How Arsenal used research into ref behaviour to rise up the league

By Tony Attwood

The last post here was Arsenal’s yellow cards dropped by 45% last season. How? Why? And how did it help?

That was the second statistical bombshell of last season – the first being that a club like Arsenal could slip down to 15th in the league by Christmas and be looking like relegation candidates, only to put in a set of performances in the second two-thirds of the season, and start the climb back up.

So the next question (put in its polite form) had to be, “What on earth is going on?”   How could a team be 15th in the league in the first third of the season and 2nd in the league through the last two thirds of the season?

To answer that we have to go back one more season – and yes there is going to be some more crowing because the only place you will have seen a report on the origins of the weirdest of all the footballing affairs we’ve researched and reported, is on Untold.  

And that’s not because we have a team of investigative journalists and those so-called “super computers” that the media quote each year to tell us who is going to win the league.  It’s just, when we see something that doesn’t make sense, and which no one is mentioning, it makes us do some work.  Quite often the results are not very spectacular but I think this is perhaps the best one we’ve done in our 13 years on line.

I touched on this in the last piece: “How a club can commit the most fouls but get the fewest yellow cards, but to save you reading all that, here’s the key point…

We found that two seasons ago if Burnley tackled 13 times they would on average commit 10 fouls.  That was about normal.  But then we noticed that Leicester, would need to undertake 21 tackles to give away 10 fouls.   That is an astonishing difference.  

Then when we looked at the number of fouls that led to yellow cards, the situation was even more extreme.  Arsenal were more than twice as likely to be penalised with a card each time they tackled, than Leicester.

The explanation given by “fans” (I use the word lightly) was that this was down to Xhaka, who was a menace on the pitch.  But that did not make any sense; he only got five yellows that season, and was not even in the top 20 list of players for yellow cards),  

Leicester’s figures were in fact totally out of line with every other club.  Which raised the questions a) how were they doing this, and b) why wasn’t everyone else following this approach?   

You can see the whole table in the earlier article but here I want to highlight Arsenal and Leicester.  Arsenal were mid-range tacklers that season, Leicester were tackling more than any other club.  Arsenal were also mid-range foulers, but Leicester were near the bottom.  Arsenal got a yellow every 4.32 fouls, more than any other club.  Leicester got one every 9.48 fouls.

Club Tackles Fouls Tackles/foul Fouls/game Yellow Fouls/yellow
Arsenal 417 259 1.61 10.79 60 4.32
Leicester 494 237 2.08 9.88 25 9.48

So Arsenal put in far fewer tackles, committed a few more fouls (generally about one per game more than Leicester) yet ended up with 35 more yellow cards!!!!    

The article in which we discussed this led to some Leicester fans saying that Leicester achieved their low yellow card rate by having the best tacklers in the League.

Our article appeared in February 2020, and then something odd happened.  Leicester’s figures changed dramatically, and their fouling level took off.  Timidly we suggested that PGMO realised their favouritism of Leicester had been rumbled.  Leicester fans took offence and said the change was because of injury to their top tackler.   We checked, but no – the dates didn’t match at all.

So we changed our commentary.  Before this we had been saying “Arsenal are treated differently from other clubs by referees”.  Now we were saying, “there is a way out of this referee bias.  Cut the tackling.”

It would of course be ludicrous to think that a blog run by a handful of fans who quite like numbers could influence anything.  So it must have been coincidence, but after we published the table showing Leicester could commit twice as many fouls as Arsenal before getting a yellow, Leicester’s numbers did change: the refs started penalising their tackling pretty much as they did for most other teams.

So we then suggested maybe Arsenal could change too.  Our view was that Arsenal were always going to be penalised by refs for their tackles possibly because PGMO refs have the subconscious views of each team’s approach and treat them accordingly.  (That would be an influence akin to that of the home team’s fans on the ref – but one that is going to be much harder to prove academically).  However, if right Arsenal needed to do just one thing.  Cut out the tackles.

And it must have been a coincidence, as I am sure we are not influencing Arsenal,  but that is what they did.

