What happened at the election?

Many explanations of the recent election result, where the SNP actually increased their share of the constituency vote, often focus on the age-old practice of “well, we told you so”.

James Kelly couldn’t even wait till the next morning to publish his (http://scotgoespop.blogspot.co.uk/2016/05/i-hate-to-say-i-told-you-so-but.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed:+ScotGoesPop+%28SCOT+goes+POP!%29) – the first comment was at 3.39 AM on the 6th. Wings took a bit longer, but is considerably more considered – first comment at 1.52 pm (http://wingsoverscotland.com/dont-say-we-didnt-tell-you/). In the course of this, the Rev Campbell confirms again that he has “been warning readers for nine solid months that the AMS electoral system couldn’t be “gamed” and that the meaningless pursuit of a “pro-independence” majority could lead to disaster”.

However, as Wee Ginger Dug points out (https://weegingerdug.wordpress.com/2016/05/06/the-curates-egg/) the SNP “didn’t only ensure that they were returned to office, they also increased the number of votes that they received. That’s pretty amazing. The SNP continues to defy political gravity and the independence movement has not been sidetracked or set back. But they were buggered by the voting system. This time the SNP didn’t break it.”. In other words, what just happened was designed in the electoral system to happen. The outcome of the 2016 Scottish election is largely how the Scottish Parliament’s electoral system works it was what was intended. There are though one or two other issues that I will work through in the course of this, but in the main what I am going to argue here is that what just happened was substantially what I have warned about in several posts in this blog.

Essentially what I have argued is that the SNP, according to the polls, were on course to win most constituency seats. The reality was that they won fewer than I would have expected, and I will come back to this. However, if we look at the table below, which sets out for each Region, how many constituency seats were won by the SNP and how many List seats they won in each region, the conclusion is clear – that unless the SNP won “only” six seats or less, they would win no List seats. As you must be able to remember, the number of constituencies won are included in the calculation to allocate List seats, and the division seems to come at or about six. So, in at least five of the eight electoral Regions, a vote for the SNP Regional List elected no one at all. Arguably the vote was wasted, or at least might have been used more effectively, in that it might have elected someone from another independence supporting party without doing any damage to the SNP. What damage could possibly be done, if the SNP would have no one elected on the List?

Region

Constituency seats won by SNP

Regional List seats won by SNP

Glasgow

9 of 9

0

Central Scotland

9 of 9

0

North East Scotland

9 of 10

0

Mid Scotland & Fife

8 of 9

0

West Scotland

8 of 10

0

Lothian

6 of 9

0

Highlands & Islands

6 of 8

1

South Scotland

4 of 9

3

 

Let’s take a specific example – the Glasgow Region where, all nine seats were won by the SNP. The table below sets out the List votes for each party, and then how they were counted in each region. As a reminder, the List vote for any party will be divided by its number of elected members (constituency and/or List). Therefore, the SNP List vote, which at 111101 was more than the vote for the Conservatives, Labour, Lib Dems and Greens combined, in round 1 was divided by ten (9 constituency members +1), and as I have argued before, this was simply too big a disadvantage for the SNP to overcome. Indeed, you will notice that instead of stopping at 7 (there are only 7 list members in each region) I have gone on to 10, since it is only then that the SNP would have gained their first List seat in Glasgow. In fact, in round 7 the 7th and last List seat was won by the Conservatives with a modified vote (having already taken one list seat, so divided by two) with 14766.5. To have challenged for that final seat, the SNP List vote would have had to be ten times that – 147665, which is 36564 more than they actually got – a required increase of 32%. Just where were those votes going to come from? The SNP constituency in Glasgow was 17,342 more than the List vote, so even if everyone who voted SNP in their constituency had voted the same way on the List, they would still come up less than half way to the number needed to secure even the last (7th) List seat. Where were these votes coming from?

Party

List vote

1

2

3

4

5

6

7

8

9

10

SNP

111101

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

Labour

59151

59151

29575.5

19717

19717

19717

14787.75

11830.2

11830.2

11830.2

9858.5

Con

29533

29533

29533

29533

14766.5

14766.5

14766.5

14766.5

9844.333

9844.333

9844.333

Green

23998

23998

23998

23998

23998

11999

11999

11999

11999

7999.333

7999.333

LD

5850

5850

5850

5850

5850

5850

5850

5850

5850

5850

5850

 

The alternative argument to #bothVotesSNP was, “first vote SNP, second vote someone else”. Had that “someone else” been Green, and even just 25% of the SNP transferred to Green for their second vote, then instead of four Labour, two Conservatives and one Green, there would have been three Labour (one less), one Conservative (one less) and three Green. For sure the SNP would have had no direct advantage from that arrangement – they would have won no additional List seats, but ‘as is’ they have no List seats either. BUT the opposition would have had fewer Unionists (Pauline McNeill and Annie Wells), and two more Green Party members who at least share the aim of independence.

