Why the swingometer lost its sway

Guto Thomas
BBC Wales political correspondent

Peter Snow's swingometer was a feature of BBC election shows

Swingometers have been a key part of election night programmes in the United Kingdom for decades.

With each declaration as key seats hovered with intent, the swingometer in the hands of pundits like Peter Snow gave a good idea of how the government of the day was faring.

So what will "swing" be able to tell us about the fortunes of Labour's Welsh Assembly Government administration in the early hours of 4 May?

The answer is pretty much nothing.

A rich and varied "patchwork quilt" of different swings is more than likely, with different parties winning votes against each other for different reasons in different parts of the country

The problem is that "swing" was devised in an era of two-party politics, when the vast majority of people across the UK (around 90%, in fact) voted for Labour or the Conservatives.

This meant that in most marginal seats, the contest was a simple two-horse race.

And while such contests can and still do exist, the nature of politics in Wales has changed the rules of the psephological game.

In Wales in 2007, there is now an established four-party system - with the addition of Plaid Cymru and the Liberal Democrats, as well as a number of significant independent candidates.

These factors immediately weaken the ability of a simple swingometer to explain or anticipate political movements around the country.

Most elections these days are no longer a two-horse race

But as well as this, there is really no expectation among observers that there will be any kind of uniform swing from one particular party to another.

In fact, the expectation is that in terms of the contests to win seats in the Welsh assembly, a rich and varied "patchwork quilt" of different swings is more than likely, with different parties winning votes against each other for different reasons in different parts of the country.

And so what we are more likely to see during the campaign itself are 40 separate by-elections being fought in every constituency in Wales.

This means that any specific swings will not be of much use in terms of predicting any general outcome.

Swing, it seems, is dead.

Fundamental change

For the same reasons, the reliance of the past on the "share" of the vote between the parties is much less likely to retain its reputation as an accurate measure of success or failure.

In Wales, the crucial question will in fact be about how many of their core voters the political parties manage to mobilise.

Pundit Robert McKenzie, when lo-tech swing was king in the 1970s

Moreover, this is an election that won't be determined so much by the total level of turnout achieved across Wales, but by the differential turnout achieved by each one of the parties, in specific key constituencies, and in terms of the second vote, across each one of the five electoral regions.

This new electoral orthodoxy is illustrated by the way that candidates have been encouraged by all the political parties to identify the local issues which most concern voters, such the threat of job losses, or proposals to reconfigure, downgrade or close the local school or hospital.

With limited evidence of the intention of any one of the parties to conduct anything approaching a sustained, daily, national campaign, then this signals a fundamental change in the way the parties hope to re-engage with the voters.

Whether this will actually work - or whether the level of public dissatisfaction with politics will remain dominant in this election - is impossible to predict.

Regardless of what happens in 2007 however, "localism" it seems is definitely here to stay, as the backbone of a new Welsh politics.