Tackles 2019/20 Tackles 2020/21 Change Fouls 2019/20 Fouls 2020/21 Change Yellow 2019/20 Yellow 2020/21 Change
584 456 -22% 421 345 -18% 86 47 -45%

All we did was do the research and publish stats which for some reason no one else ever published.   And then we commented.   Like we do.  

Thanks for reading, and to all the readers who have stayed with us through this long and tedious set of analyses over the last two seasons, thanks for your patience.  And guess what, there’s a bit more to come.

The four dangers that could stop Arsenal’s return to greatness


8 Replies to “How Arsenal used research into ref behaviour to rise up the league”

  1. Again an interesting analysis, which does however hover around the ‘confirmation bias’ syndrome.

    You see, one of the matters that make me really wild, is the frustrating reality that some referees award fouls against Arsenal for the most incipient reasons, when compared to the opposition mirroring the same ‘fouls’ that were not given.

    Contrarily, more definite fouls were awarded with an accompanying yellow card against Arsenal, whereas an equally hard tackle awarded against the opposition frequently did not get a yellow card.

    In summary, if as you garnered, Arsenal and, say Leicester, had roughly equal fouls given, the award of more yellow cards against Arsenal would seem to be an embedded mindset that a number of referees had against us.

    The problem for me is that innocuous fouls that were given against Arsenal, were not given against the opposition, and when more definite fouls by either team were given, they resulted in stronger punishments (yellow cards) being given against Arsenal.

    Referees, if we could question them, would simply say the above was my opinion, and their opinion was different. As they have the whistle and the authority, I do not see how Arsenal could ever overcome the mindset I mentioned above.

  2. Roger That I am completely with you in what you say. I have tried to argue elsewhere that referees do have a view that Team A is a ball-paying team of artists who don’t do nasty tackling, while Team B must be watched at all times for their dirty tricks.

    I do agree that was embedded in the mind-set of the referees, and it was Arteta who finally searched for a way to sort it out, and that by reducing the tackling, which is where the vast majority of punishments came from.

    I think by doing this, he is slowly taking the focus off giving Arsenal yellow cards, and slowly over time the mindset of the referees may change. Certainly if, next season, the home / away differential returns to the norm, that will be further proof for the theory that referees are biased by the home crowd. So it will be up to all Arsenal fans in the ground to do their bit.

  3. Although our yellow card figures are reduced, referees appear to be still more inclined to give them to Arsenal than to their opponents. I don’t recall any Arsenal player receiving a verbal warning from a referee before a yellow card is awarded, whereas it is a common occurrence with our opponents. Also, referees seem to be reluctant to award second yellows or straight red cards to our opponents, even when the (usually hostile ) commentators suggest that they might be warranted.

    For example, in the recent match at Selhurst Park, it would have been quite fair for up to three red cards to be awarded to Palace players and the fact that none were given is surely clear evidence of an uneven interpretation of the rules – especially in the case of Benteke’s assault on Elneny, which was excused by pundits (other than the majority who ignored is completely,) as being “not sufficiently violent” to warrant a sending-off. Presumably, Elneny received a yellow card in the same incident for not taking evasive action and therefore being provocative. Contrast this event with the sending-off of Pepe at Leeds earlier in the season, for only a head-butting “gesture” towards Alioski.

  4. John L
    Similarly with the Xhaka throat grabbing for which he received his marching orders.
    I have seen two far worse than that, one I think by Chilwell and can’t remember who did the other, but both resulting in only receiving yellow cards.

  5. When an Arsenal player is deemed to have threatened violence it is a red card. The League cup match a few years ago when Adebayor was deemed to have punched a Chelsea player from outer space. It is truly amusing how these abysmal examples of FIFA badged stooges have been retained by the FA for so many years.

  6. Arsenal are still the only team in the PL to have had a player carded for placing the ball in a sarcastic manner.

  7. Arsenal players walk a very tightrope as per the PIGMOB , but as Tony says , have learned to avoid the cards by staying well away from trouble ; or what is perceived as being trouble . Well done to MA and his coaches in steering us right.

Comments are closed.