La Liga's True Table: What xPoints Says After Seven Weeks
The La Liga table after seven weeks says Barcelona are perfect, Real Betis are the league's surprise package, and Villarreal are a mid-table afterthought.
Expected points says something else entirely: Betis are the table's biggest lie, and Villarreal might be the third-best team in Spain.
The method, in one paragraph
For each of the 69 played La Liga matches this season, we took both teams' Opta expected goals and ran them through independent Poisson distributions: the probability of every scoreline, converted to win/draw/loss probabilities, converted to expected points (3 for a win, 1 for a draw). Sum those across a team's games and you get its xPoints: how many points its chance creation and concession deserved. No finishing luck, no referee luck, no woodwork.
The true table
| # | Team | P | xPts | Actual pts | Diff |
|---|---|---|---|---|---|
| 1 | Barcelona | 7 | 17.83 | 21 | +3.17 |
| 2 | Real Madrid | 7 | 15.22 | 15 | -0.22 |
| 3 | Villarreal | 7 | 13.26 | 8 | -5.26 |
| 4 | Atlético de Madrid | 7 | 12.76 | 16 | +3.24 |
| 5 | Athletic Club | 6 | 10.13 | 8 | -2.13 |
| 6 | Alavés | 7 | 9.89 | 11 | +1.11 |
| 7 | Osasuna | 7 | 9.54 | 8 | -1.54 |
| 8 | Espanyol | 7 | 9.43 | 7 | -2.43 |
| 9 | Sevilla | 7 | 9.05 | 13 | +3.95 |
| 10 | Real Betis | 7 | 8.81 | 16 | +7.19 |
| 11 | Celta de Vigo | 7 | 8.80 | 7 | -1.80 |
| 12 | Elche | 7 | 8.41 | 5 | -3.41 |
| 13 | Real Sociedad | 7 | 8.38 | 10 | +1.62 |
| 14 | Deportivo de A Coruña | 7 | 8.21 | 10 | +1.79 |
| 15 | Rayo Vallecano | 7 | 8.11 | 8 | -0.11 |
| 16 | Racing de Santander | 7 | 8.10 | 7 | -1.10 |
| 17 | Valencia | 7 | 6.99 | 4 | -2.99 |
| 18 | Getafe | 7 | 6.59 | 8 | +1.41 |
| 19 | Levante | 6 | 6.38 | 5 | -1.38 |
| 20 | Málaga | 7 | 6.20 | 3 | -3.20 |
Real Betis are living a lie
Sixteen points from seven games. One spot off second. Nine goals scored. And 8.81 expected points, tenth in the true table.
Betis's entire season is one match: the 1-0 home win over Real Madrid on September 4, from an xG of 0.93 to 3.46. Our model gave Betis a 6.1% chance of winning that game. It was the single biggest xG upset of the La Liga season so far, and it accounts for nearly half of their +7.19 overperformance. Strip out that one evening and Betis are what their chance creation says they are: a mid-table side grinding out 1-0s (their wins over Valencia and Getafe came from a combined 3.01 xG).
The market will keep pricing Betis like a top-four team. The model says: fade.
Villarreal: the best team the table doesn't know
Flip it around. Villarreal sit ninth on 8 points. The model has them third on 13.26 xPoints, a -5.26 gap that is the largest underperformance in the league.
Look at the games: a 2-2 draw at Atlético from an xG of 3.60 to 0.78. A 1-0 loss at Alavés from 1.74 to 0.73. A 2-3 home defeat to Deportivo from 1.20 to 0.83. Three matches, three results that went against the run of play, roughly four and a half points left on the table. When they finally met a team at their level of misfortune, they produced the most dominant performance of the season so far: 3-1 over Levante from 4.27 to 0.45 xG.
Villarreal's attack is generating chances at a top-three rate. Finishing variance this extreme does not persist. Back the bounce.
The rest of the true table
Barcelona are legit, but mortal. Seven wins from seven is real dominance (17.83 xPts leads the league), and 31 goals scored is absurd. But +3.17 says even perfection has been flattered; the gap to Madrid is closer to three points than six.
Sevilla are Betis-lite. Thirteen points from 9.05 xPts (+3.95): the model has them ninth, the table has them fifth. Their finishing is running hot.
The relegation watch is about chance creation, not points. Valencia have scored 4 goals in 7 games and sit on 4 points; the model says they deserve 6.99, which is still relegation form. Málaga are winless with 3 goals scored, and their 6.20 xPts says the winless run is earned, not unlucky. Elche are the interesting one: 5 points from 8.41 xPts, and their 2-3 home loss to Real Sociedad came from an xG of 2.37 to 0.57 (the model gave Elche a 77.6% win probability). If any bottom-three side climbs, it's Elche.
What to do with this
Seven games is a small sample, and xPoints is a model, not a prophecy: it assumes finishing is luck, and over a full season that assumption is one of the safest in football analytics. The edges it flags now, Betis to fade and Villarreal to back, are exactly the kind that get priced in by November. Check back then; we'll re-run the table and see who listened.
Where the numbers come from
Every figure is computed from per-match Opta expected goals and results for 69 played La Liga 2026/27 matches (Aug 15 to Sep 20, 2026), queried live from the Dribble API on Oct 2, 2026. Expected points come from independent Poisson distributions on each side's xG. No pundit vibes, no eye test.
Want the underlying data? The same xG feeds behind this model are available as BigQuery tables, API endpoints, and downloadable files. See plans and pricing →