Election forecasts

Will any party get more than 300 seats?

A growing number of websites are making regular predictions of the election result. The Guardian is the latest to join the pack, offering daily forecasts based on a range of sources and calculations.

The greater the number of predictors, the better a picture we’re likely to get of the range of possible election outcomes. Or so you’d think. Instead all the main websites have converged on a very narrow band of numbers. Both Labour and the Conservatives are currently being forecast to win around 270-280 seats. That’s a very narrow prediction. What’s more, it’s one that is being suggested not just by the Guardian, but by May2015.com, Election Forecast, Elections etc. and even the bookies Ladbrokes.

Can they all be right?

Either everyone is making the same mistake about what’s going to happen at the election, or everyone is right about what is going to happen at the election.

What I find particularly striking is how nobody is forecasting any party to win more than 300 seats. They’re not even expecting anyone to get close to 300.

This is remarkable when you consider both Labour and the Conservatives are continuing to talk up their chances of forming a majority government (i.e. winning over 326 seats!). Some commentators still think the Tories will actually add to the 306 seats they won in 2010. As for a minority government, I’ve written before of how I think 300 seats is an absolute minimum if such an administration is to survive more than a matter of weeks.

I’ll be making my next prediction of the result in a few days’ time on 7 March, when it’ll be exactly two months until the election. As ever, I’ll be making it on the basis of how the parties will fare in individual seats, rather than extrapolating anything from polls or statistics. It’s an admittedly more granular approach than that used by any of the forecasting websites, but I’ll be interested to see if I end up putting Labour and Tories inside the same 270-280 band.


One response to “Will any party get more than 300 seats?

  1. It’s worth bearing in mind that these models factor in swingback that has happened at previous GEs but is not showing much sign of happening at this one. “Nowcasts” (that is, seat extrapolations from current polling, rather than predicted GE vote shares) show Labour on around 300 seats, and the longer the polls stay where they are, the more reasonable 300 Labour seats becomes.

    That said we got two polls showing the Tories 3% ahead today…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s