More importantly than anything though, the SNP, having won all nine seats had a vanishingly small chance of picking up list seats in Glasgow, and indeed in the first five regions in the first table.

But let’s look now at the polar opposite – South Scotland’s Regional List vote, which worked out as follows

Party

List vote

1

2

3

4

5

6

7

SNP

120217

24043.4

24043.4

20036.17

17173.86

17173.86

17173.86

15027.13

Labour

56072

28036

18690.67

18690.67

18690.67

18690.67

14018

14018

Con

100753

20150.6

20150.6

20150.6

20150.6

16792.17

16792.17

16792.17

Green

14773

14773

14773

14773

14773

14773

14773

14773

LD

11775

11775

11775

11775

11775

11775

11775

11775

 

Thus three SNP and two each for Labour and Conservative. In this case it’s hard to see what might have been done to increase the number of independence party representation. Certainly it is true that #BothVotesSNP was a policy that delivered in this region. For instance, had the SNP vote been split with the Greens then there would be no discernible difference. If we assume that 25% of the SNP vote was induced to vote Green on the List, the distribution of seats would change such that Labour and Conservative would still have had two List MSPs each, but the Greens would have had two and the SNP only one. But from the SNP point of view, #BothVotesSNP was the correct option to put to the electorate in South Scotland.

However, if we look at things across Scotland, then while #BothVotesSNP produced positive outcomes in one region (South Scotland), it certainly did not in four regions – Glasgow, Central Scotland, North East Scotland and in Mid-Scotland and Fife. Even in West Scotland, it is arguable that because of the number of seats won by the SNP, that #BothVotesSNP did not succeed in winning more seats.,

In Lothian and Highlands and Islands, where the SNP won all but 3 or 2 seats (respectively) the situation is more complex and there is scope for genuine disagreement about whether #BothVotesSNP was the right advice. That can only be said with confidence in South Scotland where the SNP won fewer than half the seats. But certainly in four regions, and arguably five, #BothVotesSNP was inappropriate advice.

What do we mean by “inappropriate”? Well I would say that a strategy advising both votes to a party that won every seat is “inappropriate” if on the List – as happened in Glasgow – it produces an outcome of four Labour, two Tories and one Green, when if SNP voters (even 25% of them) had switched to Green, then there would have been three Labour (one less), one Tory (one less) and three Greens (two more). That kind of shift across these five regions would have done nothing directly for the SNP, but would have reduced the Unionist presence in this Parliament, and by implication had more support for independence.

In “Five cold, hard facts about the election” (http://wingsoverscotland.com/five-hard-facts-about-the-election/) Wings Over Scotland is scathing about the Green Party, and in particular its commitment to independence. He bases this on the Greens demanding that there should be a petition signed by one million Scots, and a statement by Patrick Harvie that there should not be an early re-run. However, the SNP manifesto says “Holyrood should have the right to hold another referendum if there is “clear and sustained evidence” of majority support for independence, or if there is a “significant and material” change in circumstances, such as Scotland being taken out of the EU against its will.” (http://www.bbc.co.uk/news/election-2016-scotland-36181591). These two statements by the Greens and by the SNP seem to me to be saying something fairly similar, even if in slightly different ways – a petition and “clear and sustained evidence”, for instance.

But the key point that is missed by Wings is that, even if the support for independence by the Greens is uncertain, at least it is only uncertain. If the SNP could win only a small number of List seats, then the alternatives are uncertain support from the Greens and certain opposition from the Unionist parties. The uncertainty of the former, even if true, seems to me preferable to the certainty of the latter.

