Betis's surplus has a name. xGboard rates their keeping at +4.9 goals while their finishing is below xG, and Vallés leads LaLiga with +6.8 to +7.4 goals prevented.
Back to the trend of the day: Elche's goalkeeper makes six saves to hold Villarreal to a 0-0 draw
Poly Rating · from the trend of the day
LaLiga xG Table 2026-27: Points vs Expected Points
Real Betis have 19 points from chances worth about 12, the biggest overperformance in LaLiga, while Villarreal sit around six points short after their 20 shots against Elche brought only a 0-0.
- No. 1: Real Betis
- 10 places
01Final list
The leading AIs each built their own list of ten independently. Here is one result that is easy to check: every place has a reason, a source, and the list of AIs that included it.
A bright badge means this AI put the entry on its list. A dim badge means it did not.
- 1
Real Betis
Four models put them +6.1 to +7.1 points above expected. xGboard rates their keeping at +4.9 goals and their finishing below par, so the gap is Vallés, not the strikers.
Included it: ChatGPT, Kimi, DeepSeek. Did not include: Claude, Meta, Gemini, Wenxin.In their lists: ChatGPT: 1 · Kimi: 1 · DeepSeek: 1
What each AI noted 3
- ChatGPTplace 1
Betis sit 6.9 points above their model expectation despite finishing 1.8 goals below xG; goalkeeping is the remarkable counterweight.
Source ↗ - Kimiplace 1
The luckiest side in Spain: 19 points from performances worth 11.9 xPTS (+7.1) — and roughly a third of that 'luck' has a name: Álvaro Vallés (6.8 goals prevented).
Source ↗ - DeepSeekplace 1
Six points better than their chances, yet their finishing is 4.2 goals worse than xG — the gap is Vallés and timing.
Source ↗
3 AIs put it here - 2
Villarreal
They are −4.1 to −6.4 points short of expected, and the models disagree on how bad the finishing is: from −1.6 to −7.1 goals. Then 20 shots and 2.15 xG ended 0-0 against Elche.
Included it: Claude, ChatGPT, Gemini, Kimi, DeepSeek. Did not include: Meta, Wenxin.In their lists: Claude · ChatGPT: 2 · Gemini: 1 · Kimi: 2 · DeepSeek: 3
What each AI noted 5
- Claude
Named differently in its own list: open the full list.
- ChatGPTplace 2
A 7.1-goal finishing shortfall is the strongest warning signal: Villarreal have created substantially more than their results suggest.
Source ↗ - Geminiplace 1
Dominated Elche 20 shots to 7 with 2.1 xG in a 0-0 draw on October 17, expanding their league-leading underperformance deficit to -5.8 points below xPTS.
Source ↗ - Kimiplace 2
Unluckiest team in LaLiga — and today was the whole season in 90 minutes: 20 shots, 0 goals, 0-0 against Elche's six-save Matías Dituro.
Source ↗ - DeepSeekplace 3
20 shots, 2.12 xG, zero goals: the 0-0 with Elche turned a 5.1-point xPTS gap into 4.1.
Source ↗
5 AIs put it here - 3
Atlético Madrid
+3.6 to +6.0 points above expected on three models, with finishing +2.3 to +4.0 goals. They are now second on 22 points.
Included it: ChatGPT, Kimi, DeepSeek. Did not include: Claude, Meta, Gemini, Wenxin.In their lists: ChatGPT: 5 · Kimi: 3 · DeepSeek: 5
What each AI noted 3
- ChatGPTplace 5
Atlético's early 19 points exceeded their 14.9 expected points, with positive finishing and goalkeeping contributions reinforcing each other.
Source ↗ - Kimiplace 3
Second in the real table with 22 points — but their chances were worth about 13 xP after eight rounds (+6.0), the second-biggest over-performance in the division.
Source ↗ - DeepSeekplace 5
3.6 points above xPTS with +3.7 finishing and +1.3 keeper — both ends are overperforming.
Source ↗
3 AIs put it here - 4
Barcelona
A perfect 24 points is still +2.7 to +5.7 above expected. The models split on over-finishing, from +1.7 goals (xGboard) to +9.7 (MatchPulse).
Included it: Claude, ChatGPT, Gemini, Kimi, DeepSeek. Did not include: Meta, Wenxin.In their lists: Claude · ChatGPT: 9 · Gemini: 3 · Kimi: 4 · DeepSeek: 4
What each AI noted 5
- Claude
Named differently in its own list: open the full list.
