How it works
Every club gets a rating from its expected goals. Each game left becomes odds for every scoreline. Then the rest of the season is played out thousands of times, and we count where Aberdeen finish.
The examples below use the latest data, so they change as the season goes on.
Chances made and chances given away
Goals are noisy: a deflection, a wonder strike or a missed sitter can swing a game. Expected goals (xG) measures the quality of the chances instead, so it tells you more about how good a team really is. Each club gets two numbers from FotMob's xG: an attack rating (chances it creates per game) and a defence rating (chances it allows).
A handful of games is a small sample, so a club's numbers this season are never used on their own. Every club starts from last season's rating, and that starting point counts as 12 games. As the season goes on, the new games outweigh it. A club just promoted starts as a typical promoted side: a little below average in attack, well below in defence.
The scoreline odds
For any game, a team's expected goals = its attack rating × how much the opponent allows, scaled to the league average. The home side gets a 10% boost and the away side a 10% cut. Take Aberdeen's next game, at home to Hibernian on 7 Nov: that gives Aberdeen 1.35 expected goals and Hibernian 1.49.
Goals follow a well-known pattern (the Poisson distribution), so those two numbers give the chance of each side scoring 0, 1, 2, 3 and so on. Put the two sides together and you get the chance of every scoreline. Add up the scorelines where Aberdeen win, and that is their win chance.
| Hibernian goals → | |||||
|---|---|---|---|---|---|
| Aberdeen ↓ | 0 | 1 | 2 | 3 | 4 |
| 0 | 5.8% | 8.7% | 6.5% | 3.2% | 1.2% |
| 1 | 7.9% | 12% | 8.8% | 4.3% | 1.6% |
| 2 | 5.3% | 7.9% | 5.9% | 2.9% | 1.1% |
| 3 | 2.4% | 3.6% | 2.7% | 1.3% | 0.5% |
| 4 | 0.8% | 1.2% | 0.9% | 0.4% | 0.2% |
Every game left, thousands of times
The predictor takes every league game left before the split and plays each one, drawing a score at those odds, and adds up the table. Then it does it again, thousands of times. Each run is a season that could happen; together they show which ones are likely.
Right now, across those seasons Aberdeen average 37 points after 33 games, usually somewhere between 30 and 45. Sixth place after 33 games has averaged 44 points since 2010-11. Aberdeen got there in 13% of the runs.
Tested against the bookmakers
The bookmakers' odds are the best public forecast there is: they price in team news, injuries and transfers as well as form. So we replayed the predictor on all 981 Premiership games from 2022-23 to 2026-27, using only what it would have known before each kick-off, and scored it and the bookmakers against what actually happened.
The chart below shows how close the predictor gets. 0% would be no better than guessing from the league's usual home, draw and away rates; 100% would be as good as the bookmakers.
It gets most of the way to the bookmakers at every stage of the season, including the opening weeks, because each club starts from last season's rating. It doesn't beat them: they know about team news and transfers, and it only sees xG. So the predictor isn't a betting tool. It is an open way to ask what the rest of the season could look like.
When it says 30%, does it happen 30% of the time?
Both sit close to the line: when the predictor says something has a one-in-four chance, it happens about one time in four. Its draw chances (around a quarter of games) match how often Scottish games are drawn.
The limits
- Injuries, suspensions, a new manager or a January signing. It only sees xG, so it catches up as the games come in.
- Finishing. A club that keeps scoring more than its chances suggest is rated on the chances, not the goals.
- The home edge is an assumption (10% either way), not measured club by club.
- Goal difference. Clubs level on points are split at random.
- After the split. It stops at 33 games, because the last five fixtures aren't set until then.
The points total is the number to trust most. Finishing positions depend on every other club too, so treat the top-six chance as a rough guide, especially early in the season.