Click to download the image above as a full-sized PDF file

Is it all in the mind? When your team are clinging on to a one-goal lead, it always seems as if the fourth official chooses that occasion to add on, I don’t know, seven minutes or something. But it must just be in my head, mustn’t it?

This week the Fink Tank has been looking at stoppage time. And our results are very striking indeed. They suggest that injury time is, well, badly injured.

Dr Henry Stott and Dr Ian Graham took a look at the added time granted in 1,140 matches played over the past three seasons. The aim was to see if the football myth — that added time favours the big teams — could be true.

The first look at the data showed nothing amiss. When you looked at the distribution of added time, there were spikes at the exact moments when the time on the board had run out, the more goals there were the more time there was added and the second half had more added time than the first. You would expect that, because the second half features more goals, more bookings, more substitutions and, possibly, more time-wasting behaviour.

So far, so good. But then we started to look at team quality. We used a crude method. We compared the average league points over the three seasons that a team obtained with the average amount of time added that these teams had to endure when winning at home.

And immediately something disturbing popped up. With the naked eye it was possible to see that worse teams were being forced to defend their lead for longer, playing more additional time when they were ahead.

The teams playing the most added time when ahead at home were Stoke, then Watford, West Ham, Derby and Wigan.

The teams playing the least when they were defending a lead at home? Manchester United, then Liverpool, Aston Villa, Everton, Chelsea and Arsenal.

But before we go jumping to conclusions, there could be a reason for this, couldn’t there? Perhaps the poorer teams have been scrapping for their victory, fouling more, time-wasting more, substituting players more.

We needed to use a multiple linear regression analysis to get at the truth. This is a statistical analysis that allows us to isolate the impact of different events on the amount of additional time. And the results do not make pretty reading for match officials.

We have two big findings. The first concerns the disparity between the time added on for events that take place during the match and the amount added on for the same event happening in stoppage time.

Perhaps this happens because the crowd and the referees are so acutely aware of every second in injury time. But happen it does. So 22 seconds are added on for an offside taking place during stoppage time, but none if it happens in normal time.

A goal in the second half might lead to only four seconds being added, but if it takes place during stoppage time, the same event adds 38 seconds to the play. A card during the second half merits five seconds of time added on, but during stoppage time it merits 17 seconds.

The second finding is even more worrying. If a home team are clearly better than their opponents (20 points or more per season) and are leading, it significantly (statistically significantly) cuts the amount of stoppage time they have to play — by 13 seconds in the second half and five seconds in the first.

Weaker teams have to sit on their hard-earned leads away from home for longer and get less chance to chase a point. Fact.