The Community Shield played between the winner of last season’s Premier League and FA Cup is the traditional starter to the English football season, and today Arsenal beat Manchester City 3-0. I don’t know about you but when I ought to be doing a boring task I often procrastinate by playing with statistics – exciting, huh? Well, in that spirit I looked at the data for this event since 1958 to see whether the result tells us anything about the coming season. Someone else must have done this before me, but I haven’t seen it anywhere before.
Some games in the past were draws (now there is a penalty shoot out). Of the 46 games with a result, the winner ended the following season above the runner-up 32 times – 70% of the time. The winner ended the season in the top four 35 times (76%) and the runner-up 26 times (57%), the winner won the league 12 times (26%), the runner-up eight times (17%).
But in recent years the event has become dominated by the clubs to even larger extent than in the past. In the last 20 seasons the winner finished the season in the top four 17 times (85%), the runner-up 16 times (80%). Both the winner and the runner up went on to win the league title 6 times, i.e. 12 of the last 20 champions (60%) played in the game. It seems like the difference between the subsequent performance of the winner and runner-up may be narrowing, which makes sense if the teams are drawn from a smaller, dominant pool. But the differences do not appear statistically significant.
The size of Arsenal’s victory might be thought to contain extra information about the future, but since most results are closer it’s difficult to generalize. Arsenal fans will certainly be wary of complacency – in 1998 they defeated Manchester United 3-0 in the Community Shield, only to end up as runners’ up to them in the Premier League in the following season.