- ChatGPTplace 9
Even a perfect 24 points from eight matches exceeds the model's already formidable 21.3 expected points.
Source ↗ - Geminiplace 3
Outscoring their underlying 18.2 xG by 5.8 goals to bank 24 actual points against 19.5 xPTS.
Source ↗ - Kimiplace 4
Perfect 8-0-0 with 34 goals from 27.9 xG — top of the real table AND the xG table, but still +4.3 to +5.7 points ahead of what their chances were worth.
Source ↗ - DeepSeekplace 4
3.6 points above xPTS but 31 goals from 21.3 xG — the gap is finishing, not luck.
Source ↗
5 AIs put it here - 5
Sevilla
+3.8 to +4.5 points over chances worth 8.5 to 10.2 xPTS. Neither finishing (+1.1 to +2.4) nor keeping explains it all.
Included it: ChatGPT, DeepSeek. Did not include: Claude, Meta, Gemini, Kimi, Wenxin.In their lists: ChatGPT: 6 · DeepSeek: 2
2 AIs put it here - 6
Real Sociedad
13 points from chances worth 8.3 to 10.0 xPTS (+3.0 to +4.7), despite allowing more xG than they create (10.7 vs 15.1).
Included it: ChatGPT, Gemini, Kimi. Did not include: Claude, Meta, DeepSeek, Wenxin.In their lists: ChatGPT: 3 · Gemini: 6 · Kimi: 5
What each AI noted 3
- ChatGPTplace 3
Thirteen points from chances worth just 8.8 expected points place Sociedad 4.2 points ahead of the model.
Source ↗ - Geminiplace 6
Controlled territorial possession and solid chance creation are muted by wasteful box execution, yielding 3.6 fewer points than simulated.
Source ↗ - Kimiplace 5
Sixth on 13 points despite a negative goal difference — their chances were worth just 8.3–10.0 xPTS, so they are +3.0 to +4.7 ahead of the story the numbers tell.
Source ↗
3 AIs put it here - 7
Espanyol
−2.7 to −4.7 points short of expected. On xGboard both finishing (+0.2) and keeping (−0.1) are at par, so most of the gap is bad timing; BetBuilders rates the keeping at −2.2.
Included it: ChatGPT, Gemini, Kimi, DeepSeek. Did not include: Claude, Meta, Wenxin.In their lists: ChatGPT: 7 · Gemini: 7 · Kimi: 6 · DeepSeek: 7
What each AI noted 4
- ChatGPTplace 7
A 3.8-point deficit appears even though both finishing and keeping sit almost exactly on their model expectations.
Source ↗ - Geminiplace 7
Anchored by Joan García's shot prevention, Espanyol sits on 10 points despite conceding 14.2 xGA against 10.9 xPTS expected.
Source ↗ - Kimiplace 6
12th on 8 points after nine games while their chance quality says mid-table: −4.7 (Understat) / −3.9 (BetBuilders) of deserved points have gone missing.
Source ↗ - DeepSeekplace 7
2.7 points below xPTS — the biggest negative gap outside the top six.
Source ↗
4 AIs put it here - 8
Athletic Club
−3.4 to −4.2 points short. xGboard has their finishing at −4.3 goals, while their keeper is +0.8 to +1.7.
Included it: ChatGPT, Gemini, DeepSeek. Did not include: Claude, Meta, Kimi, Wenxin.In their lists: ChatGPT: 4 · Gemini: 10 · DeepSeek: 6
What each AI noted 3
- ChatGPTplace 4
Athletic have generated 11.3 xG but finished 4.3 goals below expectation, helping explain a 4.2-point shortfall.
Source ↗ - Geminiplace 10
Stout shot-suppression and disciplined defensive structures hold opponents below xG, banking +2.5 points beyond simulated parity.
Source ↗ - DeepSeekplace 6
3.4 points below xPTS despite a +0.8 keeper; their 7 goals from 8.4 xG is the drag.
Source ↗
3 AIs put it here - 9
Málaga
Bottom with 3 to 4 points from chances worth 5.6 to 7.8 xPTS (−2.6 to −3.8). Most of it is finishing: 4 goals from 6.2 xG.
Included it: Kimi, DeepSeek. Did not include: Claude, ChatGPT, Meta, Gemini, Wenxin.In their lists: Kimi: 8 · DeepSeek: 8
What each AI noted 2
2 AIs put it here - 10
Elche
Twist: before the 0-0, xGboard (the only model with a figure) had Elche 3.1 points below expected, with keeping at −0.9. Dituro's 1.09 goals prevented at Villarreal came in one game.