The focus of complaint of course was the recommendation that SNP voters should consider giving their List vote to a party other than the SNP. The argument was particularly strongly put by James Kelly on Scot Goes Pop. In a post-hoc piece he wrote “as a result [of advice not to vote SNP on the List) the SNP list vote drops, and the party doesn’t have enough votes for a list seat in six out of eight regions (compared to just one out of eight in the 2011 election).” (http://scotgoespop.blogspot.co.uk/2016/05/the-tactical-voting-lobby-were-proved.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed:+ScotGoesPop+%28SCOT+goes+POP!%29)

First of all, it’s hard to disagree with the proposition that if the advice to vote other than SNP on the List is followed the SNP vote will decline. What is contentious is the second part – that “as a result the party (SNP) doesn’t have enough votes for a list seat in six out of the eight regions”. Is that a fact? Let’s go back to look at Glasgow in a bit more detail.

The total constituency vote for all parties was 246,957, but the list vote was ‘only’ 229633. In other words, 17324 people voted for their constituency representative but for whatever reason did not use their List vote. Were they SNP voters? Well we will never know, but one thing we do know is that the SNP vote in Glasgow constituencies was 128,443, but their List vote was 111,101, so it’s clear from this that 17,342 (or 13.5%) of their vote did not follow #BothVotesSNP. But how much did this matter?

To try to say anything about this means making several assumptions. First of all, we will assume that the 17,342 all voted Green, despite the fact that we have no evidence for this – some might have been former Labour voters who had given their constituency vote to the SNP, but as some sort of contrition voted Labour on the List. But, for our purposes – since the advice was to vote for another indy party – we will assume it all went to the Greens. This would have reduced their List vote to 6,656 (it is worth noting there that the Greens took 6916 votes in the Kelvin constituency alone). In turn this also points to another defect in this assumption – that the SNP constituency vote might have included Green voters who had no candidate in their constituency. But sticking with the assumption, and repeating the Glasgow figures, the following comes out

Party

Amended List Vote

1

2

3

4

5

6

7

SNP

128443

12844.3

12844.3

12844.3

12844.3

12844.3

12844.3

12844.3

Labour

59151

59151

29575.5

19717

19717

14787.75

11830.2

11830.2

Conservative

29533

29533

29533

29533

14766.5

14766.5

14766.5

9844.333

Green

6656

6656

6656

6656

6656

6656

6656

6656

Lib Dem

5850

5850

5850

5850

5850

5850

5850

5850

 

In other words, what happens is that Labour still win four List seats, the Tories still win two, but instead of the Greens winning one the SNP take the last seat. But to reach that conclusion we have to make the somewhat heroic assumptions set out above – not least that it was SNP votes leaking off to the Greens (or other fringe indy parties) that was the cause of the SNP’s failure to win a List seat in Glasgow. And what have we achieved? We have replaced one indy supporting MSP with another indy supporting MSP. Yet, as I point out above, had more SNP voters ignored the #BothVotesSNP recommendation, instead of four Labour, two Conservatives and one Green, there would have been three Labour (one less), one Conservative (one less) and three Green – the balance for independence would have been more pronounced.

Ah but, I can hear Mr Kelly pronounce, if every SNP voter had voted twice for the party, then there would be one more SNP List MSP – 64 MSPs rather than just 63. Indeed, if this had been repeated in all five Regions where the SNP won at least 80% of the seats but no List seats, then instead of 63 SNP MSPs there would be 68 and an absolute majority. Job done.

However, following our assumptions about where the votes would come from, there would be only one Green MSP. There are currently six, so if we assume, as might have happened in Glasgow, that instead of the Greens winning a List seat, the SNP would take the list seat instead in five of the eight regions. The balance of independence and Unionist parties would be as it is now 69 indy MSPs and 60 Unionists.

It might then be argued that that is just too bad for the Greens – that politics is a dirty business other than for winners. But that omits what we might have achieved had there been even a 25% transfer of SNP voters to the Greens (or an agreed List alternative). In Glasgow, remember there would have been two fewer Unionists (one each Labour and Conservative) and two more Greens. If that had been repeated across all five of the regions where the SNP won at least 80% of the constituencies, that would have meant 10 fewer Unionists (50) and 10 more MSPs from independence parties (79).

Not realistic? Perhaps so, particular as the assumptions are unprovable. We don’t know exactly how those who voted SNP in their constituency used their List vote, if not for the SNP. Did they vote Green (or RISE)? Did they perhaps vote Labour or Lib Dem, or even Conservative? Or did they just not vote? We don’t know.

