Thread by @ListVoteSense

Well done to @Yes_Kirriemuir and @thecommongreen for this event, and for making it available on YouTube.

So much misinformation about the AMS list vote is making the rounds on social media, and the YES movement desperately needs informed voices to separate fact from fantasy.

https://twitter.com/Yes_Perth_City/status/1289863007812583424

Dr Craig Dalzell provides an interesting synopsis of the history of electoral systems used in Scotland, with the pros and cons of each: from the easy, but unproportional First Past The Post, to the Additional Member System (AMS) used for elections to the Scottish Parliament.

He points out that despite later claims to the contrary, the AMS system with its regional lists, was in fact primarily to counter fears of further Labour dominance (and not to prevent SNP majorities - at the time, of course, the SNP had but a handful of Westminster MPs).

The lists are intended to provide a more proportional representation of seats in Holyrood. The use of regional lists as opposed to a national list, effectively imposes a minimum threshold to win a seat of about 5% (lowest so far is 5.1% for a Lib Dem).

Those who recall the setting up of devolution will remember originally there were to be exactly the same number of list seats as constituency seats - creating a parliament of 146 members. Labour reduced the number of list seats from 73 to 56, thus reducing the proportionality.

Craig states anyone prepared to 'game' the system to increase seats for a platform should be prepared to see & accept the same tactic being used by opponents. He mentions how Labour's attempt to have the Cooperative Party stand on the list in place of Lab was ruled out by the EC.

On the assumption that it was at all possible to 'game' the list and elect many indy list MSPs, he maked the point that such a 'supermajority' of MSPs would be on less than 50% of the vote and unlikely to carry weight with WM, compared to an SNP majority on over half the vote.

Craig also reinforces fact that list MSPs are not 'second class' parliamentarians (Yessers who lambast list unionists as 'rejects' forget that, with only a handful of constituencies, it was the list that allowed SNP to overnight emerge as Scotland's second largest party in 1999).

He mentions an apparent 'loophole' in the system, whereby voting for a party that is not standing in the constituencies can garner seats. This a a crucial aspect of the argument put forward for those who advocate the indy list parties.

Craig doesn't go into any electoral arithmetic. This is unfortunate (indeed a critical omission), as such analysis is essential to evaluate claims about 'gaming' the system. Even a summary would take this head on, exposing the many claims from proponents of list parties as false.

He does make the point that modelling also allows you to produce seats numbers based on unrealistic scenarios. (We have to base our expections of what support parties actually have: we cannot sensibly vote for a list party polling at 0.5% and expect an indy 'supermajority'.)

Craig sensible draws attention to the difficulty of building up a new party, and how difficult it is even to attain 5% of the vote, the effective threshold to gain a single list seat, and indeed how difficult it is to gain a second list seat. My own summary from 2016 shows 10%:

He makes the perfectly obvious point (ignored by many) that a second indy party (list only) already exists: Scottish Greens. (@scotgp consistently polls at or above threshold to win list seats and is currently the only viable alternative for an indy list vote not going to SNP).

And in passing, Craig also effectively rebukes the claim that the Greens are not a pro-indy party. (I find made predominantly by pushers of new list parties vying for the mantle of the non-SNP list party option, or Yessers who even claim the SNP itself is not for independence!)

Craig reiterates the illegality of a 'dummy list party' taking the SNP whip. Proponents of 'indy only' list parties typically fail to address the issue that they will have to take policy positions in parliament - thus to attract votes, they have to have a manifesto platform.

And if such parties have considered policy based manifestos, then they are no longer 'single issue' parties: why therefore would indy supporters vote for them if they reject their policies? (Leaving aside the matter of more than one such party vying for your vote.)

It all sounds rather enticing in theory, but when the practicalities are examined - building up a party, getting enough support to clear the 5% regional threshold, having a policy manifesto, let alone competing with all the other fringe parties, the odds are not favourable.

An analysis of the electoral maths would also have revealed how a vote for a list party that fails to pass the 5% threshold could actually have the effect, not only of electing no MSPs, but also of seeing potential SNP seats lost to unionists. This danger needs to be highlighted.

Craig also draws attention to the highly unrepresentative nature of ScotPol Twitter, inviting indy 'thought leaders' to compare how many people engage with their tweets to the number of people even in their constituency. Be wary of self selecting online polls & canvass your town.

Craig concludes with the statement that 'gaming the system' is hard. I would also add that the typical assertions of 'tactical voting' to remove 'only unionist' seats needs to be tackled and exposed as nonsense. A 30 second examination of a regional vote count table reveals this.

