Can constituency betting markets predict the general election result?
The political science boffins at the University of East Anglia have had a go at using Ladbrokes constituency prices to make a prediction for the general election. You can find our odds for every seat in the UK here. This is the forecast they have come up with:
Apart from the UKIP and SNP seat totals being a bit too high, that isn’t too far away from the overall market expectations. It’s pretty close to what you would expect if Labour and the Tories got approximately the same national vote share and I think they have done a good job at trying to tease out the “real” outcome suggested by our odds. In the coming weeks, Ladbrokes will be undertaking a similar exercise in order to make sure that our seat by seat betting matches our odds for overall party seats, majority betting etc.
In 2010 there was an interesting discrepancy between the individual constituency odds and those on the overall outcome, which presented a significant arbitrage opportunity for anyone sufficiently motivated. There is a post here which explained how we tried to deal with some of the differences.
In theory, it would be nice to be able to build up all of the general election odds from the seat by seat markets, In practice, there isn’t really enough liquidity on most of these yet to produce an efficient estimate just from the constituency odds alone. Hopefully, the nearer we get to election day, the more information we’ll be able to garner from these markets.