But what this analysis does show is that in terms of support for independence – as opposed specifically for the SNP – the recommendation #BothVotesSNP actually diminished the number of members possible in this current Parliament who support independence. Is a Parliament where 69 support independence and facing 60 Unionists, better or worse than one where up to 79 support independence with 50 Unionists. Even if it means the SNP had no majority in the latter, I would prefer that, particularly as that is where we are with only 69 supporting independence. An opportunity has been lost.

However, Mr Kelly makes a further point – that those who argue as I do, suggested that “That the SNP were definitely going to win all nine constituency seats.” I would have to say that I don’t think I ever said – or saw anyone who argued – this. But I did expect them to win more than they did. For instance, just over 12 months ago, the SNP won 56 of 59 Westminster seats. If we gross that up to the 73 constituency seats at Holyrood that would be 69 seats. We might well ask what happened there?

It is fair to say that the Holyrood constituencies do not always map neatly on to those for Westminster. But if we take the West Dunbartonshire Westminster seat which Martyn Docherty took with a majority of over 14,000, the bulk of that are the Clydebank and Dumbarton Holyrood constituencies. But while the SNP increased their majority tenfold in Clydebank, in Dumbarton, while the SNP share of the vote increased, they could not overturn a majority for Labour that had been only 1639 in 2011. Contrast that with Rutherglen constituency where James (“no I will not sit down”) Kelly’s 2011 majority of 1779 became an SNP majority of 3743 this time.

Dumbarton, though includes Helensburgh which is part of the Argyll & Bute constituency, and is the closest town to Faslane where the UK Trident fleet is based. One might therefore expect a significant vote there, hostile to the SNP’s policy of not renewing Trident. Helensburgh is also relatively affluent in comparison to the rest of the constituency. Nonetheless, given the swing to the SNP since 2011, one might have expected this seat to fall to the SNP this time.

Nor does tactical voting by Unionists offer much in the way of explanation in this seat, since one might have expected the Tories to switch to Jackie Baillie to keep out the SNP. However, the Tory vote did not decline but actually increased on 2011 (both absolutely and in terms of share), albeit modestly. Indeed, the Conservative vote was the highest it has been since the first Scottish Parliament elections in 1999. Moreover, there were attempts by Unionists to promote this sort of strategy at the Westminster election last year, and that didn’t end well for them at all.

But perhaps the biggest problem with Kelly’s thesis here is that no one who argued against #BothVotesSNP suggested anything other than to vote for the SNP in their constituency. The moot was always what was best for the List vote?

In an entry on his Facebook page on 6th May, SNP stalwart, Iain Lawson observes

“Just take a look at the seat Tory Ruth Davidson has just won in Edinburgh.

The Yes vote was split by the Greens standing and that was enough to give Unionist Ruth Davidson a victory.

Of course the Greens had the right to put up a candidate BUT was this the result they wanted?

I have heard nothing else than a Green vote was a safe vote, it would increase the YES majority.

So GREEN PARTY WHAT HAPPENED?

Of course Iain is right, but only right as far as he takes it. Once we step outside of the constituency section, and we find the SNP putting up a slate of Regional List candidates, where very often, because of the number of constituency seats won by the SNP mean their List candidates have a vanishingly small chance of election, the Green Party might well ask the same question. As I have pointed out above, if even ‘only’ 25% of the SNP List vote in Glasgow had transferred to Green there would have been two fewer Unionist MSPs in our Parliament.

Taken together, James Kelly’s blog and Iain Lawson’s Facebook entry, crystallise the issue about this election. In an election we expect parties – no matter how close their views might be (and it should be recognised that the views of the SNP and the Green Party are not always particularly close – for instance on income tax) – to compete with each other for out votes. At the same time, a larger prize is at stake – independence.

Robin McAlpine observed in a Common Space article that during the election “the SNP without the context of its wider movement [Yes Scotland] looks awfully like a political party. It made it much harder to talk to the people who are tired with ‘normal politics’.” (https://www.commonspace.scot/articles/3972/robin-mcalpine-there-s-nothing-for-yessers-to-fear-from-sp16-unless-we-don-t-listen), which offers us a second explanation for ‘what happened’.

The first explanation is that the electoral system for Holyrood ‘worked’ – the dominance of the SNP in the constituency section came back to disadvantage them so much on the Regional List that they won few List seats. Of course, it can be shown that if everyone who voted SNP in their constituency had voted in the same way on the List then they might have won an additional five seats, but that argument requires some pretty heroic assumptions, as we have seen (e.g. to assume that, for instance, no one who voted SNP in their constituency was not a supporter of a fringe party – Greens for instance – and only ‘lending’ their constituency vote as they had no one else to vote for).

