Britain's favourite impartial Liberal Democrat election expert Mike Smithson is BACK - and he's writing letters again

Thanks to Tris of Munguin's Republic for alerting me to this.

It's truly inspiring that an impartial "polling and elections expert" would take the time to write letters to ordinary voters up and down the country, with no particular motive other than a simple urge to share his expertise with those less insightful than himself.  He would never dream of telling you how to vote, in spite of the fact that he's a current Liberal Democrat party member, a former Liberal Democrat county councillor, and a former Liberal Democrat candidate for parliament.  It's just pure coincidence that the "information" he provides is invariably that some disaster would befall the voter if they do anything other than vote Liberal Democrat.  And nobody should be cynical about the fact that the microscopically small print at the bottom reveals that the letter was in fact published and promoted on behalf of the Liberal Democrats.  They're simply providing an invaluable public service without the slightest thought for their own advantage.

The wording is very similar to the letter Smithson put his name to in the 2017 election, but I notice it's less constituency-specific - it just uses language like "in this area", whereas last time it was spelling out "voters who want to stop the SNP in East Dunbartonshire should vote for Jo Swinson and the Liberal Democrats".  In one sense that's lazier, but it's also probably a sign that he's just agreed to let the party use a general template anywhere they want - in other words his advice to voters is whatever the Liberal Democrats want it to be.

Smithson never had that much credibility as an "expert" to begin with - he's made a series of near-comical howlers over the years, including confident predictions that Kitty Ussher would become leader of the Labour party ("remember you read it here first!"), that the Liberal Democrats would hold the Gordon constituency in 2015 (they lost to the SNP by almost 9000 votes), that the Fixed Term Parliaments Act made early elections nigh-on impossible (that's been proved wrong twice in the space of two short years), and worst of all that the Lib Dems would be making a historic error if they did anything other than go into coalition with the Conservatives in 2010.  But to the extent he's got any reputation left to defend, he seems happy enough to squander it completely by being seen to once again pump out industrial-scale propaganda for his own political party, and indeed by allowing that party to put words in his mouth.

You can tell from the framing of the letter that it's matched to the profile of the individual voter - if they're a natural Tory they'll be told that "the data" shows they have to vote Lib Dem to stop Corbyn, and if they're a natural Labour supporter they'll be told "the data" shows they have to vote Lib Dem to stop Johnson and a Hard Brexit.  Essentially if you don't vote Liberal Democrat in this election, you're anti-science.

*  *  *

I have two more constituency profiles in today's edition of The National - this time it's Dumfries & Galloway and Cumbernauld, Kilsyth & Kirkintilloch East.

Options for pro-independence voters in Kirkcaldy and Cowdenbeath

I almost banged my head against the wall in frustration when I discovered that the SNP are not even going to be campaigning in a key target seat.  There are a very limited number of constituencies that the SNP have a high probability of gaining, most of them Labour-held, and I don't think they can afford to be giving Labour a bye in even one of them.  Candidates should be vetted properly before the deadline for nominations, and after that point they should be able to expect that the party will stand by them, unless they turn out to be serial killers or something.  I've been told that the SNP don't have the manpower to check every single social media post going back years, but if we're talking about something so obscure and old that it wasn't found during vetting, that might be a pretty good sign that "political death by ancient social media history" wasn't actually appropriate in this case.

The problem goes deeper than the fact that Kirkcaldy and Cowdenbeath is in all likelihood going to be needlessly held by Labour.  There are only 59 constituencies in Scotland, which means that each one (apart from the two smallest ones) have just under 2% of the national population.  This is going to negatively affect both the SNP's national vote share and the overall national vote share for pro-independence parties.  Before today, I would have criticised the Greens' foolishness for standing in a very tight Labour-SNP marginal like this, but it's just as well they've done so, because at least it gives independence supporters in Kirkcaldy and Cowdenbeath an additional option if they feel they can't back Neale Hanvey.  In all honesty, if I was in that constituency, I'd still vote for Hanvey, because it'll count towards the overall SNP vote tally, and he's probably not going to win anyway.

