By Tony Attwood
The correspondent known as “Ukp” recently wrote to Untold, commenting on our use of the “tackle/foul/card stats” noting that this is “where you analyse numbers of fouls Vs number of cards and deduce ref bias, when everybody knows cards are a product of quality and not quantity of fouls.”
I am not at all sure that our view of referee bias comes just from the issue of fouls vs the number of cards, but leaving that aside I wondered if the assertion that cards are a product of quality not the number of fouls, is true.
It was a question of some interest because the same correspondent also questioned the validity of our series in which TV evidence was produced to examine if there was referee bias. I can’t really examine that point in any detail because no evidence was given to suggest why and where the analyses were wrong. It was merely stated that they were.
Now the correspondent Ukp went on to say, “As for Debs and other people not giving evidence, that also seems like personal opinion, all you need to do to find out is as she said “google”. This time type in ‘Debs untold’ and check out all her previous comments, there’s enough of them with reliable stats.”
I tried that, but it didn’t work for me. And so I wondered if the correspondent had actually done this him/herself – or was it just an assertion? Which then made me go back to, “Everyone knows that cards are a product of quality and not quantity of fouls”.
Of course that isn’t literally true because obviously not “everyone knows” – but if one turns the comment into “most people who look at these things know that cards are a product of quality and not the quantity of fouls” then obviously the statistics would certainly reveal this to be so. And since I am responsible for those stats I thought I would take a look.
So I decided to compare how clubs perform in terms of tackles, fouls and yellow cards in relation to the quantity. I did this because it is hard to measure the quality of fouls objectively but quantity is easy to measure since the statistics are provided each week by the Premier League.
If the assertion that “cards are a product of quality and not quantity of fouls” is true, there would be no significant relationship between the position of a club in the fouls table and the position of the club in the yellow card table; the numbers should be random.
In the comment I am questioning there was no mention of the position of the club in the tackle table, it just referred to fouls and cards, but I have included the position in the tackle table below, just in case there is a relationship there.
To save time I’ve picked out a range of clubs from across the league, and if we look at line one below we immediately see a relationship. Arsenal are 18th in the table made up of the number of tackles, 17th in the table in terms of the number of fouls and 16th in the table of the number of yellow cards. That seems to be a relationship contradicting Ukp.
Moving on, Brighton come second, fourth and fourth, which again suggests a relationship. Indeed scouring the table we see that Leeds are first in all three categories, most certainly suggesting a relationship between the number of tackles, the number of fouls and the number of yellow cards – exactly as Untold suggested but as the correspondent utterly denied.
True, the relationship does break down on occasion, most particularly with Manchester United and above all Leicester, and indeed it was Leicester’s strange figures which led us to start this investigation a couple of years ago where indeed they were putting in the tackles but getting few fouls called and few cards given.
Here’s the table.
| Club | Pos in tackle table | Pos in Foul table | Pos in yellow table |
| Arsenal | 18 | 17 | 16 |
| Brighton | 2 | 4 | 4 |
| Chelsea | 16 | 11 | 14 |
| Leeds | 1 | 1 | 1 |
| Leicester | 7 | 19 | 17 |
| Liverpool | 19 | 14 | 18 |
| Manchester C | 20 | 20 | 20 |
| Manchester U | 14 | 5 | 5 |
| Newcastle | 6 | 9 | 2 |
| Tottenham | 10 | 6 | 7 |
| West Ham | 17 | 18 | 19 |
And in case you are interested in the original numbers here they are
| Club | Tackles | Fouls | Yellow |
| Arsenal | 445 | 295 | 46 |
| Brighton | 573 | 340 | 65 |
| Chelsea | 478 | 324 | 48 |
| Leeds | 665 | 389 | 91 |
| Leicester | 546 | 272 | 42 |
| Liverpool | 436 | 305 | 40 |
| Manchester C | 397 | 270 | 36 |
| Manchester U | 493 | 333 | 62 |
| Newcastle | 548 | 327 | 72 |
| Tottenham | 514 | 331 | 57 |
| West Ham | 472 | 282 | 40 |
So there certainly seems to be evidence that the number of fouls called by the referee has a relationship with the number of tackles and the number of yellow cards, which is what I would expect but which the correspondent not only denied but claimed that “everybody knows”.
Debs however argued that not only has Untold said there was a relationship when there was none, but has also suggested that from this we deduced referee bias. I am again not at all sure this is so. There are a number of articles noting Leicester’s strange tackle, foul and yellow card figures over the years – Leicester’s strange tackle / foul / yellow figures change as they slip down the league is a good place to start if you are interested, but I think the overall conclusion was one of puzzlement as to how and why Leicester’s figures are as they are. Referee bias was one possibility but in subsequent articles we went on to explore the notion that expectation and club familiarity were possibly to blame, reinforcing the view that having referees only oversee each club twice in a season could be a very helpful move.
A Leciester fan wrote in and said that the unusual Leicester tackle / foul / yellow figures were because of the brilliance of the club’s defenders, and that the club’s figures only began to slip (after Untold highlighted them) because the player in question was then injured, but on checking the dates we didn’t find they matched.