An easier, and potentially more fruitful approach, could have been to encourage more SNP voters to switch to Green on the List, as with a 25% switch two fewer Unionist MSPs would have been elected on the Regional List, being replaced by two Greens.

Of course it has been argued that the commitment to independence by the Greens is uncertain, but is uncertainty about that commitment (if true, as their views on the next indyref seem very similar to those of the SNP) not better than the certainty of Unionist opposition?

But what good would this do the SNP? The short answer is “not much”. Indeed, if #bothVotesSNP had stuck 100% then they might just have won the additional Regional List seats they needed for a majority, though it would have been, to use a quote associated with Sir Alex Ferguson, “squeaky bum time” – maybe 66-68 seats. But by encouraging switching we could have been looking at a Parliament of 63 SNP MSPs and 10 or 11 Green MSPs, so 74 of 129 MSPs supporting independence. With more switching on the List that figure could have been even greater, which brings me back to the Robin McAlpine quote.

How best can we take independence forward? On a party political basis or as a movement? Of course Iain Lawson is right about the Greens standing in constituencies and costing the SNP seats (eg letting Ruth Davidson in), but it is also demonstrably true that #bothVotesSNP cost the Greens List seats.

My own view is that more clearly than anything, the election just past shows the need to revive independence as a movement, in which the SNP might be the biggest part, but by no means the only one. We will turn to that in a subsequent blog.

Voting tomorrow

In today’s Herald, Magnus Gardham has a piece with the headline SNP ‘fears Nicola Sturgeon may not secure majority’ in Holyrood elections, insiders claim” (http://www.heraldscotland.com/news/14468532.SNP__fears_Nicola_Sturgeon_may_not_secure_majority__in_Holyrood_elections/?action=success#comment_16127289).

This story is very like one in this week’s Lennox, that the election was too close to call. Jackie Baillie is defending a majority of just over 1600 from 2011, during which time the SNP vote has increased nationally by 20% and half her constituency includes one of only three local authority areas to vote Yes. Of course it is said that the greater part of that came in Clydebank, and that SNP support in the Dumbarton and the Vale is at a lower level, and as for Helensburgh ………..

But just to add to Jackie’s woes, given the split in Labour in the UK on Trident, this has made it difficult for her to use the “defending jobs” argument as much as she did last time. That is not to say it has never passed her lips – at hustings she continues to make the point that she will “fight for local jobs”.

I have no access to any polling data for the constituency, but just one impression of my own. The other day, I was driving down the High Street, past the SNP election hub – full of folk doing leaflets etc. We had a couple come down our street from the SNP delivering leaflets (this one was re the Vale, another JB hobby horse). Anything from Labour has come via the Post Office.

A bit further on – Castle Street – is Jackie’s constituency office. It was shuttered and locked, just as the same office was the Saturday before Gemma Doyle got voted out (for those who don’t know, they shared). I see little sign of Labour in the constituency, other than letters to the local paper. I doubt they have given up, but if the level of campaigning is any kind of indication of their vote, it doesn’t look good for JB. Indeed, it seems in Ayrshire Labour have been busing London activists up to supplement local resources. This of course happened this time last year, and look what happened then.

So, why the article in the Lennox? Why this article? Is it to convince Labour Party – and maybe Tory Party – stalwarts to get out and vote, the cause is not yet lost? Or is to cause consternation in SNP ranks? Both of these? Indeed, we might ask “why this article?” Was Jackie Baillie holding Marc McLean’s hand as he wrote?

Or is there a real point?

Many seat prediction sites suggest that the Tories might take 3 or 4 seats (mainly in South Scotland region, holding on to Ayr, Ettrick and West Dumfries, where young Mundell is standing). Suppose the Lib Dems hang on to both the Northern Isles seats. Even if the SNP take all the rest, that leaves them with 67 or 68 constituency seats Bearing mind that this would be 92 or 93% of all the constituency seats, it is hardly a bad result (especially if they secure north of 50% of the vote). But, should, for instance Labour hang on to 3 or 4 constituency seats it would mean no SNP majority.

