SNP win the European elections - and the big losers are blundering Peter Kellner, and the BBC results programme

If I can trust the BBC website, 31 out of 32 councils in Scotland have declared, and it's now absolutely clear that the SNP are going to win the popular vote in the European elections by about 3% or so.  Their share of the vote is going to be roughly 29%, which we know must be good, because it's exactly what UKIP are getting across Great Britain, and the BBC keep telling us that's an absolutely bloody fantastic result.  In fact, it must be even better for the SNP than it is for UKIP, because you'd expect the SNP to suffer as an incumbent government, whereas UKIP are a 'free hit' protest vote party that have never governed anywhere at any time.

(UPDATE : It now looks certain that the SNP's share of the vote in Scotland will in fact be HIGHER than UKIP's across Britain as a whole after the London result is declared.)

And yet just a couple of hours ago, YouGov's Peter Kellner smugly informed the nation that Labour were heading for victory in Scotland.  Viewers of the BBC results show will have been forgiven for taking that wildly implausible claim seriously, because it was shorn of all context - it was the first time any Scottish results had been mentioned (other than a very brief comment from Kellner himself about Aberdeen).  Now, of course, the trends of early results can sometimes be unrepresentative, and if that was the case he could be forgiven for leading people astray.  But that categorically isn't what happened.  He immediately explained the percentage changes that had led him to conclude that Labour were winning, and they made no logical sense whatever - they in fact suggested that the SNP were heading for victory by about 2% or 3%.  It seems he hadn't even bothered to check the baseline figures from 2009 before coming into the studio.  The BBC's political editor Nick Robinson then had an immediate opportunity to correct an obviously flawed piece of arithmetic, but clearly he didn't know the baseline figures either, and instead eagerly seized on the latest concocted "blow for Alex Salmond" narrative.

When Mr Salmond appeared on the show later and pointed out that the SNP were in fact winning, and John Curtice implicitly accepted that to be true, I waited to hear the apology from David Dimbleby for the incorrect steer earlier.  I waited in vain.  An utter shambles, and it just ain't on.  We should have been watching a dedicated Scottish results programme - on past form that probably would have been less than perfect, but there's still no way it would have made such an unforgiveable mistake and then failed to come clean about it.

So just how good is this victory for the SNP?  The first point to make is that they've increased their lead over Labour by roughly 2% since their historic local election triumph two years ago.  Part of that can be explained by the fact that the SNP don't have to face the same challenge from rural independents in European elections that they do in local elections, but nevertheless this can probably be regarded as a slightly better result than 2012.  The most direct comparison is of course with the last European elections in 2009, when the SNP were assisted on their way to a decisive win by the unpopularity of Gordon Brown's Labour government.  Remarkably, in spite of the fact that Labour are now the challenging party at Westminster and should be sweeping all before it (especially in a supposed heartland like Scotland), the SNP have retained exactly the same share of the vote.  Brian Taylor has just claimed that the Nationalists will be disappointed that their vote "eased down" - well, as far as I can see, it's eased down by precisely 0.07%.  Let's get real here - they've replicated the 2009 result.

The other extraordinary claim that Brian Taylor made was that UKIP's snatching of a Scottish seat would somehow prevent Alex Salmond from pointing out that the political cultures of Scotland and the rest of the UK have totally diverged.  To be fair, Taylor did go on to concede that UKIP receiving 10% of the vote in Scotland and finishing fourth is still a rather different result from UKIP receiving 29% of the vote in the UK and finishing first, but he maintained that UKIP's seat would "complicate" the message.  Hmmm.  I'd suggest that's a distinction between the SNP being in a position to hammer their point home, and them being in a position to hammer the point so hard that it falls out the other end attached to a large slab of wood.

Incidentally, I had a chat on Twitter a few hours ago with an anti-independence troll who claimed that the SNP had "the UKIP bunting out".  I'll just pause briefly to note the irony of that remark in the light of Brian Taylor revealing that Labour, the Tories and the Liberal Democrats are all privately celebrating the fact that UKIP have taken a seat in Scotland, because they think that a minor scrap of success for a borderline-racist, anti-European party will somehow assist their anti-independence campaign.  That appears to be all that matters to them at the moment.