While theoretically, you can pump numbers into a spreadsheet to produce miracle results, I would add that such projections are based on fantasy and have no relation to reality.

Analysis of data reveals that increasing SNP list vote by a few % is more effective at producing seats

Craig corrently points out that the focus should be on increasing the indy vote. I'd add by emphasising that these votes will be wasted unless they go to the only two parties capable of winning seats - the SNP and Greens - and abandon impractical notions of gaming the system.

The first question comes from Pat Leigh of AFI (#MaxTheYes). He makes the point that the AFI isn’t trying to game the system, but is standing for election.

Here’s where we have to distinguish between two ideas typically combined in talk of ‘gaming the system’ with list parties.

The first idea that a list vote can be used ‘tactically’ to enable a list party to ‘target’ ‘only unionists’ without endangering SNP or Green seats.

This is absolute nonsense, and merely demonstrates those pushing it have no knowledge of how it works.

‘Just say no’ & move on.

The second idea is about ‘maximising’ Indy seats. While theoretically if you assign (taking things to extreme) 100% of SNP list vote to *one* list party, you can add a couple of dozen seats, this singularly fails to reflect the real world, where such parties barely attain 0.5%

With no polling to suggest anything but <1% support for fringe list parties, the result will only be zero MSPs.

Modelling of 2016 showed that the 1.1% won by RISE and Solidarity luckily wasn’t suffiently height enough to cost SNP seats.

The danger is if these list parties attract nearer 4% of vote from SNP: modelling shows while this is insufficient to win seats it will threaten potential SNP list seats.

If >10% of SNP vote defects to two or more parties, SNP will lose three of four projected seats to unionists

Takes a lot of hubris for AFI rep to ask Greens, already a successful party (now projected to win seats in every region) to throw in their lot with a party yet to register in a poll.

Greens look to win seats in every region, why on earth would they hand a seat to another party?

As is typical with propents of list parties, assertions are made as fact, with no data analysis to back them up: the notion that the SNP can’t increase its list seats on currently high support, and an alternative party is needed.

This fails to recognise that modelling show the SNP *can* win more seats on a higher % share. If its current list share of 50% matched constituency share of 55%, it would add 4-5 list seats.

Naturally Greens are always instantly discounted by list party devotees as though they don’t exist, and only some new unknown and untested list party will bring us to the promised land.

Those arguing for yet another party ignore fact that bigger parties do better than small ones

The theoretical possibility of having a parliament where indy seats are way above their % vote is merely an extreme scenario not based in reality.

In absence of polling indicating *any* new list party has risen above RISE (sorry!) to hit 5-10%, it win *not* increase indy seats.

The example Craig gives of a parliament where fully 80% of seats are indy will only happen if 100% of SNP vote goes to *one* list party.

I don’t see the SNP standing down from the lists, so it’s a pointless fantasy to even entertain such notions.

We need to stick to real world.

AFI Pat talks about ensuring a supermajority of indy seats, but unless SNP clears way for AFI on the list, it’s not going to happen.

And to suggest such a possibility, with no polling to show support for the AFI, is the electoral equivalent of a ‘get rich quick’ scheme.

AFI Pat says the focus is independence and we have to work under the system we have.

In which case, it would be better for fringe list parties not to stand and encourage everyone to vote SNP or Green.

The more parties the indy vote is split over, the less seats won.

If the aim is purely to increase indy seats, then if by the campaign start polls show that none of these new list parties have succeeded in attracting the 5-10% necessary to win seats, they should stand down, not take part in election, and urge indy voters to vote SNP or Green.

Next, Morgan makes obvious point that by increasing SNP list vote, it will produce more SNP list MSPs (and indeed it was the fall of this that led to four list seats in 2016).

There would therefore be no risks of seats lost to unionists if fringe parties don’t win enough votes.

He suggests all other indy parties stand down, including the Greens.

While it is sensible not to have myriad other list parties siphoning off SNP votes, modelling shows that if the Green vote declined or went entirely to the SNP there would be no benefit and likely only losses:

The current distribution of votes with one large and one smaller Indy party (SNP & Greens) actually works very well in terms of maximising indy list seats.

If you model shifts back and forth, it doesn’t lead to much change, but a stronger Green vote is best.

Craig refers to ‘barriers’ to over representation of SNP. If you model increased SNP vote it will gain seats, & be in better position to do so as a larger party by system bias

To emphasise Morgan’s point if in 2016 SNP list % had matched constituency % it would have 5 more seats

Cllr Bill Duff makes the point that, of course, we can’t take it for granted that SNP will win comfortably in 2021, so list votes are ‘free’ to try to ‘game’ the system. If the SNP vote falls and marginal seats are lost, then every SNP list vote is crucial to winning on the list.

