General Election 2017 projection and prediction, 25 May update

On Monday night Manchester was hit by a cowardly and brutal terrorist attack. It targeted young people who were together to enjoy themselves and have fun, just as previous terrorist attacks in Paris and Berlin did. It is yet another sign that those who want to destroy our democracy will stop at nothing.

We won’t back down. We will stand up united for freedom, life and joy.

Some have already tried to make political gain out of this terrorist act. Shame on them.

My thoughts with Manchester and all who lost loved ones or were injured in the attack.

And now back to business as usual, with a heavy heart.

On the reliability of polls

We already know that polls have a reputation for unreliability. The 2015 General Election is the best known example. The 2011 and 2016 Scottish Parliament elections also prove it. And hard evidence demonstrates that polls can often be off by much more than the margin of error. The methodology itself is sometimes flawed, especially the weighting of voting intentions relative to previous votes or other mysterious criteria. A pollster following British Polling Council guidelines doesn’t mean the results should be taken at face value. They all follow the guidelines and have all been wrong at some point in the past.

So let’s see what the projection of seats becomes on three possible scenarios:

  • Scenario A is based on last week’s polling average. SNP 44%, Conservatives 29%, Labour 16%, LibDems 6%.
  • Scenario B is SNP overestimated by 4%. Vote supposed to be SNP 40%, Conservatives 31%, Labour 17%, LibDems 7%.
  • Scenario C is SNP underestimated by 4%. Vote supposed to be SNP 48%, Conservatives 26%, Labour 15%, LibDems 6%.
A B C
SNP 51 41 58
Labour 0 1 0
Liberal Democrats 1 3 0
Conservatives 7 14 1

Note that even the worst case scenario would still give the SNP as many seats as Scottish Labour held from 2005 to 2015. Remember nobody then denied Labour was the dominant party and had a mandate to speak for Scotland. Whether Labour did it or not is a different story and I won’t discuss that right now.

Which seats would be impacted by each scenario is illustrated in the table below. I left out the 41 seats that would go to the SNP in any case. Food for thought…

*

At this moment I’m sticking to Scenario A for better or for worse. But B and C make sense too. Polls being off by 4% is by no means an extreme scenario; it’s roughly what we had on the FPTP vote for Scottish Parliament (SP) in 2016. Of course polls might be off by 2% ‘only’ (what we had on the FPTP vote for SP in 2011). But they might also be off by 6%, as they were in the 2015 General Election (GE).

SP 2011, GE 2015, and SP 2016 are all textbook examples of how polls being wrong impacted the projections and predictions; and, more importantly, the strategies of all parties (at least in 2015 and 2016). This is a serious issue. Especially in Scotland this year when a fifth of the seats should be treated as marginals.

And as usual turnout will be the key factor. You’ve been warned.

How did the polling data evolve?

Last week YouGov published their long awaited full Scottish poll, the first in more than three weeks. It is not as good for the SNP as subsamples of the UK-wide polls (42% compared to 45.6% average from the subsamples). The results are quite similar to previous full Scottish polls fielded in April. The YouGov poll accounts for 40% of my current ‘Poll of polls’ dataset so it will mechanically drag down the SNP vote share compared to what subsamples led us to expect.

Polling trends show the SNP first slightly up and then slightly down, leading to the same vote share as last week in my rolling average. Trends and rolling average have also moved slightly against the Conservatives and towards Labour.

*

*

How did the projection and prediction change?

I have slightly modified the SLLM (Safe, Likely, Lean, Marginal) graph and it has now become SLLMT. ‘T’ for ‘Tied’—the new category I have added for the few seats that are closer than marginal and could go either way by less than 1%. This week four seats qualify as tied: Edinburgh South, Edinburgh South West, Edinburgh West, Perth and North Perthshire.

