BS Detector: Will David de Gea's exit really cost United 10 points a season?

It was a great line, the stats coming from our collaboration with the makers of the Football Manager computer games, suggesting that the key saves De Gea is expected to make would turn five draws into wins and two defeats into draws, so that Real would go from also-rans to Spanish champions.

Now, Sky Sports have got in on the act, looking at things from the other point of view and boldly claiming that Manchester United would drop 10 points next term if they lose their Spanish goalkeeper.

Stats gurus Opta were behind the Sky Sports feature, which stated that "no Premier League side made more unpunished errors than United. It was the last line of defence that so often proved vital and the statistics for clear chances."

"A few years ago, clubs asked us if they could have some sort of qualifier that distinguishes shots that were really clear-cut chances from regular shots," explained Opta's John Coulson.

"That's something that we've added. From that, we can derive a goalkeeper's save percentage versus his save percentage when he's in a big-chance situation."

There's a slight problem, though.

Both claims are an utter load of cobblers.

Sadly, it's true. While Opta can do their level best to try and decide what makes a clear-cut chance as opposed to a common-or-garden chance, there is simply, absolutely and categorically no way of saying whether one highly-skilled, highly-experienced and highly-trained goalkeeper would have saved any given attempt on goal that other keepers would have missed. The only way you could know for sure is with some sort of Quantum Leap-style sci-fi equipment that allowed you to zap Hugo Lloris back in time and into De Gea's body at a string of crucial moments to see how he fared.

Nobody's denying that for a second. But Manchester City Joe Hart had more clean sheets (14 to De Gea's 11) and pulled off more saves for each goal conceded (2.38 to De Gea's 2.17).

Not so fast. Hart's error count was higher, for example - he made three blunders to give away a goal to De Gea's one.

But Hugo Lloris also made just one mistake to give away a goal - despite being, you'd imagine, under considerably more pressure considering that he faced far more shots - and made 89 saves to De Gea's 78.

No - well, possibly, yes - but that's really not the point. The point is that all the top teams have top goalkeepers, and to try and look back at the previous season is utterly absurd. Though De Gea made some magnificent saves, there's no possible way of saying whether any other keeper might have made them.

No, and in some contexts they are helpful. A striker's chance conversion rate, for example, should balance out over the course of a season into something sort-of meaningful. Equally, a goalkeeper's saves-per-goal ratio is good. But there are value judgements and relative judgements artificially added to the claims being made about De Gea which take them outside that zone. It's not using the numbers, Moneyball-style, to illuminate. It's fudging the facts to make them fit a story which just 'feels' right.

Not at all. Once you start doing that, you're just in the realms of opinion - and that's totally fine, but don't dress it up as facts please.

David de Gea is claimed to be so much better than any of the goalkeeping alternatives at Real Madrid and Manchester United that he'd be worth 12 and 10 points respectively to each of those teams. Those claims are absolutely impossible to substantiate, or even take vaguely seriously.