Predictions
Premier League 2026/27 predictions
Updated October 28, 2026 · 10,000 simulated seasons
Footygami's goals model plays the rest of the 2026/27 season 10,000 times every morning, from the table as it stands, 80 of 380 matches in. Arsenal win the title in 54% of those seasons and Manchester City in 41%; Coventry City (70%), Ipswich Town (65%) and Sunderland (43%) go down most often.
54%
Arsenal to win the title
56%
Brighton to finish in the top four
70%
Coventry City to go down
300
Matches left to play
The predicted table
Each club's points now, the points it finishes on in an average simulated season, and how often it wins the title, finishes in the top four or goes down. Clubs are in order of their projected points.
| # | Club | Pts now | Proj. ptsPts | Title | Top 4 | Down |
|---|---|---|---|---|---|---|
| 1 |  ArsenalArsenal | 21(8) | 81 | 54% | 99% | – |
| 2 |  Manchester CityMan City | 21(8) | 79 | 41% | 98% | – |
| 3 |  LiverpoolLiverpool | 15(8) | 65 | 2% | 63% | <1% |
| 4 |  Brighton & Hove AlbionBrighton | 16(8) | 64 | 2% | 56% | <1% |
| 5 |  ChelseaChelsea | 13(8) | 58 | <1% | 26% | <1% |
| 6 |  Aston VillaAston Villa | 11(8) | 55 | <1% | 12% | 1% |
| 7 |  BrentfordBrentford | 11(8) | 54 | <1% | 12% | 2% |
| 8 |  AFC BournemouthBournemouth | 9(8) | 53 | <1% | 8% | 3% |
| 9 |  Newcastle UnitedNewcastle | 11(8) | 52 | <1% | 7% | 3% |
| 10 |  Leeds UnitedLeeds United | 11(8) | 52 | <1% | 7% | 3% |
| 11 |  Hull CityHull City | 10(8) | 50 | <1% | 4% | 5% |
| 12 |  Manchester UnitedMan United | 7(8) | 49 | <1% | 4% | 6% |
| 13 |  EvertonEverton | 10(8) | 48 | – | 2% | 9% |
| 14 |  Tottenham HotspurTottenham | 9(8) | 45 | – | 1% | 18% |
| 15 |  FulhamFulham | 8(8) | 45 | <1% | 1% | 17% |
| 16 |  Nottingham ForestNott'm Forest | 6(8) | 44 | – | 1% | 18% |
| 17 |  Crystal PalaceCrystal Palace | 7(8) | 41 | – | <1% | 36% |
| 18 |  SunderlandSunderland | 5(8) | 40 | – | <1% | 43% |
| 19 |  Ipswich TownIpswich Town | 10(8) | 36 | – | <1% | 65% |
| 20 |  Coventry CityCoventry City | 6(8) | 35 | – | <1% | 70% |
Where each club could finish
How often each club finishes in each place across the 10,000 simulated seasons, coloured from teal for the least likely to pink for the likeliest, as in the key below. A blank square happened in fewer than one season in 200.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Arsenal
Man City
Liverpool
Brighton
Chelsea
Aston Villa
Brentford
Bournemouth
Newcastle
Leeds United
Hull City
Man United
Everton
Tottenham
Fulham
Nott'm Forest
Crystal Palace
Sunderland
Ipswich Town
Coventry City
Who will win the Premier League in 2026/27?
Arsenal are favourites: they win the title in 54% of 10,000 simulated seasons, ahead of Manchester City (41%). They have 21 points from 8 matches, and the model expects them to finish on about 81.
Who will be relegated from the Premier League in 2026/27?
Coventry City (70%), Ipswich Town (65%) and Sunderland (43%) go down most often in the simulated seasons, with Crystal Palace (36%) next. The bottom three go down.
Who will finish in the top four in 2026/27?
Arsenal (99%), Manchester City (98%), Liverpool (63%), Brighton & Hove Albion (56%), Chelsea (26%) and Aston Villa (12%) finish in the top four most often. The top four qualify for the Champions League, and a fifth place joins them in seasons when English clubs have done well enough in Europe.
How the predictions work
Every morning the bot fits the goals model behind Scorigami Watch to the last three seasons' results, recent matches counting most: each club gets an attacking and a defensive rating, and home sides a boost. It then plays each of the season's remaining matches 10,000 times, every score drawn from the two sides' expected goals, and nudges every club's strength a little at random in each simulated season, for the uncertainty in its rating. Clubs finish on points, then goal difference, then goals scored.
It knows results only: not injuries, transfers, a new manager or the bookmakers' prices. So it can't see a side that has just lost its best player, and it's slow to believe a sudden change of form.
How well has it done?
We tested it on the last ten seasons, 2016/17 to 2025/26, predicting from Matchweeks 9, 19 and 29 with only the results before them. The club it made favourite went on to win the title 26 times in 30. Of the clubs it gave between a 50% and an 80% chance of going down, 76% went down, and of those it gave 80% or more, 37 of 38 did. All 54 clubs it gave 95% or more for the top four got there. If anything it was too cautious about clear favourites: the clubs it gave a 50–80% chance of the title won it 13 times in 15.