Listen to the World Soccer Talk Podcast review of the Premier League weekend (gameweek 17) »

FRI, 7:45AM ET

Referees and The Confirmation Bias

c4140d578dbefc683e095db035e1 large Referees and The Confirmation Bias

Seen this elsewhere today?

Referees. Eh? Who’d be one? Not me anyway. I watch football, play football, love football but would never want to put myself in the position that thousands of well wishing men & women do. They are vital to the function of the sport and a good referee can make a game better. A bad referee, on the other hand, can truly ruin the game and in some extreme cases the sport itself (Calciopoli). So, in writing this I do not demean, devalue or demonise an entire profession but offer my, lay, opinion on one of the most important facets of the sport.

As Brian Clough famously put it to John Motson  (in relation to Televisions coverage of Football):

“I think what you do to referees is absolutely criminal, I do honestly and I think the standard you feel that should be coming from Referees at the moment is absolutely incredible…he makes a decision in 5 seconds or 2 seconds or 1 second or whatever it is, in the heat of the moment with 22 players, with 33,000 shouting and bellowing. All I’m saying is that you don’t make that point strongly enough. It should be overemphasised how hard it is to Referee a match.”

However, Referee’s have a flaw one that all of us possess and that is Confirmation Bias, it influences their decisions before they make it and makes it more difficult to change their minds despite the evidence in front of them, what can be done about it?

(In this article I will be quoting from an academic article by Raymond S. Nickerson of Tufts University.) Each time a quote appears indented it is from said study.

What Is Confirmation Bias?

Confirmation bias is the cognitive process whereby we reinforce our opinions by systematically accept what confirms what we already believe and discredit or point blank ignore evidence to the contrary. The problem this causes is that it can allow narratives to develop that reinforce stereotypes that may not be true. If that went over your head ask yourself; are Wolves a dirty team? If you haven’t gone in depth and go by the media’s depiction then the answer is yes. Conversely are Bolton a dirty team under Owen Coyle? If you don’t have a strong opinion you’re more likely to assume not, but in the Premier League fair play table the trotters are 5 places behind but nothing is said about it.

What this causes is an extra emphasis on Wolves’ dirty play and no emphasis on Bolton’s, this is a perceived correlation whereby:

“A form of stereotyping involves believing that specific behaviours are more common among…members of a particular group”

This would back up what we have seen, why was this yellow? Probably because it was John Carew who is not ‘known’ for horror tackles unlike, say, Karl Henry is now. Wolves’ players are going to be held by a different standard than those who play in the same league as them which is staggeringly unfair but falls within this illusory correlation that because they play for Wolves they are more physical. The reason for this is not just a spate of bad tackles, which of course there have been a few but also because

“Unusual behaviour by people in distinctive groups is more salient and easily remembered than similar behaviour who are not…of that group”

It is mental gymnastics that we will remember the fouls of Wolves to tally with what we already believe but conveniently forget moments like this.

Johnny Giles wrote an excellent article recently(which I didn’t fully agree with but that’s the way with debate) in which he broached the subject of how he was remembered:

“I took the decision that if I was to embark on a career in which honest comment was essential, in which I might have to condemn bad tackles, for example, then I would have to come clean about my well-earned reputation among my fellow professionals. It was pretty embarrassing, having to tarnish my own image in the Daily Express. Some of my friends thought I was mad to do it and warned me that, in future, I would be known only for the bad tackles, that the creative play would be forgotten.”

So despite a ‘well-earned reputation’ it took a nationally published article to refute what had been ‘known’ for decades. That is the way with confirmation bias unless you’re conscious of it you will be prone to fall into it and it takes an explicit contradiction for you to change your mind, of course some never will.

