Wednesday, May 20, 2015

Why "tactical voting on the regional list" is likely to backfire, part 2

The SNP have been criticised in past Scottish Parliament elections for saying that "the first vote chooses your MSP, the second vote chooses your government", but there's more than a grain of truth in that claim. The "second" vote, ie. the list vote, is the more important of the two, because the composition of parliament is in theory supposed to reflect how people vote on the list ballot only.

There's a common misconception that the Holyrood voting system is only "semi-proportional". I once pointed out that an eminent academic had made that mistake on a radio show, and he actually emailed me afterwards to insist that "semi-proportional" is correct! But it's not correct. A semi-proportional system would simply give 30% of the list seats in a region to a party that wins 30% of the list votes. Even if the party in question is already wildly over-represented in the region due to having won all the constituency seats, a semi-proportional system would say : "that's fine, you're still getting your list seats, and we'll just leave the imbalance of constituency seats as it is".

That's not how it works at all. Instead, the list seats are distributed in such a way as to bring the overall composition of parliament (on a regional basis) into line with the percentage of votes each party received on the list ballot. In principle, only the list vote really matters - the constituency vote shouldn't decide the final outcome at all.

The reason it's not quite as simple as that, of course, is that in some situations there aren't enough list seats available to bring the overall result into line. If Labour win 40% of the list votes in a region, but win all nine constituency seats in that region, then with the best will in the world it's not possible to distribute the seven list seats in such a way as to bring Labour down to just 40% of the total number of seats. Labour obviously won't get any of the list seats, but they'll still be significantly over-represented in the region, thanks to their performance in the constituencies.

This is the 'bug' in the system that advocates of "tactical voting on the list" claim can be exploited to vastly increase the number of pro-independence MSPs. The theory goes like this : if we know that the SNP are going to win all the constituency seats in this region, we also know that winning 40% of the list vote will do them no good, because they'll already have all the seats they are entitled to before the list seats are even allocated. Wouldn't it be better if those 40% of voters switched their list vote to another pro-independence party, like the Greens? That way, the voting system would attempt to get the Greens up to 40% of the seats - it wouldn't succeed, but because there aren't enough list seats to go around, Labour would end up with fewer seats than they are entitled to, and pro-independence parties would end up being significantly over-represented.

As we discussed on the earlier thread, this strategy could in theory work brilliantly.  In the real world, however, it is highly unlikely to work, and carries a significant risk of backfiring catastrophically.  As a voter considering going down this insane route, there are two vital questions you would really need to ask yourself first -

1) What if I make the "tactical" switch from the SNP to the Greens on the assumption that 40% of voters are doing the same - but they're not?  What will the impact of my decision be if, as is far, far, far more likely, only a tiny percentage of voters are acting in the same way?

2) What if I make the "tactical" switch on the assumption that the SNP are going to win all the constituency seats in a region - and they don't?  What will be the impact of my decision then?

Here is a fictional example to illustrate the problem.  Suppose an election prediction website was suggesting that the results in one particular region are likely to work out as follows...

Constituency seat 1 :

SNP 43%
Labour 39%
Conservatives 9%
Liberal Democrats 8%

Constituency seat 2 :

SNP 38%
Labour 37%
Conservatives 20%
Liberal Democrats 3%

Constituency seat 3 :

SNP 51%
Labour 24%
Conservatives 18%
Liberal Democrats 6%

Constituency seat 4 :

SNP 42%
Labour 38%
Liberal Democrats 10%
Conservatives 8%

Constituency seat 5 :

SNP 39%
Labour 36%
Conservatives 21%
Liberal Democrats 3%

Constituency seat 6 :

SNP 48%
Labour 28%
Conservatives 17%
Liberal Democrats 6%

Constituency seat 7 :

SNP 47%
Labour 32%
Conservatives 17%
Liberal Democrats 3%

Constituency seat 8 :

SNP 40%
Labour 39%
Conservatives 12%
Liberal Democrats 7%

Constituency seat 9 :

SNP 39%
Labour 35%
Conservatives 20%
Liberal Democrats 5%

List votes :

SNP 41%
Labour 33%
Conservatives 16%
Liberal Democrats 6%
Greens 2%
SSP 1%

Constituency seats :

