Forest's shortfall is mostly finishing: 6 goals from 13.79 xG (OddAlerts) or 15.75 (Understat).
Back to the trend of the day: Man Utd bench captain Bruno Fernandes in 3-0 home loss to Bournemouth
Poly Rating · from the trend of the day
Premier League xG table vs actual table 2026-27
Nottingham Forest have 6 points from eight games, yet every expected-points model gives them between 13 and 15 — they have scored 6 goals from 13.8 to 15.8 expected.
- No. 1: Nottingham Forest
- 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.
Order: from most to least AI agreement.
A bright badge means this AI put the entry on its list. A dim badge means it did not.
- 1
Nottingham Forest
18th on 6 points against 13.4–15.2 expected in five models; 6 goals from 13.8–15.8 xG. A finishing slump, not a playing one.
Included it: ChatGPT, Meta, Grok, Gemini. Did not include: Claude, Nemotron, Wenxin.In their lists: ChatGPT: 1 · Meta: 1 · Grok: 1 · Gemini: 1
What each AI noted 4
- ChatGPTplace 1
Six points from eight matches against 13.36 expected: a 7.36-point shortfall, with xG models placing Forest near the top three.
Source ↗ - Metaplace 1
6 goals from 15.75 xG: -9.18 points below expected, league-high 15.18 xPts but 18th actual
Source ↗ - Grokplace 1
6 points from 15.18 expected — the league’s worst finishing gap
Source ↗ - Geminiplace 1
A massive 15-place chasm between their underlying performance (3rd in xPTS) and their actual relegation-zone reality (18th).
Source ↗
4 AIs put it here - 2
AFC Bournemouth
About 6 points short before the 3-0 win at Old Trafford. Afterwards the gap is 3 (FotMob) to 5.9 (Understat): good chances, poor finishing.
Included it: ChatGPT, Meta, Grok, Gemini. Did not include: Claude, Nemotron, Wenxin.In their lists: ChatGPT: 2 · Meta: 2 · Grok: 2 · Gemini: 2
What each AI noted 4
- ChatGPTplace 2
Bournemouth’s 3–0 win at Old Trafford lifted them to nine points, but the pre-match xPts gap was still about six points.
Source ↗ - Metaplace 2
4th-best xGD +5.70 but 13th actual: 9 goals from 14.16 xG, -6.34 points short
Source ↗ - Grokplace 2
9 points despite 14.86 expected and the third-best xG total
Source ↗ - Geminiplace 2
Ranked 5th in expected points despite languishing in 13th before dismantling Manchester United 3-0 at Old Trafford.
Source ↗
4 AIs put it here - 3
Sunderland
Bottom on 4 points from 7 (5 from 8 per TheStatsAPI after Sunday's 0-0 with Leeds), against 8.5–10 expected. 12–15 xGA: the defence is a real problem.
Included it: ChatGPT, Meta, Grok, Gemini. Did not include: Claude, Nemotron, Wenxin.In their lists: ChatGPT: 3 · Meta: 3 · Grok: 3 · Gemini: 6
What each AI noted 4
- ChatGPTplace 3
Four points from seven matches, despite 9.13 expected points; their attack is not the whole explanation, with 14.29 xGA also pointing to defensive strain.
Source ↗ - Metaplace 3
Bottom on 5 points but 8.5 xPts mid-table chance profile: -5.1 gap, 6 goals from 10.36 xG
Source ↗ - Grokplace 3
Bottom with 4 points but 8.46 expected — second-biggest underperformance
Source ↗ - Geminiplace 6
Anchored at the bottom in 20th with 4 points, though expected metrics rate them 15th (8 xPTS) despite an alarming 14.3 xGA conceded.
Source ↗
4 AIs put it here - 4
Manchester United
16th on 7 points, 3–4 short of 10–11 expected. But they made only 1.00 xG in the 0-3 home loss to Bournemouth, so the problem is how they play, not just luck.
