Bournemouth
v
Liverpool
Both teams to score · 64% model probability · odds 1.44
Model Goals 3.2 Corners 11.8 (5.6H · 6.2A) Cards 4.1 BTTS 64% O2.5 62%
Odds Home 3.10 Draw 3.75 Away 2.15 BTTS 1.44 O2.5 1.53
Leeds
v
Crystal Palace
Leeds to win · 56% model probability · odds 1.70 ▼ Steaming 19.1%
Model Goals 3.0 Corners 9.9 (5.7H · 4.2A) Cards 3.6 BTTS 57% O2.5 57%
Odds Home 1.70 Draw 4.00 Away 4.75 BTTS 1.67 O2.5 1.67
Manchester City
v
Sunderland
Manchester City to win · 72% model probability · odds 1.33
Model Goals 3.1 Corners 11.5 (6.5H · 5.0A) Cards 3.8 BTTS 49% O2.5 60%
Odds Home 1.33 Draw 5.25 Away 8.00 BTTS 1.91 O2.5 1.57
Fulham
v
Manchester United
Both teams to score · 63% model probability · odds 1.50
Model Goals 3.2 Corners 10.3 (5.2H · 5.1A) Cards 4.1 BTTS 63% O2.5 62%
Odds Home 3.60 Draw 3.90 Away 1.95 BTTS 1.50 O2.5 1.53
| # | Team | P | W | D | L | GD | Pts | Form |
|---|---|---|---|---|---|---|---|---|
| 1 | Manchester City | 4 | 4 | 0 | 0 | 6 | 12 | WWWW |
| 2 | Arsenal | 5 | 4 | 0 | 1 | 4 | 12 | LWWWW |
| 3 | Brighton | 5 | 3 | 1 | 1 | 11 | 10 | WWDLW |
| 4 | Brentford | 5 | 2 | 3 | 0 | 6 | 9 | WDDDW |
| 5 | Everton | 5 | 2 | 3 | 0 | 3 | 9 | WDDDW |
| 6 | Leeds | 4 | 2 | 2 | 0 | 4 | 8 | WDDW |
| 7 | Hull City | 5 | 2 | 2 | 1 | 2 | 8 | LDDWW |
| 8 | Newcastle | 5 | 2 | 2 | 1 | 0 | 8 | WLDWD |
| 9 | Chelsea | 5 | 2 | 1 | 2 | -2 | 7 | LDLWW |
| 10 | Liverpool | 4 | 1 | 3 | 0 | 2 | 6 | DWDD |
| 11 | Ipswich | 5 | 2 | 0 | 3 | -4 | 6 | LWLLW |
| 12 | Nottingham Forest | 5 | 1 | 2 | 2 | -1 | 5 | LWDDL |
| 13 | Manchester United | 4 | 1 | 1 | 2 | 0 | 4 | LDWL |
| 14 | Sunderland | 4 | 1 | 1 | 2 | -2 | 4 | LDWL |
| 15 | Aston Villa | 5 | 1 | 1 | 3 | -5 | 4 | WLDLL |
| 16 | Bournemouth | 4 | 0 | 3 | 1 | -1 | 3 | DDDL |
| 17 | Crystal Palace | 4 | 1 | 0 | 3 | -5 | 3 | LWLL |
| 18 | Coventry | 5 | 1 | 0 | 4 | -9 | 3 | WLLLL |
| 19 | Tottenham | 5 | 0 | 2 | 3 | -6 | 2 | LDDLL |
| 20 | Fulham | 4 | 0 | 1 | 3 | -3 | 1 | DLLL |
The Premier League is the most heavily traded football competition in the world, and that shapes everything about how we bet it. Bookmaker prices here are sharpened by enormous liquidity, which means genuine value is rarer than in any other league we cover. When our model does flag a Premier League selection, it has cleared a market that is very hard to beat — so you will see fewer picks here, per fixture, than in League Two or the National League. That is by design, not a gap in coverage.
Structurally, the Premier League is a high-scoring division by English standards. Recent seasons have settled around 2.8 to 3.0 goals per game, comfortably above the Championship, driven by the concentration of elite attacking talent at the top six clubs. That gap between the top and bottom of the table also produces more short-priced favourites than any other English division, and short favourites are exactly where bookmaker margins hide. The model treats a heavy favourite the same way it treats any other selection: it needs the probability to beat the price, and at 1.20 the price rarely leaves room.
Draws land at roughly 22 to 25 per cent of matches, at the lower end of the big European leagues, and home advantage — historically worth around 45 per cent home wins — has never fully returned to its pre-2020 level. Both teams to score tends to hover near 50 per cent, which makes BTTS a genuine coin-flip market where small modelling edges matter.
Early in the season our team ratings lean on the previous campaign until enough 2026/27 matches accumulate — usually around six rounds. Promoted clubs are the hardest to price in August and September because their Championship numbers translate imperfectly, and the model widens its uncertainty on their fixtures accordingly. By October the ratings are running almost entirely on current-season data.
Where we find most of our Premier League value is away from the match result: goals lines in mid-table fixtures where the market anchors on reputation rather than current scoring rates, and correct score, where even an efficient market has to spread its probability across many outcomes.
Every Premier League selection we publish is archived at kickoff and settled against the odds available at publication. The full record, including this league on its own, is on the model results page — losses included.