdote. The difference from baseline is nowhere near statistically meaningful (p≈0.32), and the events themselves contradict the thesis. The full study below details the methodology, the two episodes, and why this setup cannot yet be treated as a signal.
For APA over the past ~3 years, when consensus next-quarter EPS estimates are revised upward over a month while Brent crude is flat to down over the same period, does APA outperform XOP over the next 20 trading days? I expect analyst upgrades against a weak crude tape to signal company-specific execution and balance-sheet strength, so capital flows into APA even as the commodity beta drags on the E&P group.
How this was measured
Daily closes for APA and XOP were aligned on common trading days over the prior three years, and 20-trading-day forward returns were computed at each anchor. From APA_estimates, quarterly rows in the same window were used as event anchors: the study kept rows where the available 30-day consensus EPS change was positive and Brent crude's trailing 30-calendar-day return was 0% or lower. Each event was anchored to the first common trading day on or after the estimate date, and APA minus XOP forward excess returns were compared against the unconditional all-day baseline.
The key numbers
Reading the numbers
Headline: across only 2 qualifying events, APA trailed XOP by about 5.9 percentage points on average over the next 20 days, and 0 of 2 events were positive. That is worse than the -0.4% baseline, but with p=0.32 the edge is not statistically clear.
The charts
This histogram shows only two event outcomes, both sitting below zero: one near -8.9% and one near -2.9%. The average of the two bars is about -5.9%, meaning APA underperformed XOP in both qualifying episodes after an upward revision with flat-to-down Brent. With just two observations, the picture is more anecdote than evidence, but it does not support the idea that upgrades against a weak crude tape helped APA.
The event group centers below zero at about -5.9%, while the 712-day baseline sits much closer to zero at about -0.4% and spans a wide range from about -19% to +25%. So the event average looks worse than the everyday pattern, not better. However, with only two event points and a p-value of 0.32, the gap is not statistically distinguishable from baseline, so this should not be read as a reliable signal.
Qualifying event windows
| estimate_date | anchor_date | eps_rev_30d | brent_ret_30d | apa_fwd20 | xop_fwd20 | apa_minus_xop |
|---|---|---|---|---|---|---|
| 2023-12-31 | 2024-01-02 | 0.04 | -0.0131 | -0.1253 | -0.0366 | -0.0887 |
| 2024-09-30 | 2024-09-30 | 0.01 | -0.0979 | -0.0182 | 0.0108 | -0.029 |
The takeaway
The short answer is no: in the 2 qualifying episodes over the past three years, APA did not outperform XOP after an upward EPS revision with flat/falling Brent — it lagged badly. Both events produced negative APA-minus-XOP returns, averaging -5.88% over the next 20 trading days, versus a -0.40% baseline for all days; zero of the two events were positive. But with only 2 events, this is essentially a coin flip dressed up as a data point: the difference from baseline has a p-value around 0.32, nowhere near what you'd call a clear signal, and the event mean isn't distinguishable from zero (p≈0.30). The direction of the two cases actually contradicts the thesis — in Dec 2023 and Sep 2024, APA's upgrade-with-weak-crude episodes were followed by underperformance — but the sample is too thin to call it a real negative edge either. Practical takeaway: there is no reliable evidence that this setup predicts APA beating XOP, so it shouldn't be treated as a signal until far more qualifying events exist.
The fine print
- Only 2 qualifying events in 3 years — too few for any statistical confidence.
- Quarterly estimates data can't capture true monthly revision episodes, so event frequency is capped at quarterly.
- Forward 20-day windows can overlap, which may overstate the significance of any test.
- APA is part of XOP, so this measures single-name relative alpha, not a hedge against the E&P group.