Just a Bit of Fun!

It would be remiss of me not to flag up th new look Scotland Votes site, newly beefed up to include details of the upcoming Westminster election, links to candidates' Twitter feeds, seat profiles, and everyone's favourite, the predictor, of which I include a screenshot:



Now, I love the predictor, as do most election geeks, as it enables us to live out our secret desire to be Peter Snow (or, for those of a more mature vintage, Robert McKenzie - I am quite sure that no one of any vintage at all has a burning desire to be Jeremy Vine), but there's a minor issue that I want to flag up - a small quibble, really, the equivalent of Craig Revel-Horwood berating Ricky Whittle and Natalie Lowe for coming down the stairs at the wrong time during their Argentine Tango on Saturday. All the same, look closely at the predictor:



Why that order of the parties? Why Labour, SNP, LibDem, Tory?

I ask simply because that order of the parties hasn't formed the basis of an actual Scotland-wide election result since 2001, and there have been five elections since then, all of which have yielded a result that wasn't this.

As such, I'd be fascinated to know what the mindset of the designer was, when they used an alignment of the parties which has only ever taken place once in a real election as the basis for their layout. Is it based on an expectation of the result? That would be an unwise approach. Is it based on the designers' suppositions of Scotland's political landscape? If so, they should have checked their facts first. Is it an institutional approach? The Weber Shandwick predictor for Holyrood - on the same site - puts the four parties in the same order. Again, this wasn't a reflection of the 2003 Election, on which the initial predictions were based (the 2007 poll not having taken place at the time), or the 2007 Election, whose outcome it was initially designed to project. It reflected an analysis that wasn't borne out by the electoral facts and that same analysis is being used again despite it being unclear what, if anything, it is based on.

And of course, if you attach any kind of significance or symbolism to that ordering, then three of the four main political parties have cause to feel aggrieved: the SNP can complain that the results of the last two Scotland-wide elections have been overlooked. The Tories can complain that the general trend of their third place has been overlooked: in the ten elections since 1994, the Tories have come behind the LibDems only twice - 2001 and 2005. And the LibDems would probably have the strongest cause for complaint of all three: on the best basis for comparison - the 2005 Election - they were the ones who came second behind Labour. That too has been overlooked.

Of course, the three parties would all envisage different orderings, so finding one based on past results would be difficult. A fairer approach would be to order the parties alphabetically: this allows an element of common sense and transparency as it's easy to see why the parties are laid out as they are. It also has the benefit of offering no symbolism whatever given the utter implausibility of the result turning out that way - the Tories haven't come first in Scotland since the 1979 European Election and haven't even reached second since 1992, while the SNP haven't been as low as fourth place in a real poll since 1987. It would seem more sensible to me.

But, frankly, the actual order doesn't matter too much, save as a possible barometer of Weber Shandwick's view of the political landscape. In truth, the predictor is, like all those that have gone before it, a decent enough tool in trying to project poll results to actual electoral outcomes (though, obviously, success can never be guaranteed), and a good way of passing a few spare minutes if you're of a persuasion that is as geekish as my own. As Peter Snow (or was it just Rory Bremner pretending to be Peter Snow?) used to say, it's just a bit of fun.