That of course is the basis of the #bothvotesSNP argument. Yet if the SNP were to end up with 62 or 63 constituency seats (which would still mean them winning 9 more than they had last time) we are still in the territory where they might win no List seats as John Curtice pointed out in his paper (http://electoral-reform.org.uk/sites/default/files/files/publication/The-2016-Scottish-Election-Briefing.pdf). For instance, that number of constituencies (62-363 suggests an average of 8 constituencies per region. Last time in Lothian they won 8 of the 9 constituencies, with a List vote of just under 40% (not a lot less than the polls suggest today) but never won a single seat.

Just for illustration – not a prediction – just some numbers, the polls suggest right now that the SNP will take upwards of 40% of the List vote, Labour 21%, the Tories 19%, Greens about 12% and Lib Dems about 8%. Let’s work those numbers through a representative region.

 

Party

List vote

1st round

2nd round

3rd round

4th round

5th round

6th round

7th round

Seats won

SNP

40

4

4

4

4

4

4

4

0

Labour

21

21

10.5

10.5

10.5

7

7

7

3

Con

19

19

19

9.5

9.5

9.5

6.3333

6.3333

2

Green

12

12

12

12

6

6

6

6

1

LibDem

8

8

8

8

8

8

8

4

1

 

In a Region with 9 seats if the SNP took them all, then working through those numbers using the modified d’Hondt system (divide List vote by elected members +1), would give Labour 3, the Tories 2 and the Lib Dems and Greens 1 each.

To simulate the #bothvotessnp policy I assumed that no one voted Green and their entire vote went SNP (very unlikely, but more likely than Unionist party voters voting SNP). This actually – from an indy supporting point of view came out worst of all – Labour and Tory get three seats each and the Lib Dems one – the SNP still get no regional list seats.

Party

List Vote

1st round

2nd round

3rd round

4th round

5th round

6th round

7th round

Seats won

SNP

52

5.2

5.2

5.2

5.2

5.2

5.2

5.2

0

Labour

21

21

10.5

10.5

7

7

7

5.25

3

Con

19

19

19

9.5

9.5

6.333333

6.333333

6.33333

3

Green

0

               

LibDem

8

8

8

8

8

8

4

4

1

 

The alternative to this would be for at least some of the SNP vote to go to another party that is not Unionist – for instance (again for illustration) the Greens. I assumed 12% of the SNP’s 40% would not switch, so giving the Greens a List vote of 40% (their own 12% + 28% of the SNP vote that switches to them).

Party

List Vote

1st round

2nd round

3rd round

4th round

5th round

6th round

7th round

Seats won

SNP

12

1.2

1.2

1.2

1.2

1.2

1.2

1.2

0

Labour

21

21

21

10.5

10.5

10.5

10.5

7

2

Con

19

19

19

19

19

9.5

9.5

9.5

1

Green

40

40

20

20

13.33333

13.33333

10

10

4

LibDem

8

8

8

8

8

8

8

8

0

 

The outcome of this is that the Greens win 4 seats, Labour 2 and the Tories 1 – Lib Dems get none. It might be pointed out here that the SNP have nothing to gain from this – they win no List seats. But then again, they win no List seats on any of the scenarios. Yet – to the extent that this might have been replicated across Scotland – do they not gain from having a party who would be the official opposition that is at least positive about independence?

However, that last scenario – or indeed the second one – is not going to happen tomorrow. We seem likely to be caught in the soggy middle ground where there are not enough votes for indy supporting parties on the List, and not enough for the SNP to win a decent number of list seats. But let me be clear there – that would be an awful lot more List votes than they could hope to win just now – as I pointed out even if the entire Green vote shifted to the SNP List, the SNP would still win no seats. Where are those votes going to come from? Labour, Tory or Lib Dem? The point is that with their vote as it stands, and the manner in which the deHondt system allocates regional list seats, the SNP, even if they increase their vote, are going to struggle on the List. Any transfer of votes to another indy supporting party than the SNP has stalled, something that the SNP themselves have encouraged with #bothvotesSNP.

Just to make the point as bluntly as possible, all other things equal (which to be fair they wouldn’t be), to win even the final regional seat, if they have won all the constituency seats, the SNP would need a List vote of more than 70% on the basis of the first scenario, or, in the second scenario 64%.

My own view is that the SNP really do need to look to win a majority in the constituencies, hope that the Unionist constituencies are concentrated in just a couple of regions (which is possible – HIghlands and Islands (Orkney and Shetland would be two), South Scotland (Ettrick, Ayr, West Dumfries) – so that they might pick up two or three list places there. Otherwise Gardham might well be right, even if only for a change.