By Tony Attwood
In a piece in The Athletic there is a rather interesting sentence: “Since the inaugural 20-team Premier League season in 1995-96, the vast majority of teams (38 percent) have shifted by just one place — or stayed in the same position — when comparing match week 10 to the final league standings.”
Now that obviously immediately raises both alarm bells and questions. Such as how can 38 percent be the “vast majority”, and what actually happened last season.
To consider this further here is the league table after ten games in the last campaign. The extra column on the right shows where those top four clubs ended up.
2022/23 (League tables from 11v11)
| Team | P | W | D | L | F | A | GD | Pts | Final | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Arsenal | 10 | 9 | 0 | 1 | 24 | 10 | 14 | 27 | 2 |
| 2 | Manchester City | 10 | 7 | 2 | 1 | 33 | 10 | 23 | 23 | 1 |
| 3 | Tottenham Hotspur | 11 | 7 | 2 | 2 | 22 | 12 | 10 | 23 | 8 |
| 4 | Chelsea | 10 | 6 | 2 | 2 | 15 | 10 | 5 | 20 | 12 |
So this suggests something quite different. It seems that on 19 October 2022 Chelsea were fourth in the league but ended up 12th, thus suggesting that the league table after 10 (or 11 in the case of Tottenham) doesn’t really tell us too much at all.
A little later they refer to the 2019 study that looked at match data from 1995 to 2017. This concluded that “all soccer leagues appear to conform to mathematical laws, which constrain the league standings as the season progresses. This means that partial standings can be used to predict end-of-season league position with reasonable accuracy.”
Now “reasonable accuracy” is the key here. And that’s important because they then talk about “a high likelihood of a return to the Champions League next season after just one-quarter of the season” for Tottenham. So clearly we are NOT talking about the positions being fixed but the possibility of the team top after ten games, ending up in the top four.
And yet despite this, they insist on saying that “looking at the league table as early as November does give a decent indication towards the final positions held by each team,” when quite clearly last season this was absolutely not the case.
So let us look at 10 games gone in 2021/22. The top four were…
| Team | P | W | D | L | F | A | GD | Pts | Final | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Chelsea | 10 | 8 | 1 | 1 | 26 | 3 | 23 | 25 | 3 |
| 2 | Liverpool | 10 | 6 | 4 | 0 | 29 | 8 | 21 | 22 | 2 |
| 3 | Manchester City | 10 | 6 | 2 | 2 | 20 | 6 | 14 | 20 | 1 |
| 4 | West Ham United | 10 | 6 | 2 | 2 | 20 | 11 | 9 | 20 | 7 |
OK, three of the top four were already in the top four after ten games, but not in the right order. And again there was a team that fell away – West Ham United in this case.
Let’s try again: 2020/21
| Team | P | W | D | L | F | A | GD | Pts | Final | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Tottenham Hotspur | 10 | 6 | 3 | 1 | 21 | 9 | 12 | 21 | 7 |
| 2 | Liverpool | 10 | 6 | 3 | 1 | 22 | 17 | 5 | 21 | 3 |
| 3 | Chelsea | 10 | 5 | 4 | 1 | 22 | 10 | 12 | 19 | 4 |
| 4 | Leicester City | 10 | 6 | 0 | 4 | 19 | 14 | 5 | 18 | 5 |
So quite clearly where clubs are after ten games might – but might not – give us a clue where they will be after 38 games.
We’ll do one more 2019/20 – and there finally we have found a league table after 10 games which does indeed get close to reflecting the final table.
| Team | P | W | D | L | F | A | GD | Pts | Final | |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Liverpool | 10 | 9 | 1 | 0 | 23 | 8 | 15 | 28 | 1 |
| 2 | Manchester City | 10 | 7 | 1 | 2 | 32 | 9 | 23 | 22 | 2 |
| 3 | Leicester City | 10 | 6 | 2 | 2 | 25 | 8 | 17 | 20 | 5 |
| 4 | Chelsea | 10 | 6 | 2 | 2 | 23 | 16 | 7 | 20 | 4 |
So the conclusion is that the top four after ten games can on occasion be a 75% reflection of the final table at the end of the season simply in terms of ending up in the top four or not. But it can also include clubs that end up much further down the league.
Indeed we can consider these four tables above showing the position of the top four after 10 games. So that has 16 clubs shown. 10 of these were in the top four at the end of the season. That is a 62.5% chance. Better than evens, but not that much better.
Nine of the listings were within one place of where the club would actually end up. That is 56%, or just a little bit over half. Four out of 16 (25%) were over three places out.
The biggest drops from a top-four place after ten games to somewhere much lower at the end of the season were Tottenham from third to eighth, Chelsea from fourth to twelfth, and Tottenham from first to seventh. And perhaps it is slightly interesting that Tottenham have done a big drop twice.
The conclusion has to be that being in the top four at this moment in a campaign is better than not being in the top four, but is no guarantee of subsequent success.