*  *  *

UPDATE: I've slightly altered the wording of this blogpost, because I noticed there was quite a bit of confusion in the comments section about what has happened and what the implications are.  To be clear: although the SNP have withdrawn support from Neale Hanvey, he remains on the ballot paper as the SNP candidate, and any votes cast for him will be fully counted by the media as SNP votes.  Therefore, if you want to maximise the SNP's national vote share, the best option is to vote for him.  Additionally, it's almost certain that he'll receive more votes in the constituency than the Green candidate (simply because he'll have "SNP" by his name), so if you want to vote tactically for the pro-independence candidate most likely to win, Hanvey is the correct choice.  Defeating Labour is probably a long shot now, but he's the only person who can conceivably do it.  Interestingly, on the betting markets, Labour are only marginal favourites to win the seat at the moment, so who knows.

In the event that Hanvey is elected, he would in the first instance become a pro-indy independent MP, because he would automatically be deprived of the SNP whip.  However, it's theoretically possible that he could become an official SNP MP after a few weeks or months if his suspension is eventually lifted.

*  *  *

I have two more constituency profiles in The National - this time it's Angus and Inverclyde.

Bombshell telephone poll gives SNP mammoth 18% lead over the Tories - and puts support for independence at 50%

Scottish voting intentions for general election (Ipsos-Mori):

SNP 44% 

Conservatives 26% 

Labour 16% 

Liberal Democrats 11%

Greens 2%

Seat projection: SNP 48 (+13), Conservatives 6 (-7), Liberal Democrats 4 (n/c), Labour 1 (-6)

Should Scotland be an independent country?

Yes 50%

No 50%

The seat projection is obviously markedly better for the SNP, and worse for the Tories, than we saw from the YouGov model last night.  So which is likely to be closer to the truth?  The YouGov projection is more sophisticated, because it takes account of different trends in different areas - for example that there may be more of a swing against the Tories in Remain-friendly East Renfrewshire than in Leave-friendly Banff & Buchan.  By contrast, the projection from the Ipsos-Mori poll is very crude and based on assumptions of a uniform swing.  But the fundamental reason why the SNP would get more seats if Ipsos-Mori are right is that the poll shows the party with a higher share of the vote than YouGov detected.  If it was possible to run a subtler projection model based on Ipsos-Mori's figures, you'd probably still see the SNP gaining more Tory seats than YouGov predicted.

So it really just boils down to whether you think Ipsos-Mori or YouGov are estimating the vote shares of each party more accurately.  There was a time when we'd have assumed that Ipsos-Mori's data collection method by telephone was bound to produce more accurate results than YouGov's online approach, but what happened in the EU referendum raised a few question marks over that.

Incidentally, the fieldwork dates don't explain the difference - the Ipsos-Mori poll was conducted over roughly the same period as the fieldwork for the YouGov model.

More to follow...

I have four more constituency previews in today's edition of The National - this time it's East Lothian, Coatbridge, Chryston & Bellshill, Glasgow South and Glasgow Central.

Long-awaited YouGov projection puts SNP on 43 seats - and suggests Jo Swinson is only 5% ahead in East Dunbartonshire

The YouGov MRP figures are finally out, and it's going to take me a while to wade through the data properly.  But like a lot of people, the first seat I checked was East Dunbartonshire, because there's been a long-running difference of opinion about whether Jo Swinson is in trouble, or will hold her seat easily.  The answer seems to be somewhere in between the two extremes.

East Dunbartonshire estimate:

Liberal Democrats 38%

SNP 33%

Conservatives 17%

Labour 8%

Greens 3%

So she's in the danger zone, but she'll probably hold on.  However, YouGov agree with the earlier Better for Britain projection in saying the SNP have the lead in the Lib Dem seat of Caithness, Sutherland and Easter Ross.  That would be a reasonably big shock.

