Poll Analysis: Redfield & Wilton 30th of April – 2nd of May 2023

After a brief lull in polling – but not dramatic political moments – Redfield & Wilton are back with the third entry in their monthly tracker (link to tables). I’d been quite publicly exasperated about the last one, which showed an astronomically high Lib Dem vote share that was completely outwith the realms of credibility. It was so disheartening to me that it was undermining the whole poll, and I’d said I would just stop including Redfield & Wilton in BBS output entirely if such oddities continued.

Whether they re-evaluated some of their weighting or methodology since then, or just by coincidence, this month looks less daft, so I don’t have to do the relative nuclear option of just ignoring the whole thing – much to my relief, because it’d have been a pain to cut the previous ones out the averaging! As a bit of a precaution however, for (arbitrary, 5-poll) averages going forward I’m only allowing one R&W poll to be in it, skipping over their older ones in favour of other pollsters.

The previous Redfield & Wilton covered the 31st of March – 1st of April. Changes are shown as (vs that poll / vs last election).

Regional Vote

This is the properly juicy bit of the poll, and the portion that should be reported most widely – the first Labour leader over the SNP since before the 2014 referendum. From the point at which Peter Murrell was arrested I’d settled into a pretty firm conviction that we’d see a Labour lead at some point this year, and here it is. There’s a lot to get into here!

Firstly, although this is Labour’s first lead in nearly a decade, it’s not actually their best figure in this vote since then – in fact, it’s 3% shy of their recent peak. It is however significantly higher than the last poll. That is in part, I think, because the Lib Dems are down by as much as Labour are up. That still leaves this firmly on the upper end of Lib Dem polling, but 10% is believable in a way 13% isn’t – and I did think it was Labour suffering in the last poll as a result.

Looking at the SNP, this isn’t merely their worst vote share of the term, or of the post-IndyRef period, but the worst since they entered government in 2007. Dropping 5% is a really sharp fall, leaving them with only a quarter of the vote overall. Their worst since taking office prior to this had been a few polls saying 26%. Funnily enough, one of those was on the 22nd of February 2011, and was followed by their majority-securing 44% a couple of months later. I highly doubt they can expect a repeat of that, but it is a reminder of how quickly things can change.

With the Conservatives unchanged since last poll but still down a bit versus 2021, the Greens are the only party coming out of this poll with an entirely positive swing versus both the election and last poll. This is their highest figure yet with Redfield & Wilton, and a joint second-highest ever.

One final thing to note is that after they weren’t included in the last two polls, Alba were directly prompted for this time. I think their 2% will therefore have come directly from the SNP’s share in the last poll. Given the SNP are polling 15% lower than their 2021 result here, Alba not doing statistically any better than at that election suggests they are extremely unlikely to prove a vehicle for dissatisfied SNP voters.

Constituency Vote

Despite the drama on the list vote, over on the constituency side of things the SNP remain in the lead, though not by much. This is their narrowest lead over Labour, and combined with the latter being on a joint-best share for this vote, has serious implications for the distribution of constituencies, effectively finally and firmly eliminating the advantage of FPTP dominance the SNP have enjoyed in most polls.

There’s not otherwise a lot needing said for this side of the poll. The Conservatives are again static, whilst the Lib Dems and Greens both tick down a little bit, though neither are dependent particularly on nationally strong constituency polling.

Seat Projection

Projecting that into seats might give us something like this:

Please see this page for how projections work and important caveats.

The combination of the SNP’s constituency logjam being burst and Labour placing ahead on the list vote leads to this also being the first poll since before the referendum where Labour come out ahead in seats. It’s only by one, mind you, in something of an echo of the SNP’s narrow win in 2007. Notably, and as would be the case with the vote shares too, this would be the smallest ever “winning” party at Holyrood, as the SNP’s 2007 result was a comparatively healthy haul of 47 MSPs.

Although the Greens don’t have quite their best vote share in this poll, the fact overhang is significantly reduced opens the door to this being the most seats they’ve ever been projected to win. It’s such a chunky pile for them that South comes out as the only region not to project to multiple Green MSPs.

Trying to form a government out of this may seem tough on the face of it, but as I’ve said before, I’m inclined to believe that a Pro-Union bloc majority automatically leads to a Labour First Minister now. The Conservatives really would have no choice but to put up or shut up on the “stop the SNP” front and back a Labour FM. An alternative option though, with 67 seats, would be what is known in Germany as an “Ampelkoalition”, meaning traffic light coalition.

That’s the current configuration of their government, with a Social Democratic (red), Liberal (yellow) and Green (duh) party together leading the country. I reckon the constitution would likely be a barrier here, especially if Labour and the Lib Dems felt they could rely on the Conservatives for exactly that reason. That said, the German Liberals are a lot more economically right wing than the UK Lib Dems, so the real Ampelkoalition had its own significant differences to overcome.

I like to drop this line in occasionally, but doing it again here: this poll is another great example of why I annoy some people by putting the list vote first in my reporting. It’s the most important vote! It’s the one that does the most to shape the overall distribution of seats! And it’s the one that secures Labour a lead here! In Germany and New Zealand they don’t even poll the constituency vote, but they are nationally rather than regionally proportional and account for overhang, so aren’t susceptible to constituency related distortions the way Scotland is.

As it so often does, the Westminster figures look very similar to the Holyrood constituency side of things. Again, this has the narrowest lead for the SNP over Labour, at just 3%. I’ve talked a lot about the SNP’s danger zone being when that gap dips into single figures, and this is a level at which it becomes highly likely Labour would win more MPs than them, just due to the SNP having a more evenly spread vote than Labour’s highly concentrated Central Belt support.

Very minor shifts on the big constitutional question, resetting once again to the 2014 figures. I reckon that for the time being we’re probably going to be settled around this point – the SNP and Greens are too busy firefighting day-to-day political issues to be making the case for Independence, and Labour’s wisest play would be to go heavy on change of government rather than forcing the debate back to constitutional ground they’ve generally suffered on.


As ever, the last little bit of analysis concerns those hypothetical and more proportional voting systems that BBS likes to play about with. The use of pure FPTP at Westminster is an affront to democracy, and though Holyrood fares far better, AMS is still deeply imperfect. The examples here simply transpose the poll findings onto more proportional voting systems – the reality is that different systems would of course result in different voter behaviour.

At this point, the SNP have lost enough support that our actual AMS system probably works reasonably proportionally, so the difference here isn’t massive. The Lib Dems were the only party whose ordinary AMS projection had them below their vote share, which this version fixes, but that’s about it.

Basically the same applies here as with RAMS.

Scandinavian Style Westminster

A proportional approach to Westminster would just squeak the SNP ahead of Labour in seat terms, though would seem about as hotly contested as FPTP. The big beneficiaries would likely be the Conservatives, as tends to be the case at the moment, and the Greens scrape the threshold this time so would also benefit from a crumb of proportionality.

If you find this or other Ballot Box Scotland output useful and/or interesting, and you can afford to do so, please consider donating to support my work. I love doing this, but it’s a one-man project and takes a lot of time and effort. All donations, no matter how small, are greatly appreciated and extremely helpful.
(About Donations)