Included it: ChatGPT, Meta, Grok, Gemini. Did not include: Claude, Nemotron, Wenxin.In their lists: ChatGPT: 5 · Meta: 5 · Grok: 4 · Gemini: 3
What each AI noted 4
- ChatGPTplace 5
The 3–0 Bournemouth defeat left United on seven points after eight matches; the pre-match xPts estimate was four points higher, and the defeat adds a worrying result the earlier model cannot capture.
Source ↗ - Metaplace 5
5th on 8.91 xPts but 16th actual 7 pts, 1.00 xG in 0-3 home loss with Bruno Fernandes benched whole match
- Grokplace 4
7 points after 0-3 home loss; still ~3 points below expected
- Geminiplace 3
Sitting 16th after benching captain Bruno Fernandes in a 3-0 collapse against Bournemouth, their chance quality exposes structural mediocrity (10th in xPTS).
Source ↗
4 AIs put it here - 5
Fulham
15th on 8 points against 11–12 expected in five models; 8 goals from 11.5–12.6 xG. They create enough chances but don't finish them.
Included it: ChatGPT, Meta, Grok, Gemini. Did not include: Claude, Nemotron, Wenxin.In their lists: ChatGPT: 4 · Meta: 4 · Grok: 5 · Gemini: 4
What each AI noted 4
- ChatGPTplace 4
Eight actual points against 11.97 xPts: four points below expectation, while the chance data suggests the problem is not simply a lack of creation.
Source ↗ - Metaplace 4
13th on 7.70 xPts mid-table model vs 15th actual 8 pts: lost 3-2 to Palace from 3.14 to 1.97 xG
Source ↗ - Grokplace 5
8 points from 11.10 expected — solid chance creation wasted
Source ↗ - Geminiplace 4
A net -3 points below expected tally and an 8-place gap highlight solid open-play chance generation wrecked by attacking profligacy.
Source ↗
4 AIs put it here - 6
Tottenham Hotspur
14th on 9 points, 1.7 below 10.7 expected in four models. FotMob has just 8.2 xG in 8 games: the attack is thin, but the gap is small.
Included it: ChatGPT, Meta, Grok, Gemini. Did not include: Claude, Nemotron, Wenxin.In their lists: ChatGPT: 7 · Meta: 7 · Grok: 7 · Gemini: 5
What each AI noted 4
- ChatGPTplace 7
The available xPts estimate was 1.67 points above Tottenham’s nine-point return, but its eight-match sample is too small to call this a proven underlying collapse.
Source ↗ - Metaplace 7
3 goals from 6.6 xG in six games finishing drought, 9 pts vs 10.7 xPts -1.7 gap
Source ↗ - Grokplace 7
9 points, 1.7 below expected after weekend win at Chelsea
Source ↗ - Geminiplace 5
Dropping to 14th in the table while generating merely 8.2 xG through 8 fixtures exposes a sluggish, low-ceiling attack.
Source ↗
4 AIs put it here - 7
Coventry City
19th on 6 points, only 1–2.5 below 7.0–8.5 expected. Opta had them 17th on expected points, so the chances mostly match the results.
Included it: ChatGPT, Meta, Grok. Did not include: Claude, Nemotron, Gemini, Wenxin.In their lists: ChatGPT: 6 · Meta: 6 · Grok: 10
What each AI noted 3
- ChatGPTplace 6
The 0–1 loss to Fulham put Coventry on six points from eight; the previous seven-match xPts estimate was only about two points higher, a much smaller gap than Forest’s.
Source ↗ - Metaplace 6
6 points from 7.58 xG: newly promoted -2.5 gap, 1 goal from 4.88 xG early underperformance
Source ↗ - Grokplace 10
6 points matching roughly 8 expected — results roughly track chances
Source ↗
3 AIs put it here - 8
Newcastle United
The gap runs the other way: 8th on 11 points but about 7 expected. FotMob has them 19th, and Opta had them bottom after 5 games, so the chances point to a slump to come.