The SNP are projected to win 43 seats overall, which would be a perfectly good result if it actually happens, but it just worries me a little because the projections in 2017 were showing much the same thing, and we all know what happened next.  If there's another late Labour surge, things could still go wrong.  But one point of encouragement: some of the seats the Tories are projected to hold look quite tight.  So things could yet go wrong for them as well, and if that happens there'll only be one beneficiary: the SNP.

*  *  *

On a completely different topic, a free Scottish Gaelic course is now available on Duolingo.  That could have a transformative effect on the language's prospects - it puts it in the shop window, and anyone with the urge to begin to learn it can now do so, wherever they are in the world.  Read more HERE.

MRP analysis shows SNP on course for a net gain of nine seats

The pro-Remain group Best for Britain have published a new seats projection based on an MRP analysis of a large number of polling interviews - similar to the method YouGov used in 2017, and that proved to be more accurate than conventional polling results.  The headline figures aren't surprising at all -

Conservatives 366
Labour 199
SNP 44
Liberal Democrats 17
Plaid Cymru 4
Greens 1

But what's very surprising are the individual seat projections for Scotland, some of which are much better for the SNP than expected, and some of which are much worse. For example, Labour are shown as being slightly ahead in Glasgow North-East. That doesn't necessarily mean that they're projected to win the seat, just that it's too close to call, but if that's typical of what's happening in Glasgow it would be very concerning. On anything even approaching a uniform swing, Glasgow North-East ought to be an easy SNP gain. But on the other hand, the SNP are shown as being fractionally ahead of the Tories in Ochil & South Perthshire, and only slightly behind them in the likes of Banff & Buchan and Gordon. Those are better results than you'd expect if the Scottish Tories really are back up to the high 20s - although admittedly the data is probably skewed by the long spell covered by the fieldwork. But the dates can't explain why the SNP appear to be virtually level-pegging with Jo Swinson in East Dunbartonshire - that's the most striking finding of the lot.

I'm not sure whether this should cause us to reevaluate our assumptions completely, or whether it's just dud data or dud methodology. I'll be interested to see what YouGov's own analysis shows.

*  *  *

Click here for a handy list of SNP election crowdfunders.

If the BBC run a 'fact check' on a BBC interview and find that only the interviewer got his facts wrong, shouldn't an apology automatically follow?

You may already have seen this, but the BBC ran a "fact check" on the Andrew Neil interview with Nicola Sturgeon on Monday night, and the results were rather amusing.  They found that each and every claim Ms Sturgeon made was accurate, but they were obviously deeply frustrated by that discovery and couldn't leave it there, so in each case they needlessly added extra text along the lines of "but what she could have additionally said was..."

For example, in respect of her claim that almost all polls show an increase in support for independence, they admit that's true, but add that: "Excluding "don't knows", the average of polls this year has been 51% for No to 49% for Yes - extremely close but still marginal support for remaining in the union".  Which is fine, but as Nicola Sturgeon didn't claim at any point that there's a pro-independence majority in the polling average, it's unclear what that gratuitous observation is doing in a purported "fact check" piece.

It gets better, though, because the article does tacitly identify one genuine inaccuracy from the interview.  The only snag is that it was Andrew Neil who was guilty of the inaccuracy, not Nicola Sturgeon.  This is what the fact check says about whether Scotland would need its own currency to qualify for EU accession -

"The EU rules for countries which want to join the EU do not explicitly say that a country has to have its own currency before it is allowed to become a member"

Which directly contradicts a claim made repeatedly by Andrew Neil in his questioning.  When Ms Sturgeon pointed out that he was misleading his viewers, he talked over her loudly and refused to even acknowledge she had challenged him on the point.  Indeed, the most bizarre part of the whole interview was when he informed a bemused Ms Sturgeon that she had "accepted" that a currency is required, even though every single viewer of the programme had just heard her vociferously say the complete opposite!

