By Tony Attwood
It was 31 December 2018 that we published Andrew Crawshaw’s groundbreaking article “How four clubs commit fouls with impunity, while others constantly get punished”
Since then we’ve been looking constantly at the issue of fouls and the way referees treat Arsenal, and so it was probably inevitable that we’d be the people to pick up on last season’s amazing development in which Arsenal sank down tackling table in order to stop getting ceaselessly penalised by referees throughout the games.
But now a new set of data has come to light which has made that discovery far more shocking, for it concerns which referee handles Arsenal in which way.
Our original story gave us an insight into the new tactical style adopted by Arsenal which coincided with the extraordinary improvement in results in the last two thirds of last season as Arsenal cut down on tackling, in order to reduce yellow cards – which dramatically improving their performances.
And of course by and large it has been ignored by the media who have treated the season as a unity, and fouls as something absolutely not to be mentioned.
But now we have another set of figures which suggest there is something even stranger that went on last season.
We have long since known that referees call out fouls by certain teams in different ways. But here’s something we missed before: the more likely a referee is to call fouls against Arsenal – the more often Arsenal get that referee for their matches!!
And that is not just a feeling – we’ve now got the stats to prove it.
The table below shows the referees for Arsenal games this last season, and is ordered in the number of fouls the referee awarded against Arsenal per game. (Figures from Who Scored).
First we notice an incredible variation from top to bottom. Oliver gives 386% more fouls against Arsenal per match than Mason. Are we really going into some games kicking the hell out of the opposition, while with others we are much, much more passive? That’s what these figures imply!
| Referee | Games | Fouls pg |
|---|---|---|
| 1.Michael Oliver | 5 | 11.60 |
| 2.Chris Kavanagh | 3 | 11.33 |
| 3.Jonathan Moss | 4 | 10.50 |
| 4.Mike Dean | 2 | 10.50 |
| 5.Andre Marriner | 4 | 9.50 |
| 6.Craig Pawson | 4 | 9.25 |
| 7.Stuart Attwell | 2 | 9.00 |
| 8.David Coote | 1 | 9.00 |
| 9.Paul Tierney | 2 | 8.00 |
| 10.Martin Atkinson | 4 | 7.75 |
| 11.Anthony Taylor | 2 | 7.50 |
| 12.Peter Bankes | 2 | 6.50 |
| 13.Kevin Friend | 1 | 6.00 |
| 14.Graham Scott | 1 | 5.00 |
| 15.Lee Mason | 1 | 3.00 |
OK that is bad enough, for although Arsenal’s style of play might vary from game to game it is really hard to see that they move from a rugby style approach which would lead to 11 fouls being called in a game, down to a passive style in which the club only gets three fouls against them.
But it is far worse than that, for the referees who gave the most fouls against Arsenal PER GAME handled 22 of Arsenal’s 38 league games!
Only one of the referees who calls fewer fouls against Arsenal got more than two games, while five referees who call foul after foul against Arsenal got three Arsenal games or more.
Of course one could call that a coincidence, but it is rather like two school children who sit next to each other handing in an exam paper with the same answers on, explaining that away by saying it was a pure coincidence. Maybe, but it’s unlikely.
And there’s worse. When we measure these referees against the number of points Arsenal got per game we can see just how much we were penalised.
Here are the six referees who we had three times or more in the season. Our average points per game through the entire season was 1.60. As we can see, yes Michael Oliver did oversee games which were above average, but the other five referees in this list, who between them controlled 19 games (which is of course half a season’s worth) all oversaw matches that gave us below our average points per game.
Have you got that. Five refs who all give us below average points per game, controlled half the Arsenal matches in a season.
| Referee | Games | Point pg |
|---|---|---|
| 1.Michael Oliver | 5 | 2.00 |
| 2.Martin Atkinson | 4 | 1.50 |
| 3.Jonathan Moss | 4 | 1.00 |
| 4.Craig Pawson | 4 | 0.25 |
| 5.Andre Marriner | 4 | 1.25 |
| 6.Chris Kavanagh | 3 | 1.00 |
Now you might be inclined to call that a coincidence too, but 19 games across five of the six referees who controlled Arsenal by far the most, and we get lower than our average points per game looks like absolute bias to me.