Specifically, Why Referees

Confirmation bias is universal and it touches on subjects I’ve already broached (Revisionism) as an impartial observer – as referees must be – it undermines their standing. Referees hold a unique position within the game, they have sole power over the course of a football game, their mistakes change matches and are widely publicised. He succeeds or fails on his own judgement and if his judgement is flawed then so is the game. Take for instance the recent case of Nemanja Vidic and Marc Albrighton. Albrighton had gone to the crowd after scoring a goal and promptly was booked, Nemanja Vidic did the very same in the same match – in the same half – and was not. Mike Dean said that Vidic was ‘pushed’ into the crowd. He came to the decision that Vidic shouldn’t be booked (and then dismissed) and fitted the facts to confirm this. Watch the incident again, Vidic is clearly not ‘pushed’ anywhere until Rio does his usual celebration and at that point he is already in the crowd. By the laws of the game and precedence in the match itself Vidic should have been off but Dean made his – incorrect – decision and stood by it. There is evidence to this effect almost every week;

Sometimes it’s because the player isn’t ‘known’ for that :

“If it was Lee (Cattermole) he would have been sent off…It is a strange one to give a yellow card for that” – Steve Bruce -

Sometimes it’s because the result of an accurate decision is unpalatable:

“Having seen it again from my armchair several times in slow motion and from different angles I can see that it was a red-card offence.” –Howard Webb -

In this particular instance it is shown that a desirable outcome can be preferred to the truth. Which as a keeper of the rules of the game is unacceptable for a Referee.

Mark Clattenburg dismissed every Tottenham player that surrounded him when discussing the controversial Nani goal in October; however he did allow Rio Ferdinand to join his conversation. I cannot know for certain but it would certainly seem that Rio was providing confirmation of what he already believed, why else would he allow the United player (not even a Captain) come that close during a conversation with his assistant?

“The idea that people tend to expose themselves more to information sources that share their beliefs than to those that do not has had considerable credibility among social psychologists.”


Why? Referees

As indicated above it is all well and good to judge a Referee from the comfort of my home with the ability to pause, rewind and review incidents on the fly, there is currently no way for a Referee to be 100% accurate, 100% of the time. However Referee’s are placed in a position whereby the influences of confirmation bias are exacerbated not alleviated. By placing the referee in a position where they must make a decision immediately then they are more likely to stick by it, this is called ‘the Primacy Effect’:

“When a person must draw a conclusion on the basis of information acquired and integrated over time, the information acquired early in the process is likely to carry more weight than that acquired later”

If a Referee must make a decision there and then, the conclusion he comes to will hold more weight than any subsequent evidence, furthermore:

“Once a belief or opinion has been formed, it can be very resistive to change, even in the face of fairly compelling evidence that it is wrong.”

Also the position of power that Referees find themselves in can be a problem in and of itself. The referee’s who we see week in and week out are the elite of each division and as such that comes with an ego. The demeanour of Mark Clattenburg as Heurelho Gomes questioned what was going on at Old Trafford is the extreme example of how this can show itself.

Being placed in the position to make the decisions in a football game does not lend itself to fallibility, if a Referee were seen to waver over his decisions he would not be long for the professional game. Consider the recent (extreme) example in Scotland where the Referee (Dougie McDonald) awarded a penalty, when in reality the Goalkeeper had touched the ball away, he immediately realised his error but rather than admit it consulted with his linesman and then indicated a different result. This isn’t uncommon, what is though is that after the game the reason he gave for the change was that he had been heavily influenced by his assistant and not that he had realised his own error. The assistant in question – Steven Craven – resigned after the facts came to light as he was disgruntled with the handling of the situation. Why did the referee not just admit his mistake? It would seem that his decision was based on self-preservation and that he identified that concocting a story about his linesman was less damaging than admitting that at first instance he was incorrect.

What does that say about the mindset of a professional referee that when confronted with being wrong it is preferable to invent a lie rather than admit the mistake, how likely is it then that a Referee will accept conflicting information if the result is so abhorrent:

“Friedrich (1993) argued that “ our inference processes are first and foremost pragmatic, survival mechanisms and secondarily truth detection strategies”

It would certainly hold that the mistake, actions and subsequent strike in Scotland bear this out.

The main issue I have with Confirmation Bias in this context is the narratives it can allow to develop. Certain players will develop reputation which will unfortunately distort their actions beyond what has actually happened. This leads to a disparity in the game which should not be tolerated it can lead to a different set of results that given the finely tuned nature of league football could be the difference between promotion/relegation and all the rewards/pitfalls of both. Over the course of a match one or two decisions will not have a profound effect, over 38 games however if those decisions are based on and enforce a bias then that can have a cumulative effect far greater than it initially seems.

Can We Fix the ‘Problem’?