If the BBC run a fact check on a BBC interview with a leading politician, and find that only the interviewer got his facts wrong, you'd expect the article to say something like -

"This means that Mr Neil was not strictly correct in the claims he made in his questioning, and the BBC would like to extend our apologies to Ms Sturgeon and to viewers for this inadvertent error".  

As they haven't done that, and as they've essentially just ignored the blatant contradiction between statements made in the fact check and claims made by Mr Neil in his questions, what conclusions are we entitled to draw?

*  *  *

Click here for a handy list of SNP election crowdfunders.

My namesake's attempt to rig the next independence referendum by giving the No side a "sabotage-by-boycott" option

When I saw that James Kelly MSP was trying to add a 50% requirement to the Referendums Bill, I assumed he had jumped the shark completely and was saying that 50% of the total registered electorate would have to vote in favour of independence to make the result valid.  To put in perspective just how ridiculous that would be, if there was a 75% turnout, and if the result was 66% Yes, 34% No, the electorate would be deemed to have voted No.  And it's not at all fanciful that unionists would seek to load the dice in such an extreme way - only a couple of months ago, an anti-independence group in the north-east (with which Professor Hugh Pennington is involved) openly called for a two-thirds majority requirement.

But it turns out that Mr Kelly is playing a slightly subtler game.  His amendment would simply require that half of the registered electorate takes part in the referendum, irrespective of whether they vote Yes or No.  If they didn't, any Yes victory would be voided.  Superficially, that doesn't sound quite so unreasonable, because of course the turnout in the 2014 referendum was 85% - so surely there'd be no danger of a referendum of any real importance falling below the 50% threshold?

Well, here's the thing.  In 2014, in spite of that record-breaking turnout, and in spite of the fact that No won by a "decisive" margin (it must be true, because the BBC used that word about 70,000 times the following day), less than 47% of the registered electorate actually voted No.  If there had been a 50% turnout requirement for a No vote to be considered valid, pro-independence groups could have sabotaged the result by urging their supporters to abstain rather than vote Yes.

But of course Mr Kelly's intention is that only a Yes majority would be voided by his 50% rule.  A No majority on less than a 50% turnout would effectively be respected.  So if a future campaign was going against them, the No side would have the "sabotage-by-boycott" option open to them, but the Yes side wouldn't.

As the chap in the Question Time audience said the other night, the UK has become like Hotel California - we can check out any time we like, but we can never leave.

*  *  *

I have two more constituency previews in today's edition of The National - this time it's Stirling and Glenrothes.  (And these are honestly written by me, and not by the Labour list MSP for Glasgow.  We all have a cross to bear, and mine is my name...)

*  *  *

If you know of any independence and/or SNP supporters who might not have registered to vote yet, make sure they know that TONIGHT is the deadline.  They can register HERE.

Are there signs that the Tory lead may be weakening slightly?

I don't know about anyone else, but I'm finding it quite hard to work out what to wish for in this campaign.  I don't want there to be a Tory government, and especially not a majority Tory government, which on the face of it means that I should want the GB-wide Tory lead to start dipping.  But that would mean a Labour recovery, and we know from what happened in 2017 that it's very difficult to keep a Labour surge "quarantined" in England and Wales - it's highly likely to spill over into Scotland, and at the end of the day it's Labour that poses the biggest threat to SNP dominance.  Not only are most SNP seats in former Labour heartlands, there's also the problem that any gain in Labour support is like a "six point swing" in football - it's bound to be partly at the SNP's expense, and therefore erodes the SNP lead much faster than any gains for the Tories.

So far the chances of that happening have looked remote, but today for the first time I'm beginning to wonder.  A new poll in Wales has shown a really significant jump in the Labour vote, way beyond anything that can be explained by random sampling variation.  Something very similar happened just before the 2017 election.

