An article on 606 that has received scant interest so far. It's about an article posted on the bbc football page about independent measuring of referees performances in the premiership. http://news.bbc.co.uk/sport2/hi/football/9379989.stm First off, it's interesting that The Premier League do not release statistical information about refereeing performances but insist that they're doing a good job. There is an obligatory nod to how video technology being implemented in most other major sports has helped improve decision-making no end and how soccer is lagging behind. The excuse we're given for video technology not being implemented is that it would be unfair as it couldn't be implemented across the spectrum of football. This of course, could be offered as an excuse in any of the other major sports but isn't. So would people agee that FIFA refuse to implement it because they cannot corruptly control the outcomes of matches with such ease if it is implemented? And also, there is research into referee's performances being conducted by two Belgians that indicates that referees are only getting 62.75% decisions correct in our games. In fact, two of the referees got as little as 40% of the decisions correct, namely Chris Foy at home to Fulham, and Howard Webb away to Manchester United. I expect many to point to the fact that the two Belgians are Arsenal supporters, and in response to that it was interesting to note that they named Mark Clattenburg as our best referee so far in our 2-3 defeat to West Brom with 93% of decisions correct. Irrespective of that, these people are paving the way for independent statistical research of referee's performances, and already their research is showing up patterns of biased refereeing for or against certain teams. As this research comes into the public domain, I wonder how The Premier League will justify using referees in matches where they've clearly shown a bias towards one or both of the participating teams in the past?
It's an interesting analysis, but fundamentally flawed. Broeckx, the referee they have doing the analysis, fully admits that he could not be unbiased when refereeing Arsenal. So how can his analysis of what decisions referees get right be unbiased? If the data is biased, then the conclusions will be inaccurate. The ref also doesn't show how many of the incorrect decisions actually went in Arsenal's favour. I'm a qualified ref with a degree in business and statistics, and I bet I could produce a similar analysis 'proving' that all referees are biased in favour of Arsenal, just by being selective about the decisions I deem to be correct. I mean the guy is seriously trying to claim that only 12.5% of penalty decisions in Arsenal's games are correct?! You don't think Wenger would be making a much bigger deal about penalty decisions if this was even vaguely accurate??? And he claims that Clattenburg, who got an eight month suspension for dodgy business dealings, and who has been suspended and banned from referring Everton games due to his shocking performance in the Merseyside derby, is the best ref in the Prem! If some genuinely independent analysis, with a panel of referees and an independent statistician, can prove there's bias, then that would be huge news. This analysis is basically just an Arsenal supporting ref, who's claiming that he thinks decisions made by refs in Arsenal games are not generally correct, and thus there must be a bias. Hardly convincing stuff.
my head hurts after reading all of this. Does it include how much the ref's get bunged to produce bad performaces and let go crucial moments to opposing teams.
Lol, no that's in the second phase of the analysis. The need to recruit the 'fake Sheikh' first... As far as I understand it, a qualified ref and a stats geek, both Arsenal fans, have reviewed a range of football matches and counted the number of errors the ref made. Apparently the number of errors in Arsenal games rose sharply after Arsenal were recognised as genuine title contenders this season, which they say is after the Villa game. They claim this suggests a bias by the refs, but don't show any evidence whether that bias is for or against Arsenal. They also make some really dodgy calls, like claiming that only one in eight penalty decisions involving Arsenal have been correct, and Clattenburg gets 93% of his decisions right. Also, if you read any of their analyses on the games, you can see that they are far too partisan to make any objective judgements. They start going on about bad decisions for Arsenal 'cancelling out' bad decisions for the other team and thus not having to be counted. They follow that up with the standard "Utd get fewer cards and more penalties at home" argument which has been statistically proven to be true for all home teams: http://www.research.salford.ac.uk/cms/news/article/?id=41
Swarbs............ Much as I hate to agree with a Manc - albeit one working in Moscow (is that right?) - I think your argument is spot on. After all, any analysis can be made to fit any pre-conceived notions regarding any team. There are lies, damned lies and then there are statistics..... The hard truth of the matter is that refs are human and we all make mistakes. I would hate to ref a Premier League game - the pace is so fast that the notion of getting ALL decisions 100 per cent right is non-existant. I would also hate to ref a Prem League game because it would involve running around a lot....something I believe would be injurious to health. As for those that argue that "technology" would elminate a lot of ref-ing mistakes; wrong. FIFA have been only talking about goal line technology. All that might achieve is resolving the old question of "was the ball over the line or did that dozy defender handle on the line" That of course does not even go half-way to addressing the current problem of refs making the wrong shout anywhere else on the pitch. Until there is a "Robo Ref" who can float around the entire pitch wielding his - or her - new tech powers we're stuck with what we've got - a fabulous game called football.