SNP 9
Others 0

List seats : 

Labour 5
Conservatives 2

Overall seats :

SNP 9
Labour 5
Conservatives 2

Now, it's easy to look at that and think : SNP votes on the list are wasted.  If those votes switched to the Greens or the SSP, they wouldn't be wasted.  But in reality, only a small percentage of people will be thinking in the same way (short of a campaign of mass-hypnosis, that is).  Let's be ultra-generous and say for the sake of argument that 3% of the electorate "tactically" switch from SNP to Green on the list, and a further 1% switch from SNP to SSP.  On the face of it, no harm is done - the Greens and SSP still fall short of the threshold for winning a seat, but the SNP still have their nine constituency seats in the bag, so the unionist parties are no better off.

But what if the constituency predictions aren't quite right?  Look closely at them - the SNP are claimed to be winning all nine seats, but six of them are too close to call, with the SNP ahead by 4 points or less (well within the margin of error).  What if, thanks to the vagaries of first-past-the-post, the SNP don't win nine constituency seats, but only three?  Where is your masterplan then?  Suddenly the SNP need every single list vote they can lay their hands on - and those wasted "tactical" votes for the Greens and the SSP may be doing some real damage.  Here's how it would work out -

List votes :

SNP 37%
Labour 33%
Conservatives 16%
Liberal Democrats 6%
Greens 5%
SSP 2%

Constituency seats :

Labour 6
SNP 3

List seats : 

SNP 3
Conservatives 3
Liberal Democrats 1

Overall seats :

Labour 6
SNP 6
Conservatives 3
Liberal Democrats 1

Without any "tactical voting" (ie. if the SNP had been on 41% of the list vote), it would have been :

SNP 7
Labour 6
Conservatives 2
Liberal Democrats 1

So in this example, "tactical voting" has completely backfired - it has cost the SNP one seat, and gifted it to the unionist parties (more specifically to the Tories).

The best way of looking at the list vote is as your "banker" vote.  You give it to the party that you want to be in government, and you can be absolutely sure that it'll count if it's needed.  It will only be "wasted" if it's not needed - ie. if the party has already won enough constituency seats.  That's something you cannot know in advance.

In 1999, Dennis Canavan stood as an independent in both the Falkirk West constituency, and the Central Scotland regional list.  Anyone who voted for him twice knew, in a sense, that one of those votes would be "wasted".  But it wasn't an irrational choice, because it maximised the chances of him being elected by one method or the other.  Without a reliable crystal ball, that's about the best any voter can do.

102 comments:

  1. I don't think it's possible to second guess the electorate anyway. Successful tactical voting assumes that you cannot benefit from voting in the normal manner AND that you have an excellent grasp of the electoral system AND a fairly good idea of the REGIONAL list makeup. (Forgive the caps, but bold not available.)

    Will the average voter be thinking tactically? No.
    Will the average voter have a grasp of how to vote tactically in a list v constituency situation? No.
    Will the average voter be aware of the electoral mood of the region? Perhaps, but not any more than than normal electoral moods and gossip.

    So you can spend an inordinate amount of time analysing this, when it's far, far easier to get out there and campaign for your preferred party (or parties), and change minds, rather than try to load the system.

    SNPout did not succeed. Mimicking them is probably not advisable. It's easier to campaign against one party than to form an alliance of several (IMHO).

    ReplyDelete
  2. I see what you're saying but why don't you also show how many green and ssp members would be elected if the snp did get the constituency seats, i.e.the polls were on the button?

    This is gamble. You prefer to play it safe - ok, I get that. But it's a bit disingenuous to only show the worst outcome. I've played with spreadsheets too, and I can see situations where we might end up with all SNP constituency seats and 3 or 4 Green seats in the Lothians.

    You might not be a gambler, but other people are. Your opinion is that it's not worth the risk, but others might think the prize is worth it. You are in a very privileged position at the moment - you are the go-to guy for numbers from the pro independence camp. I come to you for decent, humorous, insightful comment, but most of all numbers. Present both sides of the argument. Let us decide. There are regions where this makes more sense - the Lothians perhaps, and regions where it doesn't - probably the highlands. But your blanket denial is too far. It smacks of dogmatism. Trust us. We trust you.