Welsh voting intentions (YouGov):

Labour 38% (+9)
Conservatives 32% (+4)
Plaid Cymru 11% (-1)
Liberal Democrats 9% (-3)
Brexit Party 8% (-7)
Greens 1% (-2)

OK, Wales is not the UK, and we haven't seen anything similar in Britain-wide polls yet. But there is a Britain-wide ICM poll today reporting a smaller swing to Labour. Whether that's just margin of error noise, or the start of something more significant, remains to be seen.

Britain-wide voting intentions (ICM):

Conservatives 41% (-1)
Labour 34% (+2)
Liberal Democrats 13% (n/c)
Brexit Party 4% (-1)

(Note: For some reason the SNP are often edited out of the results summaries for GB polls, but on past form with ICM they're probably on around 3%.)

It's just conceivable that a 7-point Tory lead might not be quite enough to win Boris Johnson an outright majority - it depends on the geographical distribution of votes, and on the extent of pro-EU tactical voting.

Would we prefer to have 45 SNP MPs plus a Tory majority government, or 35 SNP MPs with genuine influence over a Corbyn minority government?  I suppose I'd prefer the latter, but if Labour reduce the SNP's haul of seats without actually removing the Tories from office, that would be the worst of all worlds (as we discovered two years ago).

*  *  *

I have two more constituency previews in today's edition of The National - this time it's Dundee West and Caithness, Sutherland and Easter Ross.

*  *  *

If you know of any independence and/or SNP supporters who might not have registered to vote yet, make sure they know that tomorrow is the deadline.  They can register HERE.

*  *  *

Click here for a handy list of SNP election crowdfunders.

First full-scale Scottish poll of the campaign suggests SNP vote has INCREASED

Well, it's only taken the best part of a month, but at last we have the first Scottish poll of the campaign - and it's a Panelbase poll in the Sunday Times.  Better still, the fieldwork took place after Tuesday, so we don't have to put a question mark over the results due to the rigged debate on ITV.  The trends are pretty much in line with what YouGov subsamples have been suggesting - ie. that the SNP vote has held up, that the Tory vote has sharply rebounded at the expense of the Brexit Party, that Labour have flatlined and that the Lib Dems have drifted downwards.

Scottish voting intentions for UK general election (Panelbase):

SNP 40% (+1)
Conservatives 28% (+7)
Labour 20% (+1)
Liberal Democrats 11% (-2)

The seats projection published in the Sunday Times suggests the SNP would take 41 seats on these figures, and that the Tories would hold 12, losing only Stirling.  Because of the media's tendency to treat the last poll they saw as gospel, we're now going to hear the mantra that "the polls" show that the Scottish Tories have completely ridden out the storm and are going to hold their seats - and that'll be somewhat misleading to say the least.  Whether the Tories hold seats depends not only on their own share of the vote but on the SNP's.  Even if the Tory vote is exactly the same as it was in 2017, there'll still be a net swing from Tory to SNP if the SNP's own vote increases.  And, indeed, Panelbase suggests the SNP's vote is up three points on 2017.  But here's the thing - Panelbase have not been the most favourable pollster for the SNP in recent times.  The last five Panelbase polls before tonight had the SNP somewhere between 37% and 39%, whereas the last three YouGov polls have had the SNP on either 42% or 43%.  If we imagine for the sake of argument that YouGov are closer to being right, that could mean there'd be a swing of as much as 4% from Tory to SNP, and that would bring considerably more than one Tory seat into play for the SNP.

And in any case, all of this is assuming a uniform national swing, which is very unlikely to happen.  As Professor Curtice himself points out in the Sunday Times piece, if Panelbase are basically right but there is a local swing that is 2% greater, another four Tory seats could fall.

There's also an independence question in the Panelbase poll, and it shows Yes support continuing to hold firm at an elevated level.  49% support for independence is several points higher than the average in Panelbase polls last year, and the one-point drop since the last poll is statistically insignificant.

Should Scotland be an independent country?

Yes 49% (-1)
No 51% (+1)

*  *  *

Click here for a handy list of SNP election crowdfunders.