*

*

And then you have my updated projection compared with two other models (Electoral Calculus or ScotlandVotes) and my updated prediction. This week I also came across Nigel Marriott’s predictor site. He has built several scenarios, some of which are more sophisticated than mine. His most elaborate takes into account a ‘Brexit effect’ and projects 53 seats for the SNP. Which I think is a bit overestimated, but I added his projections to the table anyway so you can form your own opinion.

*

*

I have altered my prediction to 52 SNP seats, which is possibly too cautious, or not. But I think the two Edinburgh seats I discuss below are squarely in the danger zone. Statistically they both rate as ‘Tied’ and I fear recent trends (including turnout) will see one of them gained by the LibDems.

Seats to watch

I usually don’t rely on Council elections to support my predictions but in Edinburgh they might be somewhat relevant. The maps below show the first party in each ward in 2012 (top) and 2017 (bottom). The SNP came out as the first party overall in 2017. But they were displaced to second or even third in wards that are entirely or mostly covered by the two Westminster seats I plan to discuss this time. This and other factors might very well influence the General Election result in both seats.

*
*

Edinburgh West

This one will have a much more interesting story than I expected. But not thanks to any of the main candidates. You’ll know more a few paragraphs down.

The Liberal Democrats have nominated Christine Jardine, a former BBC journalist and perennial candidate. She was also a Special Advisor to the Coalition government but her exact role remains unclear as available sources vary on this. She resigned her position as the Scottish LibDems Environment spokesperson last year after a row with Willie Rennie over gender equality. Nominating someone who stood for election five times and lost five times (and who is not even a local) seems to be a weird choice. But the LibDems obviously have high hopes for this seat. It was a LibDem seat for almost 20 years before Michelle Thomson gained it for the SNP. The Edinburgh Western seat in the Scottish Parliament was also held by LibDem MSPs except for the 2011-2016 term. The LibDems’ strong showing in the Almond and Drum Brae/Gyle wards on 4 May also hints this seat offers them a better opportunity for a gain than was originally thought.

The Conservative candidate is Sandy Batho, among others a Director of Edinburgh Zoo. Which might give new life to the old joke about pandas and Conservative MPs in Scotland. Batho stood for MP in Linlithgow and East Falkirk in 2015 and finished a distant third with 12% of the vote. He then ran for MSP in Edinburgh Western in 2016 and again finished third, this time with 14% of the vote. Conservative candidates did much better here in Council elections this month, so I expect a large increase of Conservative vote.

The SNP candidate is Toni Giugliano. He was selected last year as the SNP candidate for the Edinburgh Western constituency in the Scottish Parliament and lost to Liberal Democrat Alex Cole-Hamilton. This was one of the major upsets in the 2016 election. Giugliano gained 1,720 votes compared to 2011 but the SNP vote share went down 1.4% due to increased turnout. His defeat was also widely seen as the result of massive Unionist tactical voting, which might very well happen again in this GE.

Of course I can’t avoid a word about what you could call ‘the Michelle Thomson situation’. The facts have been widely discussed in the press, so I won’t go over them again. But the way she was deselected did not go down too well with some SNP members and of course Thomson’s own bitter reaction did not help either. How this will impact the GE result remains to be seen, but it certainly can’t benefit the SNP in a tied constituency.

Finally there’s the interesting story I mentioned earlier; that of Mark Whittet and his Scottish Independence Referendum Party. At first I did not know what to make of it but had a hunch it was suspicious and might be a plant. James McVean did some research on the man and the party and posted his findings on his Facebook page. Guess it speaks for itself and voters will not fall for it.

This seat is statistically a tie between the SNP and the Liberal Democrats, with the SNP slightly ahead. But statistics can be wrong. I have given it much thought, trying to figure out which factors would influence voters most. Quite reluctantly and contrary to what my model projects…

I rate it as a LibDem gain from SNP
LibDems 34% (+1%), SNP 31% (-8%), Conservatives 27% (+15%), Labour 7% (-5%)

Edinburgh South West

This seat is high on the Conservatives’ target list. It shouldn’t be as the SNP gained it from Labour in 2015 on a 16% margin with the Conservatives a distant third. But Scottish Parliament elections a year ago and Council elections this month have been a mixed bag here for the SNP and pretty good for the Conservatives.