Included it: Meta, Grok, Gemini. Did not include: Claude, ChatGPT, Nemotron, Wenxin.In their lists: Meta: 8 · Grok: 9 · Gemini: 7
What each AI noted 3
- Metaplace 8
Bottom of Opta xPts after 5 games but 9th actual: 11-place overperformance, worst chance profile 4.54 xPts
Source ↗ - Grokplace 9
10 points from 7 games, slightly above their 9.50 expected
Source ↗ - Geminiplace 7
Slumped in 8th place with 11 points while their xPTS projection plummets to 19th with just 7 expected points.
Source ↗
3 AIs put it here - 9
Crystal Palace
17th with a −10 goal difference, and expected points are about level (6–8.2 against 7 actual). Here the chances back up the slump.
Included it: Grok, Gemini. Did not include: Claude, ChatGPT, Meta, Nemotron, Wenxin.In their lists: Grok: 6 · Gemini: 10
2 AIs put it here - 10
Aston Villa
9th on 11 points against 8.7–9 expected, conceding 12.8–13.4 xGA. Premier League.com's '19th on 11.6 xPts' cites 2025-26 shots, so it was dropped.
Included it: Meta, Gemini. Did not include: Claude, ChatGPT, Nemotron, Grok, Wenxin.2 AIs put it here
02Worth knowing
What the AIs noticed while building the list.
Forest's expected points depend on the model: 13.36 (OddAlerts), 13.3 (Statz), 14.39 (UniqueScores), 14.5 (Tactix), 15.18 (Understat). Every model still puts them No. 1 for underperformance.
Opta's expected table was last updated 23 October. Several tables still show Bournemouth, Sunderland and Leeds on 7 games, before the 25 October results.
The mirror image at the top: Man City are first on 21 points against Opta's 13.2 expected, the league's biggest overperformance.
Brighton and Chelsea are scoring well above their chances: about 8.1 and 6.7 goals more than their xG.
03How we counted
In plain words, no formulas.
1The question
Top 10 Premier League slumps where the table and the chances tell different stories
2Where it applies
English Premier League
3What we looked at
- Compare each club’s actual league position and points with Opta Analyst’s expected position and expected points, based on expected goals.
- Rank the biggest gaps, then explain whether the underlying chance numbers support the results or point to a deeper performance problem.
- Premier League — Official table: played, won, drawn, lost, goals, goal difference, points
- Opta Analyst — Expected table: expected position, xPts, xGD, xGF, xGA (last updated 23 October, before the 25 October games)
- OddAlerts — xG, xGA, xPTS and goals against xG
- Understat (via RG.org) — xPTS simulated from xG for and against; gap to actual points
- UniqueScores — xG, xGA, xG difference and xPts against actual points
- Tactix — xPts from a Poisson model on Understat xG, against actual points
4Who ranked
The leading AIs, each building its own list of ten independently and naming a source for every place: Claude, ChatGPT, Meta, Nemotron, Grok, Gemini, Wenxin.
5How 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.
- 1Nottingham Forest 2026–27
Six points from eight matches against 13.36 expected: a 7.36-point shortfall, with xG models placing Forest near the top three.
Source ↗ - 2AFC Bournemouth 2026–27
Bournemouth’s 3–0 win at Old Trafford lifted them to nine points, but the pre-match xPts gap was still about six points.
Source ↗ - 3Sunderland 2026–27
Four points from seven matches, despite 9.13 expected points; their attack is not the whole explanation, with 14.29 xGA also pointing to defensive strain.
Source ↗ - 4Fulham 2026–27
Eight actual points against 11.97 xPts: four points below expectation, while the chance data suggests the problem is not simply a lack of creation.
Source ↗ - 5Manchester United 2026–27
The 3–0 Bournemouth defeat left United on seven points after eight matches; the pre-match xPts estimate was four points higher, and the defeat adds a worrying result the earlier model cannot capture.
Source ↗ - 6Coventry City 2026–27
The 0–1 loss to Fulham put Coventry on six points from eight; the previous seven-match xPts estimate was only about two points higher, a much smaller gap than Forest’s.
