Panelbase, 18th-20th June 2019

Panelbase and the Sunday Times seem to have settled into a pattern of monthly polling, with this June poll being their fourth month on the trot. We’re now at the point it’d be really quite lovely for another paper/pollster partnership to pop up and do so too, as Panelbase is now completely dominating Scottish polling output. Anyway, now I’ve given this gift horse’s mouth a good look, let’s get down to the data.

Display format;

Holyrood Voting Intention and Projection (Tracker)


  • SNP – 39% (+2 / -3)
  • Con – 20% (+1 / -3)
  • Lab – 16% (-2 / -3)
  • LD – 10% (+2 / +5)
  • Grn – 7% (nc / nc)
  • Brex – 6% (nc / +6)
  • UKIP – 1% (-1 / -1)
  • ChUK – 1% (-1 / +1)

Although the SNP are still polling below their 2016 result, this is the highest they’ve seen for the list vote since the snap GE in 2017. The Lib Dems will also be very happy to be in the double figures, though that hasn’t been entirely unprecedented for them. By contrast, things look rather grim for Labour who come in with their second lowest figure since the snap, and whilst the Conservatives have bounced back a little bit since last month’s poll, they are still clearly suffering for that solid Brexit Party support. As we’ve come to expect from Panelbase the Greens are completely static – though their EU polls vastly underestimated the Greens, so I’m leaning towards viewing Panelbase as being the outlier on this front.


  • SNP – 42% (+1 /-5)
  • Con – 20% (nc / -2)
  • Lab – 16% (-2 / -7)
  • LD – 11% (+3 / +3)
  • Brex – 7% (nc / +7)
  • Grn – 3% (nc / +2)
  • UKIP – 1% (nc / +1)
  • ChUK – 0% (-1 / nc)

The constituency vote reads pretty similarly to the list vote; very good for both the SNP and Lib Dems, grim for Labour, poor for the Conservatives, and pretty static for everyone else.

Projected into seats it might look like;

  • SNP – 64 (+1 / +1)
  • Con – 24 (nc / -7)
  • Lab – 20 (-2 / -4)
  • LD – 11 (+3 / +6)
  • Grn – 6 (nc / nc)
  • Brex – 4 (-2 / +4)

Keeping up the recent trend of SNP seat projections in the 60’s, here the SNP would pick up one more seat than in 2016, bringing them just one short of an overall majority. The Lib Dems would more than double their seats, even being projected to pick up another constituency, whilst this is a third poll running indicating a total Labour wipeout from constituencies. Although the Brexit vote is static vs last month they drop a couple of seats in this projection; effectively, the Lib Dem surge knocks them out of seats in the Glasgow and Mid & Fife regions.

As is now my habit, let’s also look at that translated to an imagined more proportional system a la Scandinavian countries (vs AMS projection);

  • SNP – 56 (-8)
  • Con – 25 (+1)
  • Lab – 20 (nc)
  • LD – 12 (+1)
  • Grn – 9 (+3)
  • Brex – 7 (+3)

As ever it’s the SNP that are over-represented by the current system whilst everyone else except Labour would pick up more. On these figures though they are still pushing the more proportional system enough to scrape a pro-Independence majority with the Greens.

Westminster Voting Intention (Tracker)

  • SNP – 38% (nc / +1)
  • Con – 18% (nc / -11)
  • Lab – 17% (-2 / -10)
  • LD – 13% (+3 / +5)
  • Brex – 9% (nc / +9)
  • Grn – 2% (-1 / +2)
  • UKIP – 1% (nc / +1)
  • ChUK – 0% (-2 / nc)

The big story from the Westminster portion of the poll is that, having broken back into double figures for the first time last month, the Lib Dems are further benefiting from a Brexit bounce bringing them up to 13%. With both the Conservatives and Labour on their joint worst results since the 2017 election, that’d likely see the Lib Dems take as many or more seats than the two of them combined – what a sensible system First Past the Post is. Yes, of course, the SNP therefore end up with the overwhelming majority of Scotland’s MPs again despite being up only marginally on 2017, but that’s par for the course at the moment.

I don’t normally include 0% showings but you’ll notice I’ve done so here for ChUK, as well as up in the Holyrood constituency VI. They just scraped past the 0.5% mark to round up to 1% for Holyrood regional vote, but it looks like the artists formerly known as Change UK are simply going to sink beneath the waves, following a poor EU Election result and the resulting loss of 6 of their 11 MPs at Westminster.

Constitutional (Independence Tracker)

It’s possible they’ll come back an add to the tables later, but at the moment it looks like it was just Independence that was asked about this time on this front. Standard Independence question;

  • No – 48% (-1)
  • Yes – 46% (+1)
  • Don’t Know – 6% (nc)

Excluding Don’t Know;

  • No – 51% (-1 / -4)
  • Yes – 49% (+1 / +4)

“Not much movement on Independence” has been the mantra of these posts for some time, and whilst that still holds true, the “not much” has been creeping towards independence. At 46% in the raw data and 49% when excluding Don’t Knows, this is the highest support for Independence that Panelbase have found since that anomalous Yes majority in the immediate aftermath of the Brexit referendum, as well as one of the highest figures of any pollster in that time. On purely universal swing versus 2014, that’d see 12 Yes voting areas versus 20 No.

They also asked how people would vote if Boris Johnson was Prime Minister;

  • Yes – 48%
  • No – 44%
  • Don’t Know – 8%

Excluding Don’t Knows;

  • Yes – 53%
  • No – 47%

As you’d expect that cause a bit of a stir on social media, but as ever I’d urge against reading very much into that kind of hypothetical question. Remember that in the past polling tended to suggest a Conservative majority in 2015, or a UK-wide Leave vote against a Scottish Remain would tip people towards Independence, yet both of those things happened and neither have led to any lasting increase in support for Independence.

Finally, there are a range of “How good or bad a job is X doing as Y” questions about various leading politicians, including some who where then in the running for Conservative Leader and Prime Minister. It’s worth having a week look at the tables for those, but they are beyond the scope of Ballot Box Scotland.


  1. Many thanks BBS interesting graphics. I ran some research recently on our tweaked version of D’Hondt. Putting aside my Party loyalty, I felt that our current system is not democratic enough. The hefty weighting on the Lists v Constituency vote penalises too heavily against the most popular Party in my view and as such distorts the democratic vote. I tried a simple experiment. I took out the Constituency voting all together then just applied weighting to the popular vote alone by Region. Still halving the popular vote each time down the List as we do now. Then I added the Regional list returns to the Constituency to complete. This was simple and quite satisfying and I felt more representative for the voters … So not the Scandinavian style just list, still a mix of both….. give it a try.

Comments are closed.