The Westminster constituency is mostly covered by two Holyrood constituencies: Edinburgh Pentlands and Edinburgh Southern. Last year the SNP held Pentlands by 7% over the Conservatives. But the SNP also lost Southern to Labour by 3%, with the Conservatives unexpectedly coming out first on the list vote in this constituency.

This month’s Council elections saw the Conservatives come first in three of the wards that are within the constituency, with a very significant increase of their vote share. The SNP came first in only one. In 2012 it was a four way split with SNP, Labour, Conservatives, and Greens coming first in one ward each. So recent voting trends definitely favour the Conservatives and put the SNP squarely in the danger zone here.

The Conservative candidate is Miles Briggs. He was elected to the Scottish Parliament in 2016 as a list MSP for the Lothian region after finishing third in the Edinburgh Southern constituency. Previously he stood twice for MP and once for constituency MSP, and lost all three. He is now the Conservative spokesperson for Mental health and public health in Holyrood. Briggs openly calls for Unionist tactical voting, urging Labour and LibDems voters to switch to the Conservatives as it is probably his best chance to defeat the SNP here. And, apart from the obvious ‘no to independence’ mantra, he does not mention the Conservative manifesto too often. Probably because the demographics of the constituency (younger and with higher education than average) don’t make it standard Conservative turf. And also probably because, even if the constituency voted slightly more ‘Leave’ than the Edinburgh average, it was still a resounding 72% ‘Remain’. Even the support of former UKIP voters (2% in 2015) might not be enough to get him first past the post this time.

Labour have nominated Foysol Choudhury, a businessman and prominent figure of the Edinburgh Bangladeshi community. He’s quite high-profile as co-founder and Vice Chair of the Edinburgh Mela among others. This might bring him some extra votes but certainly not enough to do better than third place in what was once Alistair Darling’s seat. Liberal Democrat candidate Aisha Mir is a complete unknown to me, and stands as the sacrificial lamb in a constituency where LibDems lost their deposit in 2015.

Finally Joanna Cherry QC, the sitting MP and SNP candidate. Before becoming an MP, Cherry had a distinguished career as a lawyer, culminating in her appointment as Queen’s Counsel in 2009. She first attracted public attention in 2014 as co-founder and convener of ‘Lawyers For Yes’. She’s one of the most visible SNP frontbenchers as spokesperson for Justice and Home Affairs. She’s also a prominent advocate of LGBT rights and women’s rights. She has also been deeply involved in local issues since the early days of her term, like fighting the Allan Park Crescent development project in a joint effort with the Residents Association and then-MSP Jim Eadie.

She gained the seat from Labour in 2015 after Alistair Darling decided to stand down, on a 23% swing. Despite her strong personal status and campaigning skills, her re-election is not a given and will obviously not be as easy as it was two years ago. But the SNP will obviously spare no effort to have her re-elected and she has already raised enough money to crowdfund her campaign up to 60% of the legal spending limit. She will also benefit from the absence of a Green candidate this year, which may bring her up to 2% extra votes.

Statistically this seat comes out as a tie between the SNP and the Conservatives, with Conservatives slightly favoured. But I will go out on a limb again here and…

I rate it as a SNP hold
SNP 38% (-5%), Conservatives 37% (+17%), Labour 21% (-6%), LibDems 3% (-1%)

Next week: Moray, Perth and North Perthshire

Saor Alba gu bràth.

Jeremy Blackwell, 25 May 2017

Jeremy BlackwellJeremy Blackwell is an analyst and statistician living and working in Edinburgh. You can follow him on Twitter at @WeAreThe59.

One thought on “General Election 2017 projection and prediction, 25 May update”

Leave a Reply

Your e-mail address will not be published. Required fields are marked *