EPL Predictions 2026: United, Villa Negatively Regress

Two Premier League clubs fit the profile of negative regression candidates relative to their market expectations.

Critical to these EPL predictions are how Premier League odds view these teams. For example, had Sunderland received a higher market rating, they would join this duo.

But these teams are projected to finish near last year’s placement, creating an inefficiency. Let’s dive into the predictions – odds are reflective at time of writing and are subject to movement.

EPL Predictions 2026: 2 Best Negative Regression Picks

Manchester United (-175 for Top-4 Finish)

United surged after sacking Ruben Amorim and replacing him with Michael Carrick. The former United player earned a top-4 spot with a 12-3-2 (W-D-L) finish.

But the Red Devils received a lot of help to reach that spot, including time spent under Amorim.

Manchester United scored 69 goals across all of 2025-26. The figure came against 72.3 expected goals, suggesting a small positive regression case.

However, United managed 69 goals off 58.94 post-shot xG, a far more predictive stat in terms of goal output.

The attack overperformed its psxG by at least 0.7 goals seven times. Meanwhile, United keepers kept out 3.5 goals above expectation.

Just under Carrick, United delivered 33 goals off 26.38 post-shot xG. The fourth-place finishers overperformed expectations by at least 0.5 psxG in seven of those 17 fixtures.

Now you place United into a heavier fixture load. Next season, the Red Devils are set to compete in four competitions, whereas they played in three last year.

That’s concerning for a team that took 13 of 20 wins by a single goal. Given the added fixture congestion, I predict United fails to meet its market expectation.

Aston Villa (+110 for Top-6, -600 for Top-10 Finish)

A world exists where Aston Villa takes an absolute dive from fourth place. While 11th or worse appears highly unlikely, it must go acknowledged how lucky Villa ran last year.

Unai Emery’s side clinched Champions League football off consistently lucky results. The team finished 2025-26 with a +7 goal differential vs. a -0.6 expected goal differential.

It gets worse when you look at post-shot xG. Villa delivered that same +7 differential against a -2.7 psxGDiff. That qualifies as a worse psxGDiff than 14th-place Leeds United.

Villa took all three points in 19 matches, fourth-best in the league. Thirteen of those 19 victories came by a single goal.

Most of Villa’s negative regression is expected to hit their attack. The unit notched 56 goals against 56.2 expected goals, but only 49.46 post-shot xG.

That sample includes five matches where Villa scored at least one goal over its post-shot expectation.

Without accounting for red cards or penalty kicks, Villa ranked 12th in expected points last year. Expect a decline against last year’s finishing position due to these metrics.