    ReplyDelete
    Replies
    1. Arthur : I'm sorry, but I'm absolutely dumbfounded by that comment. To me, "a gamble" and "a tactical vote" are mutually exclusive concepts. It's a tactical vote if you have a rational reason for believing that a) it has a reasonable chance of improving the outcome, and b) it has little or no chance of making the outcome worse. In a sense you're proving the point I've been trying to make - ie. it's possible to vote tactically in a single-member constituency election, but on the regional list the most you can do is "gamble" and hope that it's not going to be a total disaster. If that's what you're advocating, then fine - but be sure that the people you're trying to persuade fully understand that they'd be trusting in pot luck.

      Delete
    2. You are gambling too - but with the risk of denying a commanding independence majority. I'm not asking you to come down in favour of tactical voting - you don't think it's worth the risk, and I see that. But it's not for you to decide what risks other people take, and you do have a personal responsibility for this that other people don't.

      We come to you for numbers. You can still give your opinion, but don't deny us the numbers. There are some areas where tactical voting on list votes is statistically more likely to work than other areas. That seems to me to be true: I've looked at the numbers from the last election, and I can see areas where we would have increased the pro independence majority. I don't think you can deny this is the case.

      Show all the numbers. That is your responsibility. Add your opinion. That is your right. Trust us. That's only fair.

      Delete
    3. "You are gambling too - but with the risk of denying a commanding independence majority."

      No, I'm not. There is no rational basis for the belief that this utopian scale of tactical voting is possible - ie. you can't just flick a switch and expect hundreds of thousands of non-Greens to vote Green.

      "I've looked at the numbers from the last election, and I can see areas where we would have increased the pro independence majority. I don't think you can deny this is the case."

      I went through the 2011 example in detail in my post the other day. I conceded that there was one region out of eight in which any SNP supporter who voted Green had a beneficial effect, leading to there being one additional pro-independence MSP - albeit by complete chance, in a way that could not reasonably have been predicted.

      But even if you get lucky, that's the kind of effect we're talking about - the odd seat here or there. The idea that tactical voting on the list (or gambling voting) is going to lead to Patrick Harvie being the leader of the official opposition is just utter fantasy.

      To use your own word, it would be irresponsible of me and others not to explain why that is the case, because people are genuinely falling for this nonsense.

      Delete
    4. Things have changed radically since 2011. Using the last election is not a good basis to begin any analysis. The starting point should be current polling and the 2015 GE. Which all indicate an insurmountable position for the SNP in terms of Constituency Vote with vote shares well beyond the level where Tactical Voting can beat the SNP. The most likely statistical outcome is that the SNP will win a majority ONLY from Constituency seats - that's 65/72 and very, very likely.

      Delete
    5. Bottom line is still that those who are trying to merrily parcel votes out that haven't even been earned yet by advocating risking the list are going to end up looking as silly as the Pouters.

      You want votes you work for them. We worked flat out from January to get the westminster result and we're already starting again for Holyrood next year.

      NOBODY is ENTITLED to anyone's vote so if the SSP or Greens want votes they will have to persuade the voters with their policies and campaign.

      There is no other way.

      Delete
    6. Alasdair ; "Insurmountable"? You're seriously saying that a year in advance? I mean, WHAT? The SNP polled just 5% more in the general election than they did in 2011 - when they certainly did not win 65 constituency seats.

      Delete
    7. No I'm not. I'm saying that currently it is insurmountable and if the election was tomorrow it would be sensible. I am also saying that the final decision for each voter should be made on the election day. If you don't think the SNP will win 9 constituency seats in your Region, fine, vote SNP on the list.

      I'm also saying that basing the analysis on the 2011 vote, when we know for certain the picture has changed, is a mistake. It would be much better to use the evidence, and keep an up to date picture.

      Delete
    8. If you don't think the SNP will win 9 constituency seats in your Region, fine, vote SNP on the list.

      Which they certainly won't in several of the regions. Not that that's stopping the evangelicals from telling everyone to do it.

