Nottingham Forest are 15th in the Premier League with five points from six matches, but the underlying numbers suggest their position flatters neither the team nor its critics.
Forest have won one, drawn two and lost three under Oliver Glasner, scoring five goals and conceding eight. Their recent sequence reads L, D, D, W, L, L, ending with last weekend’s 3-1 defeat at Crystal Palace.
According to Opta’s “expected” table, reported by the Nottingham Post, Forest’s chances and the chances they have allowed should have produced around 10.2 points so far, enough for third place behind Arsenal and Manchester City. Opta puts their expected goals at 10.3 for and 7.2 against, a positive difference of 3.1, against an actual goal difference of minus three.
The gap is starkest in the most recent match. At Selhurst Park, Forest took the lead through Dan Ndoye in the third minute and dominated much of the first half before Palace scored three times. Opta’s figures show Forest recorded 24 shots and 3.34 expected goals that afternoon, a total that would normally be expected to produce more than a single goal.
Expected goals measures the quality of chances rather than the outcome, assigning each shot a probability of scoring based on historical data. A team that consistently creates more than it concedes would normally expect results to follow over a season. The difficulty for Forest is that six games is a small sample, and some of the shortfall may reflect finishing quality rather than luck.
Glasner has made the same argument from the opposite direction. Speaking to a small group of reporters recently, he said Forest’s seventh-place finish two seasons ago had been a statistical overperformance, with the club’s data pointing to a position nearer 14th, and that teams relying on that kind of fortune eventually pay for it. He told owner Evangelos Marinakis that his previous teams had taken three to six months to adapt to his methods, and has repeatedly asked for patience.
The forecasting models are less optimistic than the expected table. Opta’s projection for the rest of the season, which blends results to date with betting market odds and its own power rankings, has Forest finishing 17th on 43.57 points, with a 26.88 per cent chance of relegation, according to the Post. That is a marked deterioration since the summer. Before the season, Opta gave Forest an 18.1 per cent chance of going down and a 13.7 per cent chance of a top-five finish, identifying them as possible outsiders under Glasner.
By the end of the transfer window, with one point from two games, the supercomputer had already moved Forest’s relegation probability up to 25.54 per cent, projecting 45.44 points. Not every model is as gloomy. One independent forecaster, StatsUltra, rates Forest’s relegation risk at about 7 per cent and projects 46 points.
The reason for the divergence is partly methodological. Expected goals tables describe what has happened on the pitch; season projections must also weigh the strength of the squad, the remaining fixtures and how the betting market rates the team. A poor run of results feeds directly into the latter even when performances are encouraging.
Forest’s immediate fixture list will test both readings. Arsenal, the reigning champions and second in the table, visit the City Ground on Sunday. Glasner will be without the striker Igor Jesus, who has a calf injury, and possibly the defender Nikola Milenkovic.
The club has reason to want the data to hold. Last season brought a change of manager and a run of results that left Forest closer to the relegation places than their supporters had hoped. If the expected numbers are a fair guide, a team creating chances at this rate should start converting them. The table will show whether that happens before the gap between performance and points becomes a problem of its own.