In terms of eradicating the bias? no. It is universal and probably evident in this article denouncing it. What we can do though is acknowledge it. Acknowledge there is a problem and take steps to mitigate it. However, as we have seen the rules of football are quite resistant to change and particularly in adding/removing layers of protection, especially around Referee’s. To admit there is a problem brings the role of the Referee into sharp focus and as already mentioned it is already an extremely hard job at all levels of football world-wide. One must only look at the paltry ban provided to Tresor Mputu for his shameful attack on a referee to see that the protection of Referee’s is not the current priority in Football. In my opinion that should have merited a life-time ban.

However when it comes to contentious, game changing decisions there must be more of an emphasis on right rather than belief. As football has become a Billion dollar industry the price of mistakes are too large now to simply ignore. So what can we do?

It has been shown that:

“ the tendency to use [confirmation bias] can be reduced if people are asked to consider alternatives, but that they tend not to consider them spontaneously”

In this regard it would take a review to come to the ‘best’ decision of course given what has been shown before this probably should not be done by the official himself as having come to the decison independently he is more likely to ignore evidence to the contrary. It should, therefore, be the remit of a third party a qualified or senior/retired Referee who could review the video evidence and advise if anything was missed/misinterpreted in the initial instance. So yes, I am advocating the use of Video Evidence and systematic video review. It is used in virtually every other international sport extensively without issue. I do not of course advocate it’s use for every single refereeing decision but as said before those which are game changing. Goals, Penalties, Red Cards all must be reviewed if they are contentious.

Of course this is an imperfect solution as the word ‘contentious’ would be open to interpretation and interpretation is already the problem. However replacing a flawed system with a less flawed system is better in any circumstance. It’s not just a case of modernising the sport( as that’s a different discussion altogether) but this would remove a significant proportion of human error from the game because:

“Much of the discussion of confirmation bias is predicated on the assumption that…people have been interested in determining truth or falsity….But determining truth or falsity is not the only, or necessarily primary,objective…. Another possibility is that of guarding against the making of certain types of mistakes”.

If we acknowledge the problem and act to prevent it perhaps a ‘reducer’ from Cattermole could be as equally punished as a ‘mistake’ from Fabregas. If we forcibly divorce ourselves from our unintentionally pre-conceived notions of what a player meant to do and act according to what they actually did the game would be fairer as a result. Also if we acknowledge that Referee’s are human and therefore flawed, we can accept that they need help to determine if the ball crossed the line or was prevented from doing so by a hand.

Also, by admitting fallibility it is less likely that Referee’s can develop an Ego like Mark Clattenburg has, in fact his ego could exacerbate the problem as:

“In general, people tend to express a higher degree of confidence than is justified by the accuracy of their performance….[Researchers] refer to the confidence that people feel for highly fallible performance as the illusion of validity.”

If we provide a second stage of scrutiny to a Referee’s performance it would lead to a more thorough (and accurate) review of decisions. I’m aware that this would slow the game down somewhat and that by sometimes ‘letting the game flow’ is a praise for a Referee rather than an indictment. Though by limiting the decisions to the fundamentals of the game (goals, player’s violent actions and penalties) we would get less debate over the Refereeing of games (such as this article) and more on the sport itself and perhaps it would prevent accusations of bias and make corruption that bit harder as you would have to corrupt two officials rather than one.


Of course I acknowledge that quoting extensively from one article is an almost text-book definition of confirmation bias however, I have chosen, read and digested an article which presents both the pros and the cons and it is a piece that admits that:

“Can we assess the merits of our own opinions impartially? Is it possible to put a belief that one holds in balance with an opposing belief… and give them a fair weighing? I doubt that it is. But that is not to say we cannot hope…to do better than we typically do in that regard.”

If this has whetted your appetite for such a debate I encourage you to read the article quoted throughout. Finally, I reiterate that this is not a treatise against Referee’s, what it is is a realisation that they are only Human and we need to provide support rather than condemnation and instead of singing ‘Fergies Rent-Boy’ sing ‘Damn Psychological Social Conventions’ it’s not as pithy but it just might be a bit more accurate.

Chris McQuade can be followed on Twitter @kipp9

This entry was posted in General, Leagues: EPL. Bookmark the permalink.