Of course I have no proof. I am just reflecting what the figures show. But it makes me ponder.
The proof that something is seriously wrong with football refereeing and reporting
- Results from crowdless games are different from those with crowds. Why?
- How even a serious science magazine has used fake data on football
- Someone is trying to fake football stats – and doing it rather well,
- Crowdless stadia stats reveal fallacy of PGMO claims
- Proving unconscious bias by referees
- How clubs manipulate referees through their tactics
- Referees are not 98% accurate but only 75% accurate
A comparison of the same stats for Manchester City and Liverpool from this season, as well as last season would have me sold on the idea. Bias might be more difficult to notice in teams with a higher expected win rate, but it will certainly be interesting to see who the referee association prefers to give these games to. The crown would be this same stat in the 2015/2016 season with Arsenal, Leciester & Spurs.
Pls can u do this same analysis for a team like Everton or westham?? Wondering what we may discover..
On a number of occasions you have accused the media of bias by looking at arsenal stats in isolation. Not comparing with stats for other teams. On this occasion you have done the same. A comparison between the ref spread with other teams would have told us if arsenal are being deliberately given certain refs or if that was the norm in the league
we were also the team that suffered most from VAR overturns:
https://www.espn.co.uk/football/english-premier-league/story/4182135/how-var-decisions-affected-every-premier-league-club-in-2020-21
would be very interesting to know which of the six above-mentioned conmen were at the helm of the VAR, then (i’m pretty sure moss was, at least once or twice)
Untold Arsenal is a site that comments upon aspects of information associated with the Arsenal Organisation; its team members; its supporters; agencies that accrue data about Arsenal etc. It is not a research document like Nature; New Scientist and the like. UA contributors publish and comment upon different correlations that appear in such data and information available from a number of sources. UA is not a research establishment nor a research document that is peer reviewed. It does not comment about actual causes with associated effects. It merely makes occasional suggestions as to what such correlations could mean and tries to explain why the National media never highlights such correlations to readers and football supporters alike. I personally find such correlations that appear on UA useful and interesting. I do get a little anxious when comments made as replies to articles criticise the author(s), implying that such articles themselves are biased and not worthy on inclusion on the site. That more research should have been done. That such correlations are meaningless unless the author(s) had extended the research, etc.
There is nothing to stop so-called critics from taking the highlighted data themselves and doing their own analysis and then returning to the site with data and information which could either support further argument or prove any suggestions made earlier, fallacious. But this is never done! Suggestions for improving any analysis is always useful but rather than just making comment, why cannot individuals not provide further analysis themselves? UA has undertaken research on refereeing in the past, through analysis of video evidence. UA has also cited published evidence of referee bias, especially that published through the University of London and the like. I feel that Tony Attwood and other UA contributors would welcome the chance to receive articles from sources who could offer the analysis called for from any of the above commentators replying to the article. The above article, as well as previous articles, gives me the chance to explore data and information in more depth if the need requires it. I thank UA for this and hopes that it continues in this task.
AKH – thank you for your kind words. And yes I can say that we do indeed welcome other research. We’ve not been receiving so much of it recently, but I can’t recall a time when I’ve had an article offered that does give us data rather than pure personal opinion, and I have not published it, irrespective of whether it supports my view of what is happening or not.
AKH
I would like to add my appreciation for your support as well because as you rightly say, despite many requests by myself (and others of course) for critics to bring the counter evidence to the table they very rarely if ever do so, often when asked disappearing for days, weeks, then repeating the same criticism again when they eventually return, and again without supporting evidence.
The point is I/We may be wrong and they may be right, but until they can be bothered to do the research and produce the supporting evidence I’ll assume they haven’t got any and that in fact I/We are correct.
It has long been my contention that there is a very simpler reason for the obvious bias shown by referees against Arsenal and that is geographical. Show me a London referee, show me a referee from the South East – you can’t as there are none. Only Graham Scott from Oxfordshire comes even vaguely close. PGMO is run by and almost exclusively comprised of Northern refs.