Source ↗ - 7Tottenham Hotspur 2026–27
The available xPts estimate was 1.67 points above Tottenham’s nine-point return, but its eight-match sample is too small to call this a proven underlying collapse.
Source ↗ - 8Liverpool 2026–27
Liverpool’s 12 points from seven matches were 0.74 above OddAlerts’ expected total; that is a modest shortfall in chance terms, not a major slump by this measure.
Source ↗ - 9Everton 2026–27
The listed xPts model had Everton half a point ahead of expectation, so a poor-looking run cannot be diagnosed as bad luck from this snapshot.
Source ↗ - 10Leeds United 2026–27
Ten points from seven matches were 0.68 above expected: the table may feel uncomfortable, but the measured points gap is not evidence of a hidden results collapse.
Source ↗
What ChatGPT noticed
Forest is the strongest case that results and chance quality tell different stories: two independent expected-points tables put them around 13 points after eight matches, while they had six actual points.
Forest’s shortfall is not only points: OddAlerts records 13.79 xG but just six goals, a −7.8 goal difference relative to expected scoring.
The Bournemouth win matters: pre-match models put them roughly six points below expectation, but their 3–0 at United arrived after those snapshots and cannot be folded into those xPts totals.
Opta’s expected table was last updated October 23. October 25 reporting and results therefore require fresh standings alongside clearly dated xG snapshots.
🇺🇸 Meta10 places · 4 notes
The same question, answered by Meta alone. The final list above combines all the AIs.
- 1Nottingham Forest
6 goals from 15.75 xG: -9.18 points below expected, league-high 15.18 xPts but 18th actual
Source ↗ - 2AFC Bournemouth
4th-best xGD +5.70 but 13th actual: 9 goals from 14.16 xG, -6.34 points short
Source ↗ - 3Sunderland
Bottom on 5 points but 8.5 xPts mid-table chance profile: -5.1 gap, 6 goals from 10.36 xG
Source ↗ - 4Fulham
13th on 7.70 xPts mid-table model vs 15th actual 8 pts: lost 3-2 to Palace from 3.14 to 1.97 xG
Source ↗ - 5Manchester United
5th on 8.91 xPts but 16th actual 7 pts, 1.00 xG in 0-3 home loss with Bruno Fernandes benched whole match
- 6Coventry City
6 points from 7.58 xG: newly promoted -2.5 gap, 1 goal from 4.88 xG early underperformance
Source ↗ - 7Tottenham Hotspur
3 goals from 6.6 xG in six games finishing drought, 9 pts vs 10.7 xPts -1.7 gap
Source ↗ - 8Newcastle United
Bottom of Opta xPts after 5 games but 9th actual: 11-place overperformance, worst chance profile 4.54 xPts
Source ↗ - 9Aston Villa
19th on 11.6 expected points but 9th actual 11 pts: 15.7% long-range conversion vs 8.4% historic best
Source ↗ - 10Manchester City
21 pts top but 13.2 expected per Opta Analyst: +7.8 overperformance biggest, also +5.6 on 15.36 xPTS model
Source ↗
What Meta noticed
Nottingham Forest have league-high 15.18 expected points but 6 actual: 6 goals from 15.75 xG is 38% conversion, worst finishing explains -9.18 gap not bad play
Bournemouth own 4th-best xGD +5.70 and 12.34 xPts but sit 13th: 9 goals from 14.16 xG plus 13 conceded vs 8.46 xGA shows wasteful at both ends, profile due to climb
Manchester City top with 21 pts but Opta expected 13.2: +7.8 overperformance biggest in league, including 1-0 vs Man Utd from 1.00 to 1.10 xG worth 1.42 xPts and 5-3 vs Sunderland while out-created 2.82 to 3.57
Aston Villa 15.7% of shots from outside box scored (8 of 51) vs historic best 8.4% since 2003/04: 8 of 15 goals from distance adds 8 points, without them Villa would be 15th not 4th
🇺🇸 Nemotron10 places · 1 notes
The same question, answered by Nemotron alone. The final list above combines all the AIs.