      And bear in mind that even when the SNP did win all nine seats in a region last time, they still got a tenth from the list vote. Playing silly buggers to sacrifice these seats for a possible Green seat or to is nuts in more than one way. First, because it's perfectly possible that the arithmetic will give the seat to the Tories instead, and second because switching a strong SNP government for a weakened one propped up by the Greens isn't a good way to achieve independence.

      Delete
    9. James, you present two concerns:

      1) That a small number of switches from SNP to Green would be detrimental (even if a large number would be advantageous) and realistically only a small number can be expected.
      2) That the SNP might not win as many constituency seats as expected.

      If Labour were indeed to win 6 constituency seats against the SNP's 3 then a large transfer of list votes from SNP to Greens could indeed cost the SNP a seat.

      But that's a huge "if". On the one hand, you are asking us to be "realistic" in your first concern; on the other hand, you are presenting a completely unrealistic situation in your second concern.

      Furthermore, I have yet to see you present numerical evidence that the first concern is independent of the second concern. In other words, if your second concern is unfounded, it doesn't matter how small the switch is from SNP to Green. I invite you to show numbers demonstrating otherwise.

      Delete
    10. "But that's a huge "if". On the one hand, you are asking us to be "realistic" in your first concern; on the other hand, you are presenting a completely unrealistic situation in your second concern."

      Rubbish. It's not only realistic, there's a very significant risk of it happening in the scenario I put forward - if six seats are too close to call, Labour plainly have a good chance of winning all or most of them if the prediction is just slightly out. If you don't understand that, you either don't understand the nature of FPTP elections, or you're bluffing.

      "Furthermore, I have yet to see you present numerical evidence that the first concern is independent of the second concern. In other words, if your second concern is unfounded, it doesn't matter how small the switch is from SNP to Green."

      My second concern isn't unfounded. See above. And your point is incorrect anyway - even if the SNP win all of the constituencies, they might do better than expected on the list (as they did in the North-east in 2011), putting them in contention to win a list seat. In those circumstances, a small "tactical" switch to the Greens could indeed cost us a pro-independence MSP, and gift the unionist parties another seat.

      Delete
    11. This comment has been removed by the author.

      Delete
    12. Well, we'll have to agree to disagree about the prospects of Labour beating the SNP in the constituency vote. I am prepared to put my money where my mouth is: I will give you even odds on any seat you think is "too close to call" between the SNP and Labour (I don't believe such a seat exists in Scotland).

      As to the second point: as ever, show me the numbers.

      Delete
    13. Your first comment is absolutely extraordinary. If you seriously believe that you can already "call" that Labour are going to lose every single seat next year, you should be offering me heavy odds-against on Labour in a seat like Eastwood, not evens. Are you prepared to put your money where your mouth is?

      As for "show me the numbers", the 2011 result in the North-east is, as ever, easy to find via your favourite search engine.

      Delete
    14. I've already put my money where my mouth is: the offer is evens on any seat in Scotland. If you believe that some seats are "too close to call", name one of those seats. If you're right, what have you got to lose?

      2011 in the North East it is, then. What is the number of list votes which, if they had gone to the Greens instead of the SNP, would have decreased the number of pro-indy seats and increased the number of pro-union seats?

      Delete
    15. "the offer is evens on any seat in Scotland"

      As you really should have spotted, I rejected that offer because it was a nonsense. First of all, define what you mean by "calling" that Labour will lose every single seat next year. An American TV network would never "call" a race until they are more than 99% certain of the outcome. Let's be ultra-generous and say that you are merely claiming that you are 90% certain that Labour will lose, for example, Eastwood. That means, if you actually believe what you are saying (evidently you don't), you should be offering me odds of 9/1 against Labour winning Eastwood. If you're not going to do that, I look forward to you going back to that Facebook page and apologising for your cynical attempts to hoodwink people.

      "What is the number of list votes which, if they had gone to the Greens instead of the SNP, would have decreased the number of pro-indy seats and increased the number of pro-union seats?"

      That number is easily available to anyone with the assistance of their favourite search engine, and their favourite calculator brand. You seem to be mistaking me for someone who has the remotest interest in spoon-feeding a troll.

      Delete
    16. "First of all, define what you mean by "calling" that Labour will lose every single seat next year."

      Okay, I think I see the miscommunication here. For you, even describing 90:10 as anything other than "too close to call" is being "ultra-generous".

      I'm not entirely convinced that this is the sense in which an American TV network would use it. You're saying that if one Presidential candidate had an 89% chance of winning the election, and the other an 11% chance, the headlines would read "too close to call"?!

      Anyway, I never used the term "called" in the sense that you're defining it there. What I said was that I don't believe Labour will win a single seat.

      I believe that the individual most likely outcome for each seat is for the SNP to beat Labour. I would agree that this doesn't necessarily make the most likely outcome overall to be that Labour win zero seats (in the same way as even though a 7 is the most likely outcome from rolling two dice, you're more likely not to get a 7).

      But I'm making my bet based not just on the outcomes of recent polling, but also the trajectory, and the fact that I see nothing in Labour's behaviour to indicate that they have understood how to address their problems in Scotland, let alone that they have the political will to do so.

      Anyway, as I say, let's agree to disagree on that. As arthurfaeleith pointed out above, there are two sides to the calculation: your position seems to be that losing a single SNP list MSP would not be worth the risk, even if the benefit were gaining, say, ten Greens at the expense of the Tories and Labour. I do not agree, but I can at least see how someone acutely risk-averse might feel that way.

      Onto the other matter.

      I cannot prove a negative; there is nothing I can show to demonstrate that no number exists. On the other hand, you describe such a number as "easily available"... yet can't be bothered to provide it.

      I invite the reader to draw their own conclusions.

      Delete
    17. "You're saying that if one Presidential candidate had an 89% chance of winning the election, and the other an 11% chance, the headlines would read "too close to call"?!"

      Correct. A contest is not called until certainty is reached. They're even more ultra-cautious than they used to be, after the Florida debacle in 2000.

      "Anyway, I never used the term "called" in the sense that you're defining it there"

      Well, I've no intention of engaging with an alternative, fantastical definition of the term "too close to call". If you don't understand what it means, that's your problem, not anyone else's. It certainly doesn't mean "nothing more nor less than 50/50".

      "But I'm making my bet based not just on the outcomes of recent polling, but also the trajectory, and the fact that I see nothing in Labour's behaviour to indicate that they have understood how to address their problems in Scotland, let alone that they have the political will to do so."

      Translation : blind faith. This is laughable. Utterly laughable. Blind faith is no basis for a "tactical voting strategy".

      "Anyway, as I say, let's agree to disagree on that."

      No, we're not going to agree to disagree on this, I'm afraid. If you continue pumping out this rubbish, I'm going to continue pointing out that you're trying to cynically mislead people.

      "your position seems to be that losing a single SNP list MSP would not be worth the risk, even if the benefit were gaining, say, ten Greens at the expense of the Tories and Labour. I do not agree..."

      Is there anything else I didn't say that you'd like to disagree with?

      "I cannot prove a negative; there is nothing I can show to demonstrate that no number exists."

      What in God's name are you wittering about, man? Do even you have the faintest idea what you're talking about?

      Delete
    18. You've never come across the idea that absence of evidence is not the same as evidence of absence?

      Okay, so we've established that you are at least 99% certain that the SNP will not beat Labour in every seat. So if I lay a £10 bet with you, you'll give me £1000 if the SNP do in fact beat Labour in every seat? I'll take those odds.

      Delete
    19. "You've never come across the idea that absence of evidence is not the same as evidence of absence?"

      Oh yes, I've come across that idea many times. What it's got to do with the subject we've been discussing is rather more of a mystery. I presume there's very little hope of hearing a credible explanation from you?

      "Okay, so we've established that you are at least 99% certain that the SNP will not beat Labour in every seat."

      Are you on drugs? Where the hell are you getting that from?

      Delete
    20. To prove your claim, you need only provide a single counter-example to my assertion that this mythical number of votes does not exist.

      For me to prove my assertion would require me to exhaustively enumerate every combination of outcomes to demonstrate that no counter-example exists, which is of course completely infeasible. Not actually mathematically equivalent to "you can't prove a negative", I suppose, because the problem space is technically finite, but you'd have to be pretty obtuse to be making that argument.

      If I'm wrong about 99%, then, what odds would you give?

      Delete
    21. "To prove your claim, you need only provide a single counter-example to my assertion that this mythical number of votes does not exist."

      Very well. By my calculations, if 2600 people had "tactically" switched away from the SNP on the list, and if 2000 of them had gone to the Greens, and 600 had gone to the SSP, the net result would have been that the SNP would have won one fewer seat, and the Tories would have won one more.

      Having done what you asked me to do, and what you claimed was impossible, I take it you're now going to accept you were wrong throughout this thread? Oh no, of course not, you're instead going to continue denying reality with more mind-bending gibberish. Silly me.

      "If I'm wrong about 99%, then, what odds would you give?"

      Why would I give you any odds? I'm not interested in playing your bloody stupid games, mate, I'm simply pointing out that you can't seem to add up.

      Delete
    22. "Having done what you asked me to do, and what you claimed was impossible, I take it you're now going to accept you were wrong throughout this thread? Oh no, of course not, you're instead going to continue denying reality with more mind-bending gibberish. Silly me."

      On the contrary: I have now run these numbers myself and you are right.

      I stand corrected!

      In my first message to this thread I wrote: "I invite you to show numbers demonstrating otherwise". If you had simply done so, we would both have saved a lot of time.

      I trust you'll be equally ready to admit your own mistakes.

      Delete
    23. "If you had simply done so, we would both have saved a lot of time."

      No, actually, by pursuing a vexatious point, you cost me a great deal of time making the exact calculation. You could have done it yourself if it was so important to you.

      "I trust you'll be equally ready to admit your own mistakes."

      If, yet again, you can't even be bothered to explain what you're actually getting at, I doubt if there's much chance of me "admitting" anything.

      Delete
    24. I forgot to say: thank you for providing the numbers. I honestly do value my mistake being corrected.

      Delete
    25. The point you provided numbers for in the article -- that tactical voting was scuppered if the SNP didn't win the vast majority of constituency seats -- was, to me, much more obvious than the point about a small transfer of votes from the SNP to the Greens. The latter wasn't obvious to me, nor Oliver, nor various others on that Facebook thread and in the comments here. Given that you took the time and trouble to show the former numbers, surely you can see the value in providing the latter ones?

      I thought my meaning was clear from the article I linked to, but in the spirit of not making assumptions about what is obvious and what is not: it's an article about what constitutes "too close to call". (TL;DR -- 70/30 is too close, 80/20 is not too close.)

      Delete
    26. I've skimmed the article, and I can't really see that it even purports to make the point you claim - it's about what does, and does not, constitute a "statistical tie" in an opinion poll. That's an entirely different subject.

      Delete
    27. No, it's the same subject. From the article:

      "In short, a statistical tie is a polling result for which the difference between two (or more) candidates is of the nature we would expect sampling error alone to reasonably explain. Presuming the actual election result between the candidates to be exactly a tie, poll results still will--due to sampling error--differ somewhat. If that difference is in the range of differences that are reasonably a result of sampling error alone, then the poll result comparing the contending candidates is ruled a statistical tie. More informally, we might say that a statistical tie occurs when the poll results lead us to conclude the election is "too close to call.""

      Here is another article which puts this specifically into the context of an American election, and takes poll size into account. From that article:

      "Generally speaking, national polls use sample sizes of about 1,100, which translates to an MOE of 3%.

      If you look at the table, this means that if one candidate is 5% ahead in the polls, this indicates with 95% certainty that the candidate is *actually* ahead (rather than the 5% resulting from sampling error).

      In other words, a 55/45 split in a properly-conducted national poll is, with 95% confidence, not actually "too close to call".

      Delete
    28. "No, it's the same subject."

      What? WHAT? You've just quoted a lengthy passage, which is - as I correctly stated - about statistical ties in opinion polls. We have not been discussing statistical ties in opinion polls. If for some bizarre reason you think we have been discussing statistical ties in opinion polls without me being aware of that, you're going to have to explain why.

      Delete
    29. Ah, I've just discovered the root of the misunderstanding. When I said:

      "I'm not entirely convinced that this is the sense in which an American TV network would use it. You're saying that if one Presidential candidate had an 89% chance of winning the election, and the other an 11% chance, the headlines would read "too close to call"?!"

