So, that’s a big claim – but I kind of have the answer. Towards the start of last year’s Premier League season, I predicted the final standings. The prediction was based on a fairly simple premise; the table after 12 games is pretty much the table after 38 games. As I said then:
“I heard about the theory a couple of years ago and – give or take a bit of movement – the theory was sound. There’s normally a couple of big changes. And as I said before, you have to have an accepted error of one or two places around most teams.”
So, how did I do for the 2009/10 season? See for yourself below:
12 games / 38 games / movement
- Chelsea / Chelsea (spot on)
- Arsenal / Man Utd (+1)
- Man Utd / Arsenal (-1)
- Tottenham / Tottenham (spot on)
- Aston Villa / Man City (+1)
- Man City / Aston Villa (-1)
- Liverpool / Liverpool (spot on)
- Sunderland / Everton (+5)
- Stoke / Birmingham (+6)
- Blackburn / Blackburn (spot on)
- Burnley / Stoke (-2)
- Fulham / Fulham (spot on)
- Everton / Sunderland (-5)
- Wigan / Bolton (+2)
- Birmingham / Wolves (+4)
- Bolton / Wigan (-2)
- Hull / West Ham (+1)
- West Ham / Burnley (-7)
- Wolves / Hull (-2)
- Portsmouth / Portsmouth (spot on)
Pretty good, is my conclusion. Just five teams (Everton, Birmingham, Sunderland, Wolves and Burnley) finished more than two places away from their predicted finish. As many as six finished in the correct spot. The rest were within the expected error margin of two places.
If you’d bet on the champions, you’d have won. If you’d bet on the top four, you’d have won. You’d have lost on the bottom three bet, but only just. And that is a hard one to call, because there’s always one club that falls like a stone.
You can’t actually beat the bookie, obviously. But the 12 game system is probably as close as you’re going to get.