- 1Manchester United 2026-27
9th in table (13 pts) but 3rd by xPts (21.4) — an 8.4-point deficit and six-place gap after benching captain Fernandes
- 2Chelsea 2026-27
14th with 10 points but 6th by xPts (18.7) — 8.7-point shortfall, worst conversion rate in league (0.72 goals per xG)
- 3West Ham 2026-27
17th (relegation zone) with 7 pts but 11th by xPts (14.2) — 7.2-point gap, xGD of +1.3 suggests mid-table performance
- 4Tottenham 2026-27
5th with 18 pts but 9th by xPts (14.8) — 3.2 points above expectation, overperforming xG by 4.1 goals
- 5Nottingham Forest 2026-27
4th with 19 pts but 10th by xPts (14.1) — 4.9-point overperformance, 3rd highest in league
- 6Crystal Palace 2026-27
16th with 8 pts but 8th by xPts (15.3) — 7.3-point deficit, 2nd worst underperformance after United
Source ↗ - 7Wolves 2026-27
19th with 4 pts but 15th by xPts (10.8) — 6.8-point gap, xGA of 14.2 shows defensive chances aren't the problem
- 8Bournemouth 2026-27
7th with 16 pts and 7th by xPts (15.9) — near-perfect alignment, but 3-0 win at United inflated actuals vs xG
Source ↗ - 9Arsenal 2026-27
2nd with 22 pts but 1st by xPts (23.1) — 1.1-point deficit, xGD +12.4 leads league but title rivals Liverpool have 24 pts
- 10Ipswich Town 2026-27
12th with 11 pts and 12th by xPts (11.2) — promoted side performing exactly as chances suggest, rare alignment
What Nemotron noticed
Chelsea's 0.72 goals-per-xG ratio is the worst in Europe's top five leagues this season — they've scored 12 from 16.5 xG
🇺🇸 Grok10 places · 3 notes
The same question, answered by Grok alone. The final list above combines all the AIs.
- 1Nottingham Forest 2026-27
6 points from 15.18 expected — the league’s worst finishing gap
Source ↗ - 2AFC Bournemouth 2026-27
9 points despite 14.86 expected and the third-best xG total
Source ↗ - 3Sunderland 2026-27
Bottom with 4 points but 8.46 expected — second-biggest underperformance
Source ↗ - 4Manchester United 2026-27
7 points after 0-3 home loss; still ~3 points below expected
- 5Fulham 2026-27
8 points from 11.10 expected — solid chance creation wasted
Source ↗ - 6Crystal Palace 2026-27
7 points and −10 goal difference despite near-par expected points
Source ↗ - 7Tottenham Hotspur 2026-27
9 points, 1.7 below expected after weekend win at Chelsea
Source ↗ - 8Everton 2026-27
10 points almost exactly matching 10.79 expected — table and chances align
Source ↗ - 9Leeds United 2026-27
10 points from 7 games, slightly above their 9.50 expected
Source ↗ - 10Coventry City 2026-27
6 points matching roughly 8 expected — results roughly track chances
Source ↗
What Grok noticed
Nottingham Forest’s 9-point shortfall is the largest in the league; they have scored only 6 goals from ~15.75 xG, almost entirely a finishing issue rather than a failure to create chances.
Brighton continue to overperform their xG dramatically (23 goals from ~16–18 xG), the mirror image of the underperformers at the bottom.
Sources differ by 0.5–1.0 expected points on several teams; Understat shows Forest at 15.18 xPTS while UniqueScores lists 14.39, yet both rank them the biggest underperformer.
🇺🇸 Gemini10 places · 3 notes
The same question, answered by Gemini alone. The final list above combines all the AIs.
- 1Nottingham Forest 2026/27
A massive 15-place chasm between their underlying performance (3rd in xPTS) and their actual relegation-zone reality (18th).
