Predicting Britain’s Got Talent: And the last shall be first

No sooner has Eurovision finished than Britain’s Got Talent limbers up for its closing week. If you’re contemplating a bet, especially on the semi-finals, here’s one thing you should be aware of: the remarkable correlation between performance order and the vote in the three previous series.

In 2007, there were three semi-finals. In all three, the public vote was topped by the act performing last. In 2008, there were five semi-finals. In all five, the public vote was topped by the act performing last. Last year, the pattern continued in the first three semi-finals: the last to perform came first in the vote. 

Finally, in the last two semi-finals of 2009, some variety was displayed in the voting results as well as on the stage. But not too much. The vote-toppers each performed near the end – sixth out of eight – and had each been the hot favourite to top the vote before the show started.

Correlation between popularity and performance order is likely to be part cause and part effect. If the producers want to keep ratings high throughout the show, it makes obvious sense for them to put the most-anticipated acts on near the end. And those who perform late are then fresher in the public mind when the phone lines open.

Does this correlation hold in the final? You might expect it would have less effect, as each of the acts has by then become more familiar to the voting public.

In 2007, much of the hype was concentrated on one act: Paul Potts. He was favourite in the betting. He performed last. He won. In 2009, much of the hype was concentrated on one act: Susan Boyle. She was favourite in the betting. She performed third from last. She didn’t win. 

(The winning act, Diversity, performed second from last. SuBo, of course, placed second. And the act performing last of all, Julian Smith, came third in the vote. So the last three to perform were the first three in the vote, albeit in a different order.) 

In 2008, the picture was more muddled. Final spot in the running order went to a pre-show 20-1 outsider, Signature. They came a better-than-expected second. The two acts which had headed the betting before the show – Faryl Smith and Escala – performed eighth and ninth. But the winner was George Sampson, who performed seventh. And third place went to Andrew Johnston, who performed sixth.

So it’s far from a precise correlation. But it’s also been far from random. In all, of the nine top-three finishers over the finals of three series, five performed last or second-last and only one came from the first half of the final – and that exception was in Paul Potts’s year, when the final had only six acts.

There’s no guarantee that these patterns will continue, of course. But if you bet on this show before you know the running order, be prepared that to some extent it could end up being the running order itself that you’re betting on.

 Leave a reply...