Included it: Claude, ChatGPT, Gemini. Did not include: Meta, Kimi, DeepSeek, Wenxin.In their lists: Claude · ChatGPT: 8 · Gemini: 2
What each AI noted 3
- Claude
Named differently in its own list: open the full list.
- ChatGPTplace 8
Dituro's six saves against Villarreal earned a memorable clean sheet, yet Elche's wider early-season numbers still point to underperformance.
Source ↗ - Geminiplace 2
Surviving on pure goalkeeper variance (+4.3 prevented goals), accumulating +4.9 actual points above Understat's 4.1 xPTS projection.
Source ↗
3 AIs put it here
02Worth knowing
What the AIs noticed while building the list.
The models disagree badly on Villarreal's finishing: −7.1 goals on xGboard, −4.1 on statz.ai, −3.4 on OddAlerts and −1.6 on MatchPulse, for the same matches.
Before kick-off, xGboard gave Villarreal a 71% chance of beating Elche and Elche 12%. Elche created just 0.26 xG and still took a point.
Espanyol are −3.8 points short even though their finishing (+0.2) and keeping (−0.1) are at par on xGboard. Points gaps are not just shooting plus goalkeeping.
The tables come from different dates: Tactix after 7 matches (11 Oct), xGboard and statz.ai after 8, and the official table after 9. The latest round is not yet in any model.
03How we counted
In plain words, no formulas.
1The question
Top 10 LaLiga Teams Whose Points Most Differ From What Their Chances Say
2Why this topic
The reader sees which LaLiga sides sit higher or lower than their chances deserve, and which of those gaps belong to a goalkeeper, by comparing two xG providers with the real table.
3Where it applies
LaLiga (Spain), 2026-27 season so far, up to and including the Villarreal 0-0 Elche match on October 17, 2026
4What we looked at
- Gap between actual points and expected points from two separate xG models
- How much of each team's gap comes from its goalkeeper: goals conceded compared with post-shot expected goals (PSxG)
- How much comes from finishing: goals scored compared with xG
- Only the opening rounds have been played, so each gap is shown with a 'could still level out' note
- Tactix (Understat xPTS) — Actual points against expected points, after 7 matches (dated 11 Oct)
- statz.ai (Understat mirror) — xG, xGA and xPTS for each team after 8 matches
- xGboard — Expected points, xG and xGA, plus a finishing split (goals minus xG) and a keeping split in goals; snapshot of 17 Oct
- BetBuilders.ai (PlaymakerAI) — A separate xP model, with xG and xGA for each team
5Who ranked
The leading AIs, each building its own list of ten independently and naming a source for every place: Claude, ChatGPT, Meta, Gemini, Kimi, DeepSeek, Wenxin.
6How the places were set
The same entry written differently by different AIs was counted as one. The more AIs put an entry on their list, and the higher they put it, the higher its place. "AI agreement" shows how many of the AIs had it on their own list. Sources are only those the AIs found themselves; nothing was added by hand.
04Each AI's own list
🇺🇸 ChatGPT10 places · 4 notes
The same question, answered by ChatGPT alone. The final list above combines all the AIs.
- 1Real Betis — Spain, 2026–27
Betis sit 6.9 points above their model expectation despite finishing 1.8 goals below xG; goalkeeping is the remarkable counterweight.
Source ↗ - 2Villarreal CF — Spain, 2026–27
A 7.1-goal finishing shortfall is the strongest warning signal: Villarreal have created substantially more than their results suggest.
Source ↗ - 3Real Sociedad — Spain, 2026–27
Thirteen points from chances worth just 8.8 expected points place Sociedad 4.2 points ahead of the model.
Source ↗ - 4Athletic Club — Spain, 2026–27
Athletic have generated 11.3 xG but finished 4.3 goals below expectation, helping explain a 4.2-point shortfall.
Source ↗ - 5Atlético de Madrid — Spain, 2026–27
Atlético's early 19 points exceeded their 14.9 expected points, with positive finishing and goalkeeping contributions reinforcing each other.
Source ↗ - 6Sevilla FC — Spain, 2026–27
Sevilla hold 14 points despite producing only 10.6 xG and allowing 12.5 xGA.
Source ↗ - 7Espanyol — Spain, 2026–27
A 3.8-point deficit appears even though both finishing and keeping sit almost exactly on their model expectations.