First off, he didn't claim Clattenburg was the best referee in the league. He said that Clattenburg put in the best refereeing performance statistically of all the ARSENAL games only. Why do people keep misunderstanding that part of the text? Second off, as I already said in the op, Clattenburg took charge of the WBA game which we lost, so whilst this doesn't completely give the researcher a carte blanche, it's certainly an indicator that he's analysing games objectively. And finally, when I say this is huge, I'm not referring to this particular piece, but to the fact that it may induce a flurry of independent research and analysis of referee's performances. That's not just welcome by me, it's long overdue. I can appreciate that a United supporter would try to dismiss this sort of research out of hand though.
Thanks mate, I was worried I was gonna get overwhelmingly shouted down by Gooners, good to know at least one of you shares my scepticism! Even Robo Ref will struggle given the level of disagreement over almost every decision made, and the importance of judgement in all but the most straightforward of situations - I've seen four different opinions from Arsenal fans about Barton's tackle on Diaby alone. Some say it was fair, some say foul, some say book him, some say send him off. Interesting that the analysis by this "DogFace" says that it was definitely a straight red, and thus that the ref made three errors (not blowing for a foul, not giving Barton a card and not sending Barton off) no matter what the commentators, MOTD, other referees and many of the fans say! With judgements like that, it's not hard to see how these 'stats' can be manipulated, particularly when the ref gets no credit for correctly not producing a card or sending a player off when they don't deserve it. Robo Assistant Ref would make a big difference though - they should have those little R2D2 style things that they use for photo finishes in the Olympics. Two could be programmed to follow the last defender and provide a 180 degree image of the pitch in front of them, and one could be hung from a gantry for a top down view of the whole stadium. Whenever there's a possible offside call the 4th (or 5th) official could just look at the feed from the cameras and immediately radio the ref to say whether or not it was offside. There'd be a slight delay, but that's probably worth it to make sure the decision is right. Definitely agree with that. The sooner we get a more formal and rigorous assessment of refereeing performances (i.e. not the 'having a senior ref look over selected performances bollox that we have right now) the better. Although this isn't exactly breaking new ground - the methodology seems like a pretty sizeable rip off of the research from Salford uni that I linked to above, that was done about four years ago. Couldn't resist that little dig could you? I didn't actually dismiss it out of hand, I looked at the methodology, the sample, and the judgments made in creating the data set before pointing out how fundamentally flawed it is. I could provide numerous examples of poor quality statistics, and immense bias in the data itself, but that would probably bore everyone to tears. Just look at the Barton example above to see how the research has created and inflated the level of 'errors' to make refs look bad.
I think it's difficult to eradicate bias completely from football, and that includes responses on this thread also
5/1 ratio for pens given for Utd at OT as oppose to those given against does suggest a partisan approach though - even given that most teams will be defending more at OT
You'd think so. But given that Utd's ratio of goals scored to conceded is 4.5/1 (scored 179, conceded 40) over the same period, the penalties are actually roughly in proportion to the goals. Which implies we're not much more likely to score one of our goals at OT from a penalty as our opponents. You know what they say about damn lies and statistics, right?
