Given that Saturday marked yet another blooper reel worthy gaffe by Portsmouth’s David James (above), one that essentially gifted Middlesborough a 1-0 home victory, perhaps an editorial in Sunday’s Observer wasn’t the best occasion for the keeper to proclaim “years ago, the only statistic that counted in football was how many pints you could drink in a night.”

These days, football is a very different beast. The English game is ever more Americanised in its obsession with stats. Top football clubs are now using a model of statistical analysis similar to that used by Billy Beane in Major League Baseball to tell us how we won, how we lost, how to pick the side, or even how to buy players.

Most people know that I like a stat or two, so I’m not dismissing their value. As a kid I spent hours poring over football annuals, obsessing over clean-sheets records, attendances and county-league statistics. But data is a complicated business. Statistics are meant to be absolute, but once you start asking how they have been collated, or what they mean, you find yourself needing not just one stat but several. You can see how I became obsessive.

Peter Schmeichel best showed how numbers can be fiddled. Years ago there was a story going round that Schmeichel got the hump because of the introduction of ProZone, so decided to prove a point. The very next match, so the tale goes, every time the ball was down the other end, Schmeichel did sets of sprints across the edge of his area to raise his high-intensity running stats. Anyone watching probably thought: ‘Oh look there’s Schmeichel keeping himself warm’; but he ended up beating one of the forwards on stats for that game.

Where does coaching, that age-old skill, come into all of this? If footballers are recruited on their statistical performances, then where is the opening for managers to coach the best out of a player? The young footballer who shows flashes of brilliance but needs an arm around the shoulder, or a kick up the backside, may never get a chance under a stats-obsessed manager. A decent old-school coach doesn’t need to look at a load of stats to work out how good a player is. I certainly can’t see Harry Redknapp doing it – he knows his players and he doesn’t often buy a bad one.

Beane’s stats revolution may work for a team emerging from administration and needing a cost-efficient solution to get into the play-offs, but, like Beane’s Oakland A’s, they’re never going to win the title. Pints aside, the only statistic that really counts in football is the result.