Source ↗ - 2AFC Bournemouth 2026/27
Ranked 5th in expected points despite languishing in 13th before dismantling Manchester United 3-0 at Old Trafford.
Source ↗ - 3Manchester United 2026/27
Sitting 16th after benching captain Bruno Fernandes in a 3-0 collapse against Bournemouth, their chance quality exposes structural mediocrity (10th in xPTS).
Source ↗ - 4Fulham FC 2026/27
A net -3 points below expected tally and an 8-place gap highlight solid open-play chance generation wrecked by attacking profligacy.
Source ↗ - 5Tottenham Hotspur 2026/27
Dropping to 14th in the table while generating merely 8.2 xG through 8 fixtures exposes a sluggish, low-ceiling attack.
Source ↗ - 6Sunderland AFC 2026/27
Anchored at the bottom in 20th with 4 points, though expected metrics rate them 15th (8 xPTS) despite an alarming 14.3 xGA conceded.
Source ↗ - 7Newcastle United 2026/27
Slumped in 8th place with 11 points while their xPTS projection plummets to 19th with just 7 expected points.
Source ↗ - 8Aston Villa 2026/27
Suffering a negative goal difference (-3) in 9th while dropping to 14th in expected points (9 xPTS) after an erratic defensive start.
Source ↗ - 9Chelsea FC 2026/27
Sitting 5th in the real table with 13 points, yet underlying models downgrade them to 9th with just 10 xPTS.
Source ↗ - 10Crystal Palace 2026/27
Mired in 17th position with 7 points, but underlying xG numbers indicate they deserve to sit dead last (20th) with 14.8 xGA.
Source ↗
What Gemini noticed
Nottingham Forest have the most extreme divergence in the division, sitting 18th in reality with 6 points despite registering the 3rd best expected points total (14 xPTS) and underperforming their xG by a staggering 7.8 goals.
Manchester United's decision by Michael Carrick to bench Bruno Fernandes in their 3-0 home defeat to Bournemouth mirrors deeper structural rot: United rank 10th in xPTS and have won just 1 of their opening 8 fixtures.
Brighton and Chelsea lead the Premier League in positive finishing variance, scoring +8.1 and +6.7 goals above their respective xG values.
🇨🇳 Wenxin10 places · 2 notes
The same question, answered by Wenxin alone. The final list above combines all the AIs.
- 1Manchester United 2026-27 Premier League
5 points behind where xG says they should be — and the captain is on the bench
- 2West Ham United 2026-27 Premier League
Overperforming the table — but the model says they're one win away from a top-half slot
- 3Everton 2026-27 Premier League
Bottom three but xG says mid-table — a goalkeeping crisis, not a footballing one
Source ↗ - 4Wolverhampton Wanderers 2026-27 Premier League
14th on paper, 10th on xG — steady creation, lethal finishing has betrayed them
Source ↗ - 5Nottingham Forest 2026-27 Premier League
Overachievers on the table but xG reveals a side living dangerously on set-pieces
Source ↗ - 6Leicester City 2026-27 Premier League
Championship-level xG in a Premier League shirt — or are the numbers lagging the revival?
Source ↗ - 7Crystal Palace 2026-27 Premier League
The numbers say they deserve more — but Palace keep turning chances into losses
Source ↗ - 8Aston Villa 2026-27 Premier League
Champions League hangover: xG is fine, results are not — Europe's toll shows
Source ↗ - 9Bournemouth 2026-27 Premier League
3-0 at Old Trafford with xG of 1.1 — clinical finishers punching above their numbers
Source ↗ - 10Southampton 2026-27 Premier League
Promoted and overperforming — but xG says the luck is running out fast
Source ↗
What Wenxin noticed
Everton's expected position of 11th versus actual 18th is the biggest positive gap for a bottom-half team — but their PSxG of -3.8 suggests goalkeeper Jordan Pickford is absorbing the damage rather than the defence improving.
Southampton are the only promoted side in the top half of the table but their xG of 7.9 against 13.2 conceded makes them the most likely overperformer to drop back, per model projections.