There is an understanding that a league table does not truly “take shape” until clubs have played 10 of their allotted matches in that season’s competition.
It is an ancient and arbitrary threshold we have created for ourselves, but it has merit.
First, it is a nice round number. Second, it’s… double figures.
It’s also 26.3158 per cent of the way through a 38-match Premier League season, which makes it feel odd to infer much about the final positions each team will occupy when the music stops in late May.
We wouldn’t take this approach in other walks of life.
If you walked out of a basketball game after the first quarter, say, would you be certain of the winning team? If you watched 26 per cent of a film at the cinema and then went home, could you reliably predict its ending?
There will be plenty more twists and turns in the 2024-25 season between now and those 10 full-time whistles up and down the land on Sunday, May 25.
However, broadly speaking, there is statistical evidence to suggest that the pack is unlikely to shuffle dramatically from the current standings over the next seven months. Since the inaugural 20-team Premier League season in 1995-96, more than a third of teams (38 per cent) have shifted by just one place — or stayed in the same position — when comparing matchweek 10 to the final standings.
Maybe we innately trust the league table after 10 games because, deep down, we know just how much it reveals about what lies ahead.
With nine games played each, the above league standings could have been predicted back in the summer.
We can talk about some clubs having easier opening sets of fixtures than others, but last season’s final top four are already back in their familiar Champions League spots. At the other end, Leicester City’s recent results have seen them push clear of the relegation zone but fellow promoted sides Ipswich Town and Southampton are two of the bottom three. The promoted sides propped up the league after 10 games of last season — and all three were relegated.
Leicester began the 2000-01 season well, but it did not last – they finished 13th (David Jones – PA Images/PA Images via Getty Images)
Every fan wants to keep their respective dreams alive for the whole season, but what does the data say on how much the table differs over time?
A 2019 study looked at match data from 1995 to 2017 and found that the team who topped the Premier League after matchweek 10 had a 77.3 per cent chance of finishing in the top three. So, Manchester City, Liverpool, and Arsenal supporters, you might as well start planning those 2025-26 midweek European trips now.
The notable outlier from last season were Tottenham Hotspur. Ange Postecoglou’s men sat at the summit with eight wins and two draws after 10 games, but that near-perfect record did not last. They lost three successive games in November and ended up in fifth place.
Neverthless, a closer look at the findings from that 2019 study highlighted just how stable the wider league table becomes after 10 matchweeks.
In the Premier League, the researchers found that 77 per cent of the statistical variance in the final league standings was explained by matchweek 10. By matchweek 20, that rose to 87 per cent. And after matchweek 30, it was 94 per cent.
In other words, looking at the league table as early as November gives a decent indication of the final positions each club will hold.
We can look at this ourselves by analysing the relationship between the partial standings after every matchweek with the final league positions during the 2023-24 season.
For the data folk who are interested, this is calculated using a correlation analysis (Spearman’s rank), which provides a value (between -1 and +1) on the strength of the relationship between the sets of data points.
A higher ‘r-value’ denotes a stronger relationship.
Tracking this across each matchweek, we can see the strength of the relationship increase — but note how stable things look after just a quarter of the campaign.
Let’s be clear, the smaller changes in league positions can be the most crucial. The difference between a team finishing in first or second is not significant mathematically but immeasurable in its significance. We should still get the final-day drama of title races, European pushes and relegation battles, but the wider analysis of league standings shows that a decent projection can be made early on.
The implications of this are interesting.
While it might seem like club owners appear a little overzealous when parting company with struggling managers before the Christmas period, the data suggests things are unlikely to change dramatically in the ensuing weeks and months.
We have come to accept the chaotic desperation that oozes out of clubs when sacking a manager, but the message is simple — there is mathematical merit in taking action sooner rather than later.
While the table’s configuration shuffles less than some might think, there will continue to be memorable examples of teams who experience a significant drop — or a significant uplift — after 10 games.
Looking across the Premier League era, a few seasons stand out.
Last season, Tottenham’s slump mentioned above was the most notable at the top end of the table, but there was also a notable swing in another part of London.
Brentford were sitting in mid-table before a drop in form saw them settle for a disappointing 16th-placed finish. After matchweek 10, neighbours Chelsea sat just below them, but a strong end to the season saw Mauricio Pochettino’s side climb five places to qualify for the UEFA Conference League.
For those with longer memories, cast your mind back to 2008-09, when promoted Hull City won six of their opening nine games — including away against Arsenal and Spurs — to find themselves ‘pushing for a European spot’… only to win just two of their remaining 29 and scrape survival, finishing a point above the relegation zone in 17th.
Meanwhile, Tottenham were bottom after matchweek 10, following the worst start in their history (two points from eight games). Harry Redknapp replaced Juande Ramos as manager in late October and pushed them to an eighth-placed finish — the biggest shift in the Premier League era.
The fact is, you probably do remember these examples because they are so unusual.
Here’s another, from 2005-06.
Charlton Athletic fans will not need reminding of their Darren Bent-inspired early ‘title push’ that saw them win their opening four league games, only to drop down to a more realistic 13th position after the final round of matches the following May. It is a nosedive that has only been matched by Hull’s 2008-09 in the 32-year Premier League era, with no team falling more than 11 places.
Similarly, a pre-Sheikh Mansour Manchester City looked to be on for a strong top-half finish in autumn 2005, before Stuart Pearce’s side fell eight spots to 15th by the end of the season.
These fun quirks of yesteryear highlight the caveats that must be considered in the wider context of a season’s final placings.
The Athletic has previously analysed the impact of fixture difficulty in skewing the table in the early part of a season and a favourable run of games can look more pronounced than a squad’s quality might suggest. Also, a team’s over- or under-performance can similarly provide a false perception of true performance. Across a full season, you would expect — although not guarantee — any quirks from the opening few fixtures to even themselves out.
Nevertheless, the numbers don’t lie. While teams are yet to even play all 19 opponents once, evidence suggests that a large part of the league table variance can be explained after just a quarter of the season.
Which means 14th-placed Manchester United, among others, have plenty of work to do.
(Top photos: Getty Images; design: Eamonn Dalton)
Read the full article here