Source ↗ - 8Elche CF — Spain, 2026–27
Dituro's six saves against Villarreal earned a memorable clean sheet, yet Elche's wider early-season numbers still point to underperformance.
Source ↗ - 9FC Barcelona — Spain, 2026–27
Even a perfect 24 points from eight matches exceeds the model's already formidable 21.3 expected points.
Source ↗ - 10Celta de Vigo — Spain, 2026–27
Celta's positive 3.9-goal keeping component has not prevented the club from sitting below expected points.
Source ↗
What ChatGPT noticed
The largest opposite-direction gaps belong to Betis (+6.9 points) and Villarreal (-6.0 points) in the retrieved xGboard snapshot. Their league positions conceal sharply different underlying results.
Villarreal's finishing is 7.1 goals below expectation, whereas Betis outperform expected points despite negative finishing. A goalkeeper can materially change that story.
Espanyol's -3.8-point gap alongside finishing of +0.2 and keeping of -0.1 shows why actual-minus-expected points cannot be explained solely through shooting and goalkeeping totals.
Before the Villarreal–Elche fixture, xGboard assigned Villarreal a 71% win probability and Elche just 12%. The reported 0–0 illustrates why a likely result is never a guaranteed one.
🇺🇸 Gemini10 places · 3 notes
The same question, answered by Gemini alone. The final list above combines all the AIs.
- 1Villarreal CF
Dominated Elche 20 shots to 7 with 2.1 xG in a 0-0 draw on October 17, expanding their league-leading underperformance deficit to -5.8 points below xPTS.
Source ↗ - 2Elche CF
Surviving on pure goalkeeper variance (+4.3 prevented goals), accumulating +4.9 actual points above Understat's 4.1 xPTS projection.
Source ↗ - 3FC Barcelona
Outscoring their underlying 18.2 xG by 5.8 goals to bank 24 actual points against 19.5 xPTS.
Source ↗ - 4Valencia CF
Plagued by cold finishing (-4.1 goals below xG) and opponent hot runs, stranding them 4.3 points below their statistical expectation.
Source ↗ - 5Real Madrid
World-class individual finishing converts tight 1.4-xG games into regular 3-point hauls, driving a +3.9 points-above-model divergence.
Source ↗ - 6Real Sociedad
Controlled territorial possession and solid chance creation are muted by wasteful box execution, yielding 3.6 fewer points than simulated.
Source ↗ - 7RCD Espanyol
Anchored by Joan García's shot prevention, Espanyol sits on 10 points despite conceding 14.2 xGA against 10.9 xPTS expected.
Source ↗ - 8Real Valladolid
Conceding goals on opponents' lowest-probability attempts leaves them 3.1 points behind their xPTS floor.
Source ↗ - 9Girona FC
Regression in offensive conversion has pulled their actual tally 2.8 points below expected baseline simulations.
Source ↗ - 10Athletic Club
Stout shot-suppression and disciplined defensive structures hold opponents below xG, banking +2.5 points beyond simulated parity.
Source ↗
What Gemini noticed
Goalkeeping accounts for the single largest individual component of early-season points divergence: Elche and Espanyol have gained over 7 combined points purely from their keepers' PSxG outperformance.
Over small sample sizes (under 10 rounds), finishing variance routinely swings league tables by 4 to 6 points, but historical data shows nearly 80% of teams regress toward their xPTS mean by matchday 25.
Villarreal's 0-0 stalemate with Elche on October 17 stands as LaLiga's starkest single-match xPTS divergence of the round, failing to convert 20 shots and 2.1 xG into all three points.
🇨🇳 Kimi10 places · 4 notes
The same question, answered by Kimi alone. The final list above combines all the AIs.
- 1Real Betis (LaLiga 2026-27)
The luckiest side in Spain: 19 points from performances worth 11.9 xPTS (+7.1) — and roughly a third of that 'luck' has a name: Álvaro Vallés (6.8 goals prevented).
Source ↗ - 2Villarreal (LaLiga 2026-27)
Unluckiest team in LaLiga — and today was the whole season in 90 minutes: 20 shots, 0 goals, 0-0 against Elche's six-save Matías Dituro.
Source ↗ - 3Atlético Madrid (LaLiga 2026-27)
Second in the real table with 22 points — but their chances were worth about 13 xP after eight rounds (+6.0), the second-biggest over-performance in the division.
Source ↗ - 4FC Barcelona (LaLiga 2026-27)
Perfect 8-0-0 with 34 goals from 27.9 xG — top of the real table AND the xG table, but still +4.3 to +5.7 points ahead of what their chances were worth.