It was, I felt, also interesting that we spotted another anomaly in the way Leicester got penalties at a record level subsequent to their tackle / foul / yellow ratios being questioned. Again after we highlighted this the figures began to change radically, and Leicester stopped getting nearly so many penalties. The article “Leicester heading for all time record number of penalties” is perhaps of some interest here.
I am asked sometimes why we allow people like Debs to comment with these wild and inaccurate allegations on Untold, and the answer is, because these commentaries do represent quite a widespread view among a group of supporters and I think we need to recognise this. But a little bit of work can show the allegations are wrong, and that, I think, is always helpful. Comparing allegations to facts always seems to me to be quite helpful.
I think this is a tacky interesting one. Whilst there is a relationship between the number of fouls and the number of yellows, one bad foul equals one yellow (usually). As you say, hard to measure. Rotational fouling, where a team sets out to foul particular players, but not badly enough to get a card, and not by the same player so that they don’t get a card, is where this can go weird in the stats (I would think).
Where low level fouls are continually being used to break up play, or rotational fouling is being used against a particular player there is a simple solution which could be borrowed from rugby whereby after a series of fouls by the same team, not player, the referee calls the captain over and tells him that the next foul of any description by any player will be punished with a spell in the sin bin.
We do not have a sin bin in soccer so a yellow card would be the equivalent. This would give a player like Saka who is being targeted some protection from the regular fouls he is subjected to.
After say Saka being fouled four times by the opposition the ref would warn the captain the next time Saka is fouled irrespective of by whom will result in a yellow card being issued to the culprit. Surely that would help eradicate all these niggly low level fouls deliberately designed to break up the flow of the game or intimidate a particularly dangerous player.
You have to look at the fouls committed overall by the whole team not individual players who are clever with implementing rotational fouling.
in our previous match against Southampton, there was atleast 2 fouls committed by S’oton players that’s a standard yellow card offense but they blatantly got away with it. definitely it depends on the players and clubs they playing for
Untold’s claim has been that referees are specifically biased against Arsenal. This article contradicts such an assertion and it even attaches statistics to support that. If yellow cards issued correspond to tackles and fouls, then where is the bias? From the look of it Arsenal are getting their fair share.
there’s a video i stumble couple weeks ago on yt. it shows compilation of serious fouls committed by opposition players on Arsenal players and got away with it. watching that really make my blood boil but being supporter of Arsenal football club it really don’t surprise me much. there is myth that PL referees like to let plays flow but sometimes, the type of fouls some players got away with is egregious. not saying Arsenal is the only team that was unfairly treated but there’re countless evidences and incidence that show PL referees is a disgrace and some players do get away lightly or treated differently. i’m pretty sure majority of other club supporters and fans agree with that
Do sites that record tackle/foul/card statistics record these correctly if the referee plays advantage? No foul would be awarded, but the card count could certainly be affected. I only ask because I see major inconsistencies with the application of the advantage rule for certain teams in the PL.
Hi Tony. You are wrong primarily because you have done a poor job in gathering your stats. Sample volume is quite small. Secondly you are wrong because you have selected mainly data that projects your agenda. I’ll try to correct these shortcomings in the data I’ll put up. The sample size will be more and at least it will cover the entirety of the last concluded season 2020/21 even though I accept that one seasons data is still too small, but still it is better than what you give us. Secondly data for all the clubs in the season will be shown. I won’t select the ones that make my point… So here we go
Using the the same criteria of total tackles attempted, fouls conceded and yellow cards, I’ll only give the rankings of the clubs in these areas in the 2020/21 season. You can check the full figures on the EPL website (for tackles and yellow cards) and footstats.com for number of fouls. The table below will give club, total tackles ranking, fouls ranking and yellow cards ranking In that order
1. Man city -19,19,19
2. Man utd -13,3,3
3. Iverpool -15,14,20
4. Chelsea -9, 6, 14
5. Leicester 3, 10, 5
6. Westham 16, 18, 16
7. Tottenham – 5, 4, 10
8. Arsenal -20, 20, 17
9. Leeds -1, 7, 5
10. Everton -4, 16, 8
11. A.villa -14, 5, 4
12. Newcastle -17, 15, 5
13. Wolves -11, 10, 11
14. C.palace -5, 12, 9
15. S’ton – 2, 8, 12
16. Brighton -8, 7, 18
17. Burnley -18, 17, 15
18. Fulham -10, 1, 2
19. Westbrom -12, 13, 13
20. Sheffield – 7, 2, 1
A few clubs had close to consistent figures e.g man city, Arsenal, westbrom. But the vast majority had very little if any correlation between tackles/fouls/cards
I think after studying the above data, the only valid conclusion to make is that the conclusion that cards are a product of tackle quantity is a product of untolds imagination. Also, we seem to have done well on all 3 stats. I attribute this to our good work and not a gift from the refs as some people would prefer to do.
My lords, I rest my case