01 · 2 legs
Safe
- Double chance Brentford or draw (1X)62%
- Over 1.5 goals77%
Premier LeagueGtech Community Stadium, London16°21 km/h
Läs på svenskaAnna Karlsson · · A data-driven preview · How the predictions work
The model probabilities paint a picture of balance, with both Brentford and Chelsea given a 36% chance to win. Yet, there's a subtle lean towards the hosts. Despite the small sample size of their head-to-head record since 2022, where the visitors hold a slight edge, the current form and underlying metrics suggest a different narrative. Brentford are unbeaten in their last six league matches, while Chelsea have found the net in eight consecutive games. These streaks indicate a competitive clash, likely played to each side's strengths — both teams favor a low defensive line with a direct approach.
The expected goals figures, Brentford at 1.36 and Chelsea at 1.42, further highlight the tight contest ahead. The model suggests a 1-1 draw as the most probable outcome, making it hard to separate the two in terms of offensive prowess. Looking at the over/under market, there's a slight tilt towards over 2.5 goals at 53%, suggesting goals might flow despite both sides' defensive profiles.
However, the sharpest angle lies in the value for under 0.5 goals. With odds at 20.0 across multiple bookmakers like Betsafe, Betsson, and Nordicbet, the model's 7% rating presents a substantial 31% edge. While unlikely, the potential payout makes this a tempting long shot.
01 · 2 legs
02 · 3 legs
03 · 4 legs
Bets where the model's probability beats what the odds imply (EV+).
| Bookmaker | 1 · Brentford | X | 2 · Chelsea |
|---|---|---|---|
| Betsafe | 2.62 | EV+ 2% 3.65 | 2.50 |
| Betsson | 2.63 | EV+ 3% 3.66 | 2.51 |
| Nordicbet | 2.62 | EV+ 2% 3.65 | 2.52 |
| Pinnacle | 2.66 | EV+ 5% 3.76 | 2.56 |
Odds updated 18 Sept, 10:36
2.74 → 2.66
3.71 → 3.76
2.39 → 2.56
Pinnacle · 22 recorded price levels · 11/09/2026 → 18/09/2026 · fixed scale 5 percentage points
Head-to-head based on Premier League data since 2022.
Solid line = created (xG), dashed = conceded (xGA).
Last 5 matches. Circle size = pass volume, line width = combinations between players. Attacking left to right.
The same network weighted by offensive threat (xT): circle size = threat created by the player, line width = threat of the combinations. Last 5 matches, attacking left to right.
| Performance | Brentford | Chelsea |
|---|---|---|
| Points | 6 | 7 |
| xPoints | 5.9 | 6 |
| xG per match | 2.1 | 1.6 |
| xGA per match | 1.4 | 1.6 |
| xG within 8s of winning the ball | 0.65 | 0.28 |
| xGA within 8s of losing the ball | 0.18 | 0.34 |
| Playing style | Brentford | Chelsea |
|---|---|---|
| Build-up efficiency | 0.29 | 0.33 |
| Field tilt | 0.46 | 0.45 |
| xT per match | 1.6 | 0.82 |
| xTA per match | 0.8 | 1 |
| Won balls, offensive half | 34 | 26 |
| Pressing intensity | 0.24 | 0.24 |
| Pressing efficiency | 0.28 | 0.26 |
| Pressing efficiency, offensive half | 0.25 | 0.23 |
| Entries into the box per match | 16 | 15 |
| Entries into the box against | 14 | 17 |
| Pass completion % | 0.71 | 0.76 |
| Pass completion % under pressure | 0.67 | 0.74 |
| Passes per match | 292 | 322 |
| Passes against per match | 355 | 422 |
| Switches of play per match | 17.9 | 17.5 |
| Long balls per match | 28 | 25 |
| Set pieces | Brentford | Chelsea |
|---|---|---|
| xG from free kicks | 0.04 | 0.03 |
| Corners per match | 4.4 | 5.2 |
| Corners against per match | 3.2 | 5.5 |
| xG per corner | 0.09 | 0.04 |
| xGA per corner against | 0.06 | 0.01 |
| First touch, offensive corners % | 0.67 | 0.33 |
| First touch, defensive corners % | 0.69 | 0.55 |
| Other | Brentford | Chelsea |
|---|---|---|
| Throw-in control | 0.69 | 0.61 |
| The goalkeepers | Caoimhin Kelleher (Brentford) | Emiliano Martinez (Chelsea) |
|---|---|---|
| Saves | 15 | 15 |
| Save % | 79% | 71% |
| xG prevented | 34% | 69% |
| Claims | 0 (0%) | 0 (0%) |
The model gives Brentford a 36% win probability. Full 1X2 picture: Brentford 36%, draw 28%, Chelsea 36%.
The model's most likely final score is 1-1 at 12% probability.
The model rates over 2.5 goals at 53% and under at 47%.
The probability of both teams scoring is 57% according to the model.
The same rules as the bookmakers’ builders: legs that contradict or already follow from your slip cannot be combined — for example correct score 2–1 with under 2.5 goals. The difference is that we don’t look the rules up in a table; we compute them straight from the score matrix.
Predictions come from PlaymakerAI's match model and reflect the model's assessment — not betting advice. Gamble responsibly.