Yep – it looks like City was robbed

(as dictated to Mrs Grumble)

Here’s a job for analysts with plenty of spare time on their hands: find out if the All Blacks are disadvantaged by referees when it comes to playing extra time away from home.

Alf raises the issue after reading a fascinating piece in The Guardian by Howard Davies – a Manchester City supporter of more than 50 years – who was dismayed in the last minute of last Sunday’s match against Manchester United.

Manchester United scored the winning goal in that period of play.

For the first time for years, the good guys matched the red devils kick for kick – until the last one, five and a half minutes into four minutes of extra time. We was robbed, innit?

Davies proceeds to reason that rules are rules, teams must play until the whistle blows, refs do their best, there is no systematic bias in the game – and so on.

Then he looks into the evidence that – in recent seasons – more extra time has been added at the end of games that Manchester United were not winning.

On average, 178 seconds are added when they are in the lead at the end of full time, and 254 seconds when they are behind.

Davies asks: how do we explain this phenomenon? Are most premier league referees closet United fans, or in receipt of regular distributions from Sir Alex’s famous wine cellar?


but academics at the LSE have produced a better explanation. Using data from a number of seasons of La Liga, in Spain, Luis Garicano, Ignacio Palacios-Huerta and Canice Prendergast systematically analysed referees’ practices in a substantial paper in the Review of Economics and Statistics.

They ask if it is true that referees do add more time at the end of games when it might benefit the home team, and if so, why?

The short answer is that they do.

“When the home team is behind in a close game, more injury time is added than when the home team is ahead in an equally close game,” they say. On average, in Spain, injury time is about three minutes. But if the home team is behind by one goal, the length of time added is 35% above average, whereas if it is ahead by one goal, added time is 29% below average.

So it is demonstrable that “referees use their discretionary power to favour home teams, but only in close games”.

Davies ventures his own one-eyed view that they don’t do this if the home team wears light blue, but their dataset sadly doesn’t allow him to prove that contention definitively.

He goes on –

Furthermore, there is evidence of bias related to what goes on during injury time. When visitors score, injury time is 15% longer than when the home team scores.

“In other words, referees are more speedy in blowing the whistle for the end of the game if the home team scores, thus giving the visitors less time to respond”.

It’s lucky that Craig Bellamy didn’t score the seventh goal on Sunday, or the teams might still be playing.

So far, so good. But what’s going on, if we rule out corruption, as the study’s authors do?

Their hypothesis is that crowd size and partisanship impose pressures on referees.

The title of the paper, Favouritism under Social Pressure, summarises the argument. When crowds are bigger, referees become more biased.

So with the biggest ground in the Premier League, United are likely to benefit the most from this pressure-induced bias. QED. So we was, indeed, robbed, and we will be again.

Davies reminds us that Fifa did reduce referees’ discretion after the 1998 World Cup, so the minimum amount of added time is now declared by the fourth official.

But, as we saw on Sunday, it is a minimum and does not fully resolve the problem. Another option, especially for derby games, might be to increase the number of away fans allowed to attend.

The Spanish research shows that where the crowd contains a substantial number of fans supporting the visiting team, the referee’s bias in favour of home teams is mitigated.

There were more than 3,000 vocal Blues in the crowd, singing Blue Moon for all they were worth, but they were still overwhelmed by more than 70,000 homers.

A final question: How much does all this matter?

In Spain the result was affected in 2.5% of the games studied. That is a small percentage, admittedly, but it is enough to be of significance in a close division.

And if referees are influenced by fans in decisions on extra time, where their discretion is heavily constrained, might they not be influenced in other areas too – cards and penalties, for example?

It seems highly likely, but I will leave Spurs fans to argue their case for Keane’s penalty that never was against Chelsea last weekend. If they want to pay for some more data mining, I’m sure our LSE academics will be happy to oblige.

So there we have it. Soccer referees are influenced by the home crowd.

Alf reckons it’s reasonable to suppose much the same thing happen in rugby.

Anyone want willing to do the numbers to prove it?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: