‘Tactical Voting’ in the Scottish Parliament Election: a mass delusion deconstructed.

A lot of ‘virtual ink’ has been spilt by sundry bloggers, opinion formers and Yes pundits alike over the past few days, promoting the superficially sensible notion of ‘tactical voting’ in the Holyrood list to ‘maximise’ pro-indy MSPs (ostensibly at the expense of Unionists).

This week alone saw yet another ‘insta-list’ outfit arrive on the scene: former SNP MSP Dave Thompson’s ‘Alliance for Independence’ (AFI), which will now become the sixth pro-independence party.

The AFI looks set to join newcomer ‘Independence for Scotland Party’ (ISP), along with Solidarity and RISE, in the battle for that crucial 5% of the vote (the effective Regional threshold to win a single list seat). Many of the faithful are also now openly fantasising about the Second Coming of Salmond as a Messianic leader of a further putative list party (naturally, the fact that this is pure speculation doesn’t get in the way of a good story).

An entirely self selecting online poll (in other words totally worthless) even purported to show the potential reception for a putative list party.

Unfortunately, more confusion than light has been the outcome of these assorted blog posts and discussions, as convinced proponents of “tactical voting” battle it out in a war of words with those who (correctly) warn of the insanity of voting for fringe list parties, that not only won’t see elect any of their candidates elected, but would crucially siphon off votes required to elect SNP and Green MSPs (the only two pro-independence parties capable of winning seats), thus potentially allowing Unionist politicians to elbow their way in.

The single most important aspect of the discussion witnessed over the last week, is that it has once again highlighted just how widespread is the failure to understand the practical operation of the Additional Member allocation process — the effective 5% threshold that excludes minor parties due to the list vote being regionally based, and the Droop Quota (as opposed to the Hare Quota) that also rewards largest ones — let alone the impossibility of ‘gaming’ the seat allocation algorithm.

These failures to comprehend how the d’Hondt based Additional Member (AMS) process works has led to the dangerous and delusional fantasy that parties and voters can ‘beat’ the system through ‘tactical voting’. These delusions and assertions then achieve the status of factoids, propagating remorselessly throughout the Yes community, misinforming people and further compounding the damage.

Sadly, this issue rears its head in every Holyrood election, and was also the subject of much argument in the European elections last year. Every time you think the matter has been well and truly settled, another misguided, but evangelical proponent pops up, like another head of the Hydra.

But each time the facts are explained, it seems to make no difference whatsoever to the belief (and this is the operative word, belief), that tactical voting using ‘list only’ parties is the Holy Grail to maximising the number of pro-indy list MSPs.

Even if it was possible to ‘game the system’, and to pack Holyrood with a supermajority of indy MSPs (leaving unionist voters vastly underrepresented and feeling disenfranchised), how do you think that would be seen by potential converts to Yes?

Those who remember the 1979 referendum will recall the notion widespread in the Highlands, Islands and Borders, that a Scottish Assembly would just be ‘dominated by Central Belt Labour’, a fear that compelled many to vote against devolution at that time.

Do we really want to make potential Yes voters (you know, the sort of people we need to ensure we win indyref2 with 55%+), start to worry that if Scotland becomes independent, they’ll just get an SNP + Indy Alliance ‘one party state’, with no effective opposition?

Let’s talk about 2016…

One thing that cannot fail to be observed, is that so much of the discussion about tactical voting for a fringe indy party derives from the result of 2016.

2016 has become a veritable shibboleth for swathes of the social media indy movement, the election results fixing mindsets in an ediface of unshakeable opinion - nay belief - seemingly impervious to any rational argument or presentation of facts.

Because of 2016, it is now received wisdom that the SNP will not gain from list votes.

Because of 2016, it is received wisdom that a list vote for the SNP is a guaranteed waste, and should be put to better use by voting for another pro-independence party.

A casual familiarity with the way seats are allocated in the Regional lists recognises that parties which excel in constituency seats, start at a disdvantage in the list, as the algorithm used includes constituency seats already won.

One pundit highlighted the difference in the SNP constituency results between 2011 and 2016: the SNP was much less successful in constituencies in 2011 than 2016, and thus was able to benefit from the list in 2011 in a way it wasn’t able to do in 2016 (we’ll let slip, for the moment, that he didn’t also highlight the reduction in the SNP list vote: a crucial additional factor).

It therefore stands to reason that, with the SNP on schedule to do even better in the constituencies next year (possibly winning 70 out of the 73 seats, according to the latest opinion poll) that a list vote for the SNP is completely wasted. Based on preceding discussions, the only sensible alternative is the Greens.

Who could possibly argue with that? Seems perfectly logical to me. Case closed.

There’s just one small problem: this is merely an untested assertion, not based on any actual modelling to see just how true it actually is.

You see, it doesn’t actually matter how logical a claim appears, how much sense it seems to make, or indeed who said it — the only thing that matters is the modelling of data.

In electoral arithmetic, opinions don’t count, whoever they are from.

Modelling various scenarios is the only way we can cut through the copious verbiage and separate fact from fiction.

Opinions Don’t Matter

This important — and vital — distinction merits reinforcement.

‘Opinions’ on this subject are highly subjective, and ‘obvious’ inferences from a casual examination of election data can not only lead to utterly erroneous conclusions, but also lead us up entirely the wrong garden path in terms of political strategy.

As the saying goes: ‘a falsehood is half way round the world before the truth has got its trousers on’.

On the subject of ‘tactical voting’ for ad hoc ‘list only’ indy parties, the message I want you to take home right now is this:

It doesn’t matter what the ‘opinions’ of sundry Indy bloggers and pundits are.

And it also certainly doesn’t matter what my ‘opinion’ is either!

The only thing that matters, is not ‘opinions’, but actual predictions produced by modelling different and specific election scenarios. In other words, sticking data into the ‘black box’ of AMS list algorithms and seeing what it spits out in terms of seat allocations.

The results, surprisingly often, conflict with the ‘obvious’ inferences that seem to be the ‘common sense’ interpretation of precedent. When you model the predictions of those predictions, they don’t always bear out.

The discussions of Indy pundits on this issue is akin to a community of scientists endlessly pontificating on an assortment of theories, but without ever conducting a single experiment!

Understand this in no uncertain terms: a claim made without any empirical testing whatsoever is nothing more than an assertion.

Such assertions are based on an (all too often thoroughly) imperfect understanding of how the seat allocation algorithm works in practice, based on the distribution of votes among parties, the regional effective threshold, and, not least, the realistic electoral prospects of new and untested fringe parties.

In fact, palpably false assertions have, because of their appeal, attained online immortality as factoids, endlessly regurgitated across social media as the ‘undeniable truth’.

Luckily, we have the ability to model these various assertions, and see whether in fact any of them have any basis in electoral reality at all. Testing will allow us to separate facts from fantasy in terms of the allocation of seats in AMS.

The 5% Regional ‘Threshold’

Some countries operating a PR list system have no threshold, some have an official threshold (e.g. 5% in Germany).

The Scottish Parliament has a mixed member system, with additional 56 members elected from eight Scottish Parliament Regions: seven members are elected from each region.

The fact that these MSPs are elected from regional lists and not a single national list is crucial, as it effectively imposes a ‘threshold’ of approximately 5% of the regional vote for a list party to win a single seat in each Scottish Parliament Region. Based on precedent, the harsh reality is that a new, utterly unknown, indy party is unlikely to achieve more than a fraction of one percent, especially if it has never even registered in an opinion poll.

Unfortunately, the need to be sanguine and completely realistic about the prospects of tiny list parties escapes too many on the independence side. Even the Solidarity party of a household name like Tommy Sheridan failed to achieve more than 0.6% of the vote in 2016; the touted RISE scraped a mere 0.5%.

In the last election (2016), the minor parties in total attained only 4.5% — the entire history of Holyrood elections since 1999 shows even this percentage would have been too small a vote share for one party to win a seat (the lowest on record being 5.1% for a Liberal Democrat elected in Central Scotland in 2007).

The fortunes of the Greens in 2016 perfectly illustrates this effective regional threshold of about 5% to gain even one MSP, and over 10% to gain two:

Updated Polling from Panelbase

Previous analysis modelled predicted outcomes for a list only indy party based on the 2016 election results.

The present discussion will show modelling based on the latest Panelbase poll (conducted 30 Jun — 03 Jul 2020). While obviously this is only a proverbial ‘snapshot in time’, modelling scenarios with this poll will at least better inform us of the most likely outcome in 2021, than the results of 2016.

The projected seats from this Panelbase poll are:

(‘Delta’ denotes change from 2016.)

For the purposes of this article, we will use ‘Party X’ as the designator of the potential indy list party, as the analysis is the same, whether it was RISE in 2016, AFI /ISP/Wings now, or whoever next.

When modelling the effect of a swing from the SNP to ‘Party X’ based on the 2016 result, it was shown that not only could a realistic vote for a pop up indy list party not elect any MSPs, it could also endanger pro-indy MSPs: for example, a 5% swing to an Party X candidate in the Highlands and Islands would have seen the SNP lose a seat to the Tories.

For Party X to have gained a seat, it would have required a 15% shift of votes to allow it to take a seat from…the SNP:

Again, in both Lothian and Mid Scotland & Fife, the 15% swing needed to elect one Party X MSP in each region would have been at the expense of the Greens in each case.

And in the South and West of Scotland regions, a 15% shift of SNP votes would have just led to Party X taking over a seat from the SNP and Greens respectively.

But, does anyone see the problem with these discussions?

There is absolutely no way of knowing, prior to the election, exactly how the votes distributed among the parties will play out in the D’Hondt list algorithm — and which way the seats will fall — until after the votes are counted.

This is especially true for parties polling just above or below the 5% ‘threshold’.

So in the cases modelled on the 2016 results above, it would have been impossible beforehand to say in this Region, you should have voted Party X, but in that you should have voted Party Y. Only by number crunching the results, can that be seen after the fact. You cannot ‘game’ or ‘beat’ the AMS list by attempting to vote tactically. I haven’t been the first to say that, and sadly I won’t be the last.

In my current model based on the Panelbase poll, if this was the actual election vote count, around an extra 3,000 votes for SNP in South of Scotland region would see them win the last list seat instead of the Tories. But there’s just no way to know this prior to an election.

It’s high time to start modelling some scenarios….

Swing from SNP to ‘Party X’

Let’s start by modelling the effects of SNP voters defecting to a single new indy list party, which I’ll call ‘Party X’ (whether it be AFI, ISP, Wings, whatever).

Here are the list seat predictions based on the Panelbase poll, showing the distribution of list seats by region:

(The ‘ Delta’ symbol denotes the variation from the Panelbase prediction, not the 2016 election, as we are, of course, specifically wanting to analyse the effects of the redistribution of votes on the prediction based on that poll.)

As we increase the Party X vote, no seats change hands until 9% of SNP voters defect, giving Party X a 4.37% vote share (i.e. four times the combined vote achieved by Tommy Sheridan’s Solidarity and RISE in 2016.)

Such a share of the vote would be completely unprecented for a completely unknown, minor fringe party, but we shall persevere to see what the algorithm predicts.

At 9% swing from SNP to Party X, one seat finally swaps hands, based on this new poll: it is again the SNP which loses a seat, but not to the Tories in the Highlands this time, but rather to Labour in Glasgow. Indy bloc, one seat down.

At a 10pc swing, this loss to Labour is retrieved: but the lost Glasgow seat just goes to Party X instead, along with the SNP’s list seat in Central Scotland Region. Party X on 4.86% of the vote has not increased the pro-indy contingent, just taken two seats from the SNP. No gains for the indy bloc.

At 11% swing, Party X also take a Tory seat in North East Scotland — but the SNP loses a Highlands and Islands MSP to the Lib Dems. In other worths, stalemate, with indy gains exactly matched by indy losses:

It takes a swing of 12% from the SNP to Party X (so it attains almost 6% of the vote, Green territory) for the indy bloc to make overall gains over the unionists, as Party X picks up the Lib Dem seat won from the SNP in Highlands and Islands, and two Tory seats in the South and West of Scotland respectively:

I shall stop there, as this sort of swing is now well beyond any rational expection, and has entered purely into ‘the realms of fantasy’. Further projections might only encourage those convinced that such a massive defection from the SNP to Party X is actually feasible, outside of their own fevered imagination.

Obviously when you model, say, a 30% transfer of the SNP vote to Party X, it will pick up more seats — but then it will be on about 15% of the overall vote, vying with Labour and the Tories. Anyone seriously entertaining swings of 20%, 30%, 40% or ‘just 50%’ from the SNP to ‘Party X’ shouldn’t be left alone with scissors.

Here’s a summary of the seat changes with swings from the SNP to Party X:

In summary: unless and until Party X achieves a level of popular support comparable with the Greens (almost 6%), it can actually present a threat to pro-independence MSPs.

The ISP was initially touted as purely a vehicle to harness ‘wasted SNP list votes’, but recently the tune has changed somewhat, and they now seem to be offering themselves as a distinct political rival to the SNP:

If that is the case, they need to come clean and be honest about this, and abandon the pretence that their raison d’etre is purely to ‘maximise indy list MSPs’.

SNP Swing to Party X and Party Y

Having seen that it takes fully a 12% swing from the SNP to Party X (taking it into Green territory with 6% of the vote), to allow the indy camp to start to marginally increase its majority, can you imagine the danger presented by two competing insta-list indy parties? They could both soak up otherwise useful votes that could help the SNP or Greens, and the swing vote split between them could stop either of them winning a seat.

Let’s pursue the madness further, and see the effects of asking SNP voters to choose between the People’s Front of Judea, and the Judean People’s Front.

With Dave Thompson’s just stated intention to register the AFI as an official party with the Electoral Commission, we are literally now in this territory.

In the modelling presented here, the SNP swing vote will be split almost equally between Parties X and Y, with X having a slight majority of the split. (Readers can decide for themselves whether they want Party X to be the AFI or ISP.)

As a reminder, here is the baseline prediction from the latest Panelbase poll:

At an 11% swing to Parties X and Y, the SNP loses its Glasgow seat to Labour. Indy bloc down one seat.

At 13%, the SNP now also loses its seat in Central Scotland to Labour AND one of its Highlands and Islands seats to the Lib Dems. Indy bloc now down three seats.

There is no further change in seats until there’s an 18% swing of SNP votes (taking the SNP down from 49% of the list vote to just over 40%).

Party X now takes from Labour the seats originally lost to it by the SNP in Central Scotland and Glasgow, so the indy bloc is now just one seat down rather then three!

At a 20% swing, Party Y now steals a seat from the Greens in Central Scotland, so the indy bloc is still behind by one seat, compared to having no Party X or Y seeking SNP votes.

At 21%, Party X finally gains a seat from the Tories in North Scotland, so the indy bloc has finally returned to where it was at the start. The only change from the beginning being that the Lib Dems are up one seat and the Tories are down one.

It takes a 22% swing (slashing the SNP list vote to under 39%) to bring both Parties X and Y above 5.3% of the vote, and for the indy bloc to finally pull ahead of the unionists, going two seats up, as Party X wins the seat from the Lib Dems in Highlands & Islands, originally lost to it by the SNP, and a Tory seat in the West of Scotland.

Here’s the seat table and movement chart based on these swings from the SNP to two minor parties:

The modelled results conclusively demonstrate the insanity of ‘tactical voting’ for pop up list parties as a realistic vehicle for winning those promised legions of indy list MSPs.

Not only do they fail to win any seats without a wholly unrealistic defection of SNP voters, they resolutely fail to remove Unionist list MSPs (one of the frequently stated purposes of a list party). Indeed, if they exceed any realistic expectation and get near the 5% threshold, they can actually lead to the SNP losing seats to Labour and the Lib Dems (based on the current polling).

If you fall for the list party snake oil, and think voting for the AFI or ISP is going to bring about hordes of indy MSPs, you’ll just wake up the day after the election only to find that you were one of the ‘valiant 0.3%’.

SNP to Green

Now we have dispensed with the fantasy snake oil of the pop up list parties, let’s now examine the only two pro-independence list parties actually capable of winning seats: the SNP and Greens.

The constant refrain is that an SNP list vote is wasted, so surely it then makes sense for SNP votes to go to the Greens instead? Let’s find out.

Here’s our starting prediction, based on the Panelbase poll:

It takes a 6% swing from the SNP — to put the Greens up from 7.8% of the vote to 10.7% — for the Greens to win just one extra seat — from the SNP in Highlands and Islands:

A 7% swing allows the Greens to gain one more seat, in Glasgow this time, again from the SNP!

At a 9% swing, SNP defectors to the Greens only allow the Greens to win the SNP’s seat in Central Scotland. Greens up 3: SNP down 3.

It takes an 10% swing from the SNP for the Greens to start picking up seats from the Unionists — a Tory seat in North East Scotland — to put the indy bloc just one seat up from square one:

But you’ll see that to achieve this, the Greens will need increase their baseline support by 60%, to almost match Labour, so we are again veering directly into fantasy land, and it would be fruitless to examine further swings.

Given that many proponents of the ISP refuse to vote for the Greens on the basis of their policies, then they should therefore be voting SNP on the list to keep the Greens out.

Green to SNP

For the sake of completion, what would happen if Green votes go to the SNP? (Maybe if SNP voters defecting to the Greens doesn’t help, maybe Greens should think about voting for the SNP instead?)

Here’s a reminder of the starting line up:

At a 4% swing to the SNP, the Greens lose a seat in Lothian to the Tories:

At an 8% swing, the SNP takes the Greens’ seat in Central Scotland, and the SNP takes a Tory seat in the South of Scotland, back to no overall change.

At 10%, the Tories will regain their seat on the South of Scotland, and the SNP will gain a seat from the Greens instead. The indy bloc is back to one seat behind.

At 12%, the SNP gains a seat from the Greens in Glasgow. Indy bloc still one seat down.

At a 14% swing, Labour will gain the Greens NE Scotland seat. Indy bloc just two seats down.

We see that, despite the supposed handicap of having an all but a clean sweep of the constituency seats (70 out of 73), the SNP is still able to gain seats, as its list percentage increases, albeit not quite enough to match the loss of Green seats.

If the SNP were to increase their list vote to match their constituency vote share of 55%, this would be the list seat allocation:

Thus, by increasing its list vote share 5%, the SNP would now have twelve list seats, triple the baseline prediction from the poll of four seats.

(Obviously in this model, it is predicated on the wholesale transfer of Green voted to the SNP, but the point still stands.)

The total of the indy bloc as a whole is down two, as the gain of SNP seats has been outweighed by the loss of Green ones.

But the modelling already clearly demonstrates nevertheless the falsity of the assertion that an increase to the SNP list vote will not result in more seats and is ‘wasted’.

Well, we’ve got eight extra so far.

Let’s continue the movement of Green votes to the SNP, ad absurdam, purely to see how the SNP seat allocation holds up and to test the validity of the assertion that ‘a list vote for the SNP is wasted, as it can’t win on the list’.

These are how the seats would turn out if all of the Green vote went to the SNP, so there is only one independence party. (Naturally, as with all scenarios, Green votes would not be guaranteed to go to the SNP, but let’s see how D’Hondt crunches the numbers.)

So, with a rise of the SNP vote to 56% (exceeding its constituency vote by 2%), the party has gained every one of the seats projected to have gone to the Greens, without any losses to the Unionist parties.

Not only that, but the SNP has won list seats in all but one region, a supposed ‘impossibility’ according to the pundits.

The pattern of seat change is more readily visible here in the seat table and chart:

Remember, this list seat calculation is on top of the SNP winning 70 out of the 73 constituency seats following the Panelbase poll.

In our present modelling, if the SNP had every single Green vote, it would thus win 14 list seats to add to its 70 constituency seats, leading to an 84 seat representation, and a 39 seat majority.

Obviously this is pure and utter fantasy politics, but these outcomes need to be modelled specifically to demonstrate beyond a shadow of a doubt the lack of truth in the assertion that ‘an SNP list vote is wasted’.

This is clearly not the case.

The assertion that ‘a list vote for the SNP doesn’t make sense’ — on proper examination, itself doesn’t make sense.

The call for #BothVotesSNP is thus a perfectly honest and reasonable one to make for a party looking to maximise its representation at Holyrood and achieve the overall majority that will only strengthen our position regarding indyref2.

Calling for #BothVotesSNP can be made with confidence that an increased SNP list vote will add list seats, contrary to the belief that this is impossible.

At a realistic increase in the SNP vote share, of course, one or two losses to Unionist parties may occur, but this has to be weighed against the strategic impact of a massive SNP majority in Holyrood, to boost the case for indyref2: a rock solid SNP majority will have a much stronger bargaining position than a minority SNP government relying on the Greens, as at present.

These last two scenarios demonstrate, perhaps surprisingly, that the vote distribution between SNP and Greens does not that much alter ‘indy bloc’ seats in parliament, a slight increase to the indy bloc will only occur if the Greens increase their current poll share by 60%, which won’t happen.

Thus the current share of the indy vote between SNP and Greens is about as optimal as it gets, under current polling.

Looking back to 2016, despite the success of the SNP in the constituencies (and the factoid now engrained in the minds of so many Yessers about ‘wasted list votes’), if the SNP list vote had matched their constituency vote in 2016, they would actually have won four or five more seats (four from the Greens and just one from the Tories, but the point stands that a list vote for the SNP isn’t wasted):

Unionists Unite! You Have Nothing to Lose but your Split Vote…

Another boil that desperately needs to be lanced (as it is yet another factoid underpinning the supposed need for a list indy party to ‘tactically vote’ for) is the myth newly doing the rounds that the Unionist parties have a massive built in systemic advantage, because there are three of them and they can thus attain many more seats than they would ever be able to get than if there were only two or even one of them.

So what’s your opinion: how would the predicted results vary in any of the following scenarios?

  • Lib Dem vote goes to LAB
  • Lib Dem vote goes to CON
  • CON takes LAB vote, plus Lib Dems
  • LAB takes CON vote, plus Lib Dems
  • CON takes LAB and Lib Dem vote
  • LAB takes CON and Lib Dem vote

Obviously, these are all utterly absurd scenarios, but they need to be modelled just to deal with the misinformation spread on social media about this issue. The only way to answer this factoid once and for all is to crunch the numbers.

Let’s see how your predictions hold up.

Here again is the baseline prediction:

CON with LAB (LAB takes all LD vote)

The unionist bloc is two seats up.

LAB with CON (CON takes all LD vote)

The unionist bloc is three seats up.

CON with LD (CON takes all LAB vote)

The unionist bloc is two seats up.

LAB with LD (LAB takes all CON vote)

The unionist bloc is three seats up.

CON takes LAB and LD vote

The unionist bloc is four seats up.

LAB takes CON vote and LD vote

The unionist bloc is five seats up.

In every possible permutation of the unionist vote, they are seen to gain seats (albeit marginally) by combining their vote.

In other words, Unionism actually suffers from their split across three parties, getting fewer seats than if there was only one or two unionist parties.

So you can see, the assertion that Unionist parties ‘massively benefit’ from their ‘split vote’ is yet another utterly baseless factoid presented as “common sense” within the independence community.

Back to reality….

But you haven’t answered the question: should I vote SNP or Green in the list?”

A corollary of the inability to game the system for small parties, is that it’s also impossible to game the system for large ones too. It doesn’t matter which side you are on.

As we have seen, based on current polling and with realistic swings, voting for one over the other doesn’t seem to make much difference to the number of pro-indy MSPs, but just redistributes seats between them.

So it all comes down to party preference:

If you are SNP or undecided, vote SNP

If you are Green, vote Green

And if you don’t like either of those options, you have to decide where your priorities lie: whether to prioritise policy disagreements with these parties over independence (and waste your vote in the process), or use your vote wisely by contributing towards electing pro-indy MSPs and thereby furthering the cause of independence.

But understand that if SNP voters had stuck with #BothVotesSNP in 2016, the SNP would have won a majority of seats. And whether you like it or not, Scotland’s appetite for independence is judged by the fortunes — or otherwise — of the SNP alone.

This might not be your preferred choice on Polling Day…but it is your only choice.

Summary

The modelling here demonstrates, beyond any possible doubt, that these ubiquitous ‘opinions’ (which persist in seducing swathes of the indy movement into fantastical notions of ‘tactical voting’ to ‘maximise indy list seats’) are not only entirely erroneous, but self defeating delusions which actually endanger the seats of pro-indy SNP and Green MSPs.

Unless someone can show you actual modelled outcomes to prove their assertions about list ‘tactical voting’, it’s not worth the paper it’s written on.

But they won’t because they can’t.

That’s the beauty of modelling: it doesn’t matter who you are, or what your preconceived opinions are, if you use the same data you’ll get the same results.

‘Tactical voting’ is impossible with the AMS.

Additionally, as we have seen, you cannot ensure Party X candidates win seats only at the expense of Unionists and not the SNP or Greens.

Thus the concept of ‘targeting’ Unionist MSPs with ‘tactical voting’ is even further nonsense.

The brutal but honest truth: anyone who still persists in their beliefs about ‘tactical voting’, in the face of all the algorithmic evidence to the contrary, is as impervious to the facts as any Brexiter or anti-vaxxer.

So, independence supporters, let common sense prevail:

  • If you vote for a no-hoper fringe list party (with no consistent polling showing a bare minimum of 5% support), your vote is 100% guaranteed to be wasted, and will contribute nothing to electing pro-independence MSPs. Not only that, it may even deprive SNP or Greens of seats they could have won with your support.
  • Therefore, in the list, if you truly want to further the cause of independence, you have to vote SNP or Green.

To reiterate: it is impossible to ‘game’ the AMS system through ‘tactical voting’.

All we can do is try to maximise the vote for the only two independence parties that will win seats.

Let no one mistake me: I don’t oppose the concept of tactical voting because I want to exclude other parties, I oppose it because it simply won’t work. It is certainly not ‘cheating’, just pointless and potentially self-defeating for the independence movement.

The best thing the indy movement can do is abandon the ‘tactical voting’ fantasy — and make no mistake it is a fantasy — right now.

Imagine waking up after election night in 2021 to find in every region, two or more pop up list parties getting 2–3% or so of the vote, leading to potential SNP or Green seats going to unionists.

We cannot allow ourselves the navel gazing indulgence of numerous, factional list parties, in what will undoubtedly be the most important election in our lives, the ‘independence election’. #BothVotesSNP does produce the goods.

So, the next time a prolific blogger or opinionated indy pundit proffers you the intoxicating ‘drug’ of a miracle ‘Party X’ to use the ‘wasted SNP vote’ to win swathes of list only indy MSPs through ‘tactical voting’:

‘Just Say NO’

Back away quickly; and

Run, run for the hills…

If you still see indy supporters suffering from the ‘tactical voting’ delusion, please do them a favour and point them in the way of this article.

And anyone still clinging to the snake oil fantasy of ‘tactical voting’ should be locked in a room with pencil and paper, and made to calculate each and every one of the various scenarios, by hand, until the penny -hopefully - finally drops :)

#ListVoteSense

Follow ListVoteSense on Twitter.

List of articles by ListVoteSense.