Source ↗ - 5Real Sociedad (LaLiga 2026-27)
Sixth on 13 points despite a negative goal difference — their chances were worth just 8.3–10.0 xPTS, so they are +3.0 to +4.7 ahead of the story the numbers tell.
Source ↗ - 6RCD Espanyol (LaLiga 2026-27)
12th on 8 points after nine games while their chance quality says mid-table: −4.7 (Understat) / −3.9 (BetBuilders) of deserved points have gone missing.
Source ↗ - 7Valencia CF (LaLiga 2026-27)
19th with 5 points — but their chances were worth 8.9 xP: a −3.9 gap where both the keeper (−2.4) and the finishers (−1.0) are below par.
Source ↗ - 8Málaga CF (LaLiga 2026-27)
Bottom with 4 points from four draws — winless not because they are outclassed (7.8–9.3 xPTS) but because they finish −2.2 below their chances.
Source ↗ - 9Real Madrid (LaLiga 2026-27)
Fourth on 18 points — two points behind Betis despite a far better chance profile (16.0–16.6 xPTS): the models say Madrid, not Betis, should be Barcelona's nearest chaser.
Source ↗ - 10CA Osasuna (LaLiga 2026-27)
The league's most keeper-hurt team: 18 goals conceded against 12.6 xGA (−5.4) turned a chances-worthy 11.3 xP into just 8 real points.
Source ↗
What Kimi noticed
Betis's +7.1 'luck' is actually measurable skill in one position: strip out Vallés' league-leading 6.8 goals prevented and their gap falls by roughly a third — no other over-performer has such a clear single-player explanation.
Villarreal's 0-0 with Elche was their season in miniature: 20 shots to 7, zero goals, and six saves from 37-year-old Matías Dituro — their −6.4 xPTS gap is both bad finishing (−4.1 vs xG) and below-par keeping (−3.5 vs xGA).
The models disagree on how bad Villarreal's finishing really is: Understat credits them 17.1 xG, BetBuilders only 14.4 — a 2.7-xG swing on the same eight matches, which is why both numbers are quoted.
Everything here is an eight-or-nine-round sample: statz.ai's own back-test shows teams running hot go on to win slightly less often than rated — the tradeable signal is fading the flattered (Betis, Atlético), not backing the unlucky.
🇨🇳 DeepSeek10 places · 4 notes
The same question, answered by DeepSeek alone. The final list above combines all the AIs.
- 1Real Betis
Six points better than their chances, yet their finishing is 4.2 goals worse than xG — the gap is Vallés and timing.
Source ↗ - 2Sevilla
4.5 points above xPTS with only +1.1 finishing and -0.6 keeper — the early table is flattering them.
Source ↗ - 3Villarreal
20 shots, 2.12 xG, zero goals: the 0-0 with Elche turned a 5.1-point xPTS gap into 4.1.
Source ↗ - 4Barcelona
3.6 points above xPTS but 31 goals from 21.3 xG — the gap is finishing, not luck.
Source ↗ - 5Atlético Madrid
3.6 points above xPTS with +3.7 finishing and +1.3 keeper — both ends are overperforming.
Source ↗ - 6Athletic Club
3.4 points below xPTS despite a +0.8 keeper; their 7 goals from 8.4 xG is the drag.
Source ↗ - 7Espanyol
2.7 points below xPTS — the biggest negative gap outside the top six.
Source ↗ - 8Málaga
2.6 points below xPTS at the bottom; the chances say they should have nearly twice as many points.
Source ↗ - 9Valencia
2.2 points below xPTS; only 4 goals from 6.2 xG.
Source ↗ - 10Celta Vigo
1.8 points below xPTS; a quiet early-season gap that could still level out.
Source ↗
What DeepSeek noticed
Betis’s +6.1 gap is not a finishing hot streak: they have scored 9 from 13.2 xG (-4.2), so the points above expectation come mostly from the goalkeeper end and from timing.
Villarreal are the unluckiest side by this measure: 14.6 xG and 9.1 xGA should put them around 13.1 xPTS, but they have 9 points after the 0-0 with Elche, where Dituro made six saves.
Barcelona’s gap is a finishing story: 31 goals from 21.3 xG (+9.7), while their goalkeeper end is almost neutral (7 GA from 6.9 xGA).
The two xG models disagree materially on Villarreal: Tactix/Understat gives them 13.1 xPTS, while MatchPulse has 14.6 xGF and 9.1 xGA — different shot data, same broad conclusion.