Michael Oliver – Northumberland, Chris Kavanagh – Manchester, Jonathan Moss – West Yorkshire, Mike Dean – Cheshire, Andre Marriner – West Midlands , Craig Pawson – South Yorkshire , Stuart Attwell – Warwickshire , David Coote – Nottinghamshire , Paul Tierney – Wigan , Martin Atkinson – West Riding, Anthony Taylor – Manchester , Peter Bankes – Merseyside, Kevin Friend -Leicestershire , Graham Scott – Oxfordshire , Lee Mason – Bolton.
I rest my case
@Akh, I concur with most of your assertions as to the status of the work done by untold. And indeed the site should be commended for its work. However where I always differ is when the work done is given a higher status than it should and presented as established knowledge. From your argument, I assume you look at work done by untold as raw information, and before you form a theory or opinion on issues, you look up other information that ties up loose ends. However when articles are presented as already established knowledge, using only previous own works as foundation then you leave the entire body of work open to criticism. A few examples
1. Walter conducts subjective ref reviews, concludes refs are biased against arsenal. There are follow up articles on untold, on other issues, discussing ref bias against arsenal as established facts, referencing a work which at best should be viewed as the authors personal opinion.
2. Tony says money spent does not improve the performance of a team in the league, using case study of Aston villa season before last who barely escaped relegation, and some others who spent much but achieved little. Nitram counters this assertion, but it doesn’t stop Tony writing other articles suggesting that the “fact” of money not being important in climbing up the table is already established knowledge based on his previous article.
3. Some years ago Walter did a series of articles “confirming” that arsenal had a disproportionately higher level of injuries than any other team in the league based on analysis of 10years of injury stats. He attributes this to be the reason for our poor performance in. League and blames referees for allowing other teams to “kick us off the park”. There were follow up articles blaming the refs for performance in the league based on the Walter study. Finally;
4. When RVPs letter to the fans came out. There was a series of articles on untold I believe written by Anna aimed at proving that RVP had nothing to do with the write up,. With follow ups castigating the media for being so eager to publish stuff that was patently fake. There have been others, Ozil, Higuain, Suarez, AST stories etc.
Now everyone has a right to an opinion, wrong or right, however when you start abusing people based on the fact that they don’t share same “opinion” with you, because you conclude that you have provided the “incontrovertible evidence” in your study and as such such matters should no longer be considered opinion, but hard facts, then you have begun to go astray. I believe most people know this, that’s why “in my opinion” is a common phrase. However I think untold let’s itself get too boastful about it’s research that’s why it readily denigrates others with dissimilar opinions. Aaa today, AST tomorrow, media everyday.
Also, you are wrong when you say people with dissimilar views “never” do their own analysis and post on the site. This happens all the time. I can remember nitram doing so everytime Tony goes on about spending and going down. In times past when the blog had little censoring there were other contributors who always presented stats, analysis for contrary views, but they always got quarreled and I believe eventually banned. I remember Rupert, Tom.P etc
Arome I would normally take up a number of issues with you, but I think the last point you make overrides them all – censorship. I do not censor expect for people who write abusive or illegal comments. Thus a person who writes “You’re a fucking tosser fuck off back to Tottenham” might well have his/her comments published, but this is a tiny number. I do not censor comments just because I disagree with them – rather I normally argue back.
What is concerning here is that you make this accusation without any evidence, but present it as a fact. That is concerning.
Tony I’ve been a reader of untold for years and you have justified banning commenters for so many reasons, to mention a few
1. That they are making arguments that are not in line with the majority view here. You justify this by saying there are a thousand and one blogs that agree with such poster
2. That it’s your blog and you have the right to decide what you give voice to. Common to hear you say you have given the poster opportunity to post a few comments, now it’s time to say goodbye
3. That a poster has not posted sufficient evidence.
4. That the poster’s comments is quite unpopular and not sitting well with other regulars of the blog who are fans of the blog because of certain views it projects
The point I’m making is that over the years you have banned many commenters for reasons other than being abusive or dismissive of the articles. Some of the banned have been among the most polite of posters, for example, Rupert cook never returned abuse. Also if abuse was an important factor, Menace would have been gone long ago.