Political geeks everywhere would do well to head over to FiveThirtyEight for a fascinating fracas over projecting seat totals in this election. At stake? Up to sixty seats and the future of the trusty old BBC Swingometer.
In the red corner is Nate Silver, a baseball stats expert whose made his name by routinely debunking very well paid US pollsters.
And in the blue corner is Robert Ford of the University of Manchester, fighting on behalf of the PoliticsHome forecasting model and the UK political science establishment. It’s no-holds barred.
The nub of the argument: is “uniform national swing” hypothesis fit for purpose? In their purest form, these traditional models assume the same change of vote in each seat, a simple calculation that has dominated our analysis of elections for more than a generation.
But this hypothesis is best suited to two big parties fighting for votes in a country that, constituency by constituency, tends to change its vote in the same way. It pretty much failed in forecasting seat changes in 1997, Britain’s last big change election. And its most basic assumptions are being tested to destruction by differential swing in constituencies and the rise of the Lib Dems.
Nate Silver’s heresy has been to propose a “proportional swing” model, which assigns one party’s 2005 vote to another. This means that vote changes are related to the standing of the parties in each seat. So Labour would, for example, lose more votes in seats where it’s proportion of the vote share was high in 2005.
Silver’s more 3-D model gives the Conservatives and Lib Dems a bigger boost in seats, based on opinion polls as they stand. At first glance it seems to make a bit more sense than uniform swing, which gives Labour an extraordinary advantage. It makes a better stab at judging the impact of the Lib Dems, which some basic models tend to conveniently ignore. And it means you don’t end up with negative votes in some seats — one of the quirks in uniform swing.
You can read the first Ford counterattack here, Silver’s punchy reply, Ford’s more measured counterattck, and the fourth (and final?) round of the “Nerdfight” (where they find a lot to agree on). This is Silver’s take on why this whole debate is so interesting:
An election which looks as though it will feature both a major shift in votes away from one of the two main parties (as in 1997 or 1979) and a major surge by a third party, the effective sample size from past UK elections is basically zero. We’re literally in uncharted territory, and so claims that such-and-such model would have performed better on such-and-such occasions should be viewed skeptically. Moreover, while the differences between uniform swing and proportional swing models have generally been fairly small, they are quite dramatic here, which alone attests to the uniqueness of the election.
And for those who don’t love reading about regression analysis of incumbency advantages, here’s his intuitive explanation of why proportional swing makes more sense in this race.
I would posit one intuitive argument in defense of our approach. This has been a very dyanmic election. You have three viable parties. You have had, for the first time, televised debates. You have the strong possibility of a hung parliament and the politics surrounding that. You have some indications of a boost in turnout. You’ve had five years since the last election, during which an awful lot has gone wrong ranging from the financial crisis to the fallout from Britain’s participation in the Iraq War (something which was still ramping up in 2005) to last year’s MP scandals. We should expect things to get pretty jumbled up, and the factors that determined who voted for whom in the last election may not apply so much this year.
Generally speaking, uniform swing anticipates about the cleanest possible break from the previous election: a relatively efficient transfer of votes from Labour to LibDems and to a lesser extent Conservatives — one which might allow Labour to preserve a relatively high number of seats with a relatively small number of votes because of the advantageous way in which the votes happened to be distributed for them in 2005. The less orderly the break is, however, the less that 2005 can serve as a reliable benchmark (in some sense, the more reversion to the mean that there is), and therefore the more precarious Labour’s position. I expect a relatively ‘messy’ election, and therefore I continue to anticipate potentially large divergences from uniform swing.