The other important question is how this result tallies up with recent opinion polls, and what it can tell us about which pollster is most reliable in respect of the referendum.  As it turns out, all of the pollsters who produced European voting intention figures - No-friendly and Yes-friendly alike - have overestimated the SNP.  That doesn't surprise me, because ever since PR was introduced for European elections all of the parties have consistently failed to break through the 30% barrier in Scotland, regardless of their showing in the opinion polls.  Probably a more realistic guide is the gap between the SNP and Labour, and on that measure YouGov were closest to the truth - but only in their sole full-scale Scottish poll, which underestimated the SNP lead by just 1%.  By contrast, YouGov's subsample results (including some very large subsamples) tended to wrongly suggest that Labour were winning in Scotland, which is yet another powerful indication that the sampling and weighting methods that YouGov use for Scottish respondents in GB-wide polls is totally misconceived.

But if this means that the methodology that YouGov use for full-scale Scottish polling is more reliable than some other pollsters (and that can only be a very tentative conclusion), is that good news or bad news for the Yes campaign?  Probably neither.  YouGov are still a No-friendly pollster, but only just - they've moved a long way towards the middle in recent months, and their most recent poll showed a No lead that was only 1% higher than the current average for all BPC pollsters.

Perhaps a more encouraging indicator is the fact that across Britain as a whole, online pollsters have done quite well.  Indeed, one of the most inaccurate polls of all was the highly-regarded ICM telephone poll for the Guardian, which put the Conservatives in first place for the European elections.  That's significant because online polls have tended to be more optimistic for the Yes camp than telephone polls (albeit telephone polls have been very few and far between in the referendum campaign, and have thus far only been conducted by Ipsos-Mori).

All in all we probably shouldn't jump to too many conclusions about the inaccuracy of certain polls, because there's a special challenge in trying to take account of differential turnout in an election where only about a third of people bother to vote.  A referendum in which perhaps 75% or 80% of people turn out to vote will be an entirely different proposition, and it could well be that pollsters who got it wrong tonight may yet prove to be more accurate in September.

The talk of the town among independence supporters in recent days was the question of which party was best-placed to deny UKIP a seat.  As I and a number of others have been predicting, the answer turned out to be the SNP.  When the d'Hondt formula is applied for the final seat (which will divide the SNP and Labour votes by three and the Tory vote by two), the SNP will be significantly ahead of Labour, the Tories and the Greens - and probably in that order, unless the Western Isles throws up a major shock.  Doubtless the Greens will still try to make the case that any extra votes for the SNP would also have been divided by three, and therefore tactical voting for the Greens could have closed the gap on UKIP at a faster rate.  But really that's a hopeless argument - if the Greens were to have any chance of claiming a seat, they needed to give some indication of being capable of breaking through the 10% barrier on their own merits, before taking into account the effect of the relatively small number of 'politically aware' people who were considering a tactical switch from the SNP.  Not a single opinion poll placed them anything like that high, and although the 8% share they claimed tonight is at the absolute higher end of my expectations, they were clearly never seriously in the running.  In relative terms they would have needed to increase their raw vote by a full third to overhaul the four parties that were ahead of them for the final seat, whereas the SNP would only have needed to increase theirs by less than a tenth to pip UKIP at the post.

I was firmly taken to task on a Hearts forum (ie. the football club) by an SNP supporter on Thursday, who suggested I was "incredibly patronising" for saying this -

"Maggie Chapman is a very fine candidate (and an asset for the Yes campaign), and if your number one priority is to have her as an MEP representing Scotland, then by all means vote Green - even though the chances of success look very slim indeed. But just be aware that it isn't going to stop David Coburn or UKIP."

I suppose I can see how that might look patronising when taken out of context, but in my defence I would say two things -

1) The point I made did turn out to be true.

2) When I wrote the post it was already deep into polling day, and it was becoming frustratingly clear on social media that some SNP supporters were being misled by James Mackenzie's spin on an unweighted YouGov subsample that put the Greens on 11%, and that had the SNP behind Labour.  In those circumstances, there didn't seem to be much point in pussyfooting around.

However, if anyone felt offended or patronised, I do apologise, and I'll try to be more tactful in future!

Final thought - this result is another hammerblow for the No campaign's resident referendum "seer", Ian Smart.  If his oracular insights into the trends shown by recent local by-elections had been on the right track, Labour should have won easily in Scotland tonight, rather than being beaten by the SNP in yet another national election.  Smart has also been relentlessly talking up the Scottish Tories recently - and that theory has fallen flat as well.  Ruth Davidson's troops have increased their share of the vote, but only by a measly 0.39%.