Moderator Ruth raises again the notion that there is somehow a limit on the seats the SNP can win on the list, and how to know if voting for another party in a regional list will maximise indy seats.

Analysis shows the more the SNP list vote goes up the more seats it will win.

2016 result was unfortunate in that the four list seats won by SNP has led people to assume (erroneously) that SNP will ‘max out’ on list and vote for them is pointless

They don’t realise poorer showing due to a reduced %: much less & unionists would have won these seats instead

The second point Ruth makes about knowing which other party to vote for in a region, I have looked at in my most recent article.

The answer is it’s impossible to know without looking at the results whether the SNP or Greens would have been best placed to win a final list seat.

Craig says you cannot control how people vote & get them to follow a plan

In light of above, I’d add it’s also impossible, as no one can know before results which parties will have the votes closest to winning each round: every party opposes every other

medium.com/@listvotesense…

In 2016, if Lothian SNP vote was up 0.16% from Greens (<500 votes), SNP would have taken last seat instead of Greens.

But it’s SNP v Greens, not indy v unionists: all compete together. Having yet another indy party there splits vote further, risking SNP seats going to unionists.

This shows you just how finely balanced votes can be in the list: but it is impossible to know before you see the results, how the numbers play out in the regional seat allocations.

And if a constituency is lost or gained, there’s another factor you can’t predict.

As Craig says, all parties can do is just promote themselves.

And I’d add that in order to secure indy seats and not waste your vote, it has to be for the only two parties capable or winning seats: SNP or Green.

It’s just a matter of voting for which best suits your views.

Craig mentions George Galloway standing in South Scotland.

While in 2016 SNP won three list seats there, current polling has it winning none, just losing out to Tories for the last seat.

But Galloway could well steak enough Tory votes to hand it to SNP.

https://twitter.com/listvotesense/status/1288475170513068034?s=21

https://t.co/au8Zh0hP08

Pat Lee AFI mentions touring the country to discuss 'openly & honestly the concept of #MaxTheYes.'

But therein lies the nub - the validity of the concept just can't be determined by *any* sort of discussion. The only way to separate fact from fiction is to perform data analysis.

As I have written before, this is analagous to a group of scientists pontificating over various hypotheses, but without ever conducting a single experiment to see what, if any, claims are actually supported by the evidence!

Moderator Ruth makes point again talk of list parties is confined to the Twitter Yes bubble.

Pat believes he can galvanise the Yes movement to work as a coordinated whole to vote en masse for the AFI.

Craig mentions how hard it was even for the Greens to get media attention.

Bill again, makes point that the ‘alphabet soup’ of these extra parties (who’s involved & what they stand for) is too much for a ‘political anorak’ to keep track of, let alone average voter, who will only be aware from papers

The Greens are already a tried and tested alternative

Jean asks about the prospects of standing well known independents in regions ‘where the SNP or Greens don’t stand a chance’.

Currently polling, however, shows the Greens are likely to win a seat in every region now:

Where SNP wins varies too: now predicted to lose South seats, but win in Glasgow/Central. Most talk as if SNP will win same as 2016.

Additionally with ~6% needed to win a seat, if an independent got 20%, excess is wasted, as there would be no second candidate (as with a party).

Pat Lee of AFI is given opportunity for a final plug to promote his platform.

He makes the all too typical - and utterly false - assertion that voting AFI will remove unionists (and not affect SNP seats), with the usual rhetoric about ‘seat warmers’.

Stirring talk, but not true

If 9% of the SNP list vote (on current polling), defects to a list party in Glasgow, the result will not be #MaxTheYes, and AFI seats, but rather the SNP losing its predicted list seat to Labour:

If 13% of currently polled SNP list vote defects and is split between two list parties (eg AFI, ISP, RISE, Solidarity), then three of four predicted SNP seats will be lost to unionists. Not good.

Ian White asks about whether the Greens are standing in all regions. Naturally they will be, (and based on current polling stand to win seats in every region).

Craig makes the good point in terms of Plans A and B, that it’s the *pressure to produce results* that’s paramount. If the ability to pressure successfully doesn’t exist, then it doesn’t matter what plan we are promoting. If you can’t pressure A, you won’t be able to pressure B.

And that concludes my commentary!

Thanks again to @thecommongreen and @Yes_Kirriemuir for this interesting discussion.

Here's a link to the document referenced throughout by Craig: 'Within Our Grasp':

commonweal.scot/policy-library…