By Tony Attwood
‘Time is up’ for Mikel Arteta at Arsenal after Burnley defeat …
So said Talk Sport on 14 December 2020. Fortunately, as is the case most of the time, they were totally wrong.
Indeed so bad at predicting the future is most of the media, it is a wonder that the scribblers and talkers are not so embarassed by their inccuracy that they don’t stop.
But the question remains, is there a way to predict, accurately, the outcome of football matches?
If you are a regular reader of Untold you will know that we like statistics, and do find that they give quite a good indication of the way a match might go. Indeed recently the Athletic (behind a paywall) recently tried copying our technique of looking at the way the season has been going and comparing it with past seasons. (A method that tends to predict that yes Arsenal will win the league. See for example What normally happens to the club that is top after 15 games of a season?)
However many other commentators seem to be much keener on quoting possession in terms of football matches. In fact if there is one figure that is going to be quoted concerning a game it is likely to be possessioin. But what no one asks is, “does possession tell us very much about anything, other than the amount of possession a club has in one game?”
The table below is in the order of possession percentage ranging from Manchester City who have two thirds possession across their matches this season, tp Nottingham Forest that has just over one third possession.
And yes the top five clubs in terms of possession occupy places in the top half of the league, although only just. Chelsea are tenth in the league but third in the possession table.
And it turns out that although there is some relationship between league position and possession, we should note that Newcastle are third in the league but ninth in terms of possession.
So in general terms, teams that have more possession are doing better than teams that have less. But Brentford who are ninth in the league are 18th in the possession table. It turns out to be not a very profound statistic in terms of making us aware of any underlying trend
But what about goals? Do the teams that score the most goals have the most possession?
Again there is a tendency for goal scoring team to be high on the possession charts, but again there are exceptions. Chelsea for example are third in terms of possession but 12th in terms of goals scored.
And Wolverhampton who are 19th in the league, and bottom in terms of goals scored, are 11th in terms of possession. Brentford who are ninth in the league and seventh in terms of goals scored are 18th in terms of possession. Here’s the table…
| Team in possession order | Possession% | Goals | Goals Pos | Lge Pos |
|---|---|---|---|---|
| 1. Manchester City | 66.2% | 45 | 1 | 2 |
| 2. Liverpool | 61.5% | 34 | 4 | 6 |
| 3. Chelsea | 58.7% | 20 | 12 | 10 |
| 4. Arsenal | 57.6% | 40 | 2 | 1 |
| 5. Brighton | 56.6% | 32 | 5 | 8 |
| 6. Manchester United | 53.2% | 27 | 9 | 4 |
| 7. Tottenham | 51.3% | 37 | 3 | 5 |
| 8. Leicester | 51.2% | 26 | 10 | 13 |
| 9. Newcastle | 49.6% | 32 | 5 | 3 |
| 10. Leeds | 49.4% | 25 | 11 | 14 |
| 11. Wolverhampton | 48.5% | 11 | 20 | 19 |
| 12. Aston Villa | 47.5% | 20 | 12 | 11 |
| 13. Fulham | 47.3% | 30 | 7 | 7 |
| 14. Crystal Palace | 46.4% | 17 | 15 | 12 |
| 15. West Ham | 44.9% | 15 | 16 | 17 |
| 16. Everton | 44.5% | 14 | 18 | 18 |
| 17. Southampton | 44.1% | 15 | 16 | 20 |
| 18. Brentford | 42.5% | 30 | 7 | 9 |
| 19. Bournemouth | 39.4% | 18 | 13 | 16 |
| 20. Nottingham Forest | 38.8% | 13 | 19 | 15 |
And this reveals one of the big problems with the way that football is reported in England: the reporting is full of assumptions. Generally speaking, possession of the ball is good – and the more of it the better. But that’s it.
As we might expect, goal scoring shows a closer link with league position… This table is ordered by the number of goals scored.
| lge position | Team | Goals scored | Goals scored position |
|---|---|---|---|
| 2 | Manchester City | 45 | 1 |
| 1 | Arsenal | 40 | 2 |
| 5 | Tottenham Hotspur | 37 | 3 |
| 6 | Liverpool | 34 | 4 |
| 3 | Newcastle United | 32 | 5 |
| 8 | Brighton and Hove Albion | 32 | 6 |
| 7 | Fulham | 30 | 7 |
| 9 | Brentford | 30 | 8 |
| 4 | Manchester United | 27 | 9 |
The biggest difference here is Manchester United who are fourth in the league but ninth in the goals scored table.
As for goals conceded, that shows an even bigger variation between that metric and league position
| Lge pos | Team | Goals against | Goals against pos |
|---|---|---|---|
| 3 | Newcastle United | 11 | 1 |
| 1 | Arsenal | 14 | 2 |
| 2 | Manchester City | 16 | 3 |
| 10 | Chelsea | 19 | 4 |
| 4 | Manchester United | 20 | 5 |
| 6 | Liverpool | 22 | 6 |
| 17 | West Ham United | 24 | 7 |
| 18 | Everton | 24 | 7 |
| 5 | Tottenham Hotspur | 25 | 9 |
| 8 | Brighton and Hove Albion | 25 | 9 |
Everton and West Ham are keeping the ball out of their net much of the time but they haven’t got the balance at the other end of the pitch.
What the tables confirm is that the best way to go up the league is to score goals – which is pretty obvious, but it does show where clubs like Everton and West Ham are going wrong. They have a solid defence but clearly they are not attacking well enough to make use of that fact.
Overall, the much-lauded possession figure is the least relevant in terms of success, despite the regular howling at players who “give the ball away”. If a club is wanting to improve, the best thing it can do is score the goals.
We’ll look at how best to do that in a later piece.
Darren Bent must be a fully paid-up member of the AAA and AOB.
When you score 1 point on your specialist subject for Celebrity Mastermind, it’s probably best to keep quiet.