      What I was thinking about, and trying to describe, was the situation in which opinion polls had one candidate at 89% of the vote and the other candidate at 11% of the vote. I see now that this is not what I described, and that you were talking about confidence intervals, and I didn't pick up on this, which is why you thought I was on drugs.

      I must admit I have been very low on sleep throughout this whole discussion, as I am severely jetlagged. Not quite the same as being on drugs; but close. Anyway it was a genuine misunderstanding, and I'm sorry for the noise.

      Delete
  3. James, I have sent you an email showing all possible results of SNP/Green tactical voting for the 2011 election.

    ReplyDelete
    Replies
    1. Sunshine on CrieffMay 21, 2015 at 10:25 PM

      Showing all possible results? I bet you haven't.

      Delete
  4. Tactical voting wouldn't work because its too complex - UK Labour didn't know they were about to get thumped till after the exit poll - imagine in a AMS system they had told everyone to vote Green on the list.

    Voters can't be delivered as a block, that is why even having a co-operating party standing down in a constituency can backfire as voters seeing there is for example no Green candidate vote Labour.

    If we don't want to become a laughing stalk like SNPout then only vote non-SNP if that is where your heart is.

    And of course there is always a chance of the Greens or SSP doing a Clegg and ditching the SNP in favour of a coalition gov't with Labour.

    ReplyDelete
    Replies
    1. Yes, that's the other point. James Mackenzie made absolutely clear in 2011 that the Greens were equally open to a deal with either Labour or the SNP - and there was no indication that a deal with Labour would have been contingent on constitutional progress for Scotland.

      I'm not aware of the Greens having ruled out a deal with Labour after 2016.

      Delete
    2. If the Greens were to go into the 2016 election without having ruled out a deal with Labour, that would shape how I voted on the list (I gave them my vote in 2011, but would definitely not give them it if I thought they would prop up Labour). You may think that seems wrong-headed but the counterpoint is to ask just how much of what the Greens stand for are LABOUR likely to give them in return for support anyway?

      In this respect, I absolutely agreed with those during the recent campaign who said the leaders should have been way more honest about who they really would and wouldn't deal with if the need arose. Yes, you want to win but not acknowledging possibilities when the outcomes can affect people's lives is ridiculous (and anyway leaves parties open to the charge of dishonesty when they deal with someone in the end, which is needless when we know deals are a possibility).

      Delete
  5. "the first vote chooses your MSP, the second vote chooses your government"

    That's a fair appraisal. Vote SNP on the list if you want an SNP majprity government. Vote Green if you want an SNP-Green coalition (as that is what the Green's are set to promote, and the SNP would likely accept in the circumstances). A similar point applies for the SSP, although it's harder to tell what influence they could have.

    Incidently, if I was eligible to vote next year, I would be voting SSP on the Lothian list, as Colin Fox would be a huge asset to the parliament in my view.

    ReplyDelete
  6. I think you will get folk deliberately trying to muddy the water, every time you have a post on this subject.

    ReplyDelete
    Replies
    1. I suspect you only think that because your understanding of the issue is superficial, and you're confused by an explanation of the deeper ramifications.

      Delete
  7. Labour's internal pollster has released a post-election poll.

    http://www.gqrr.com/uk-post-election-1

    ReplyDelete
    Replies
    1. It's a GB poll, but uses a full-scale Scottish sample that is down-weighted to fit in with the overall sample.

      Obviously the VI is pretty boring because it basically matches the election result, but the "under the bonnet" stuff (i.e. who did you consider voting for) is interesting.

      Delete
  8. If it ism't semi-porportional then it is 3/4 proportional, the SNP got +50% of the seats on 45% of the vote. There should be one national list instead of several regional ones.

    ReplyDelete
  9. I also sense that James has specifically picked a set up of votes which may not be realistic(confirmation bias is a bitch) to show us how stupid we are....

    ReplyDelete
  10. Describing the benefit of Split Voting as a "gamble" or "dangerous" is a complete nonsense given the accuracy of Scottish polling and the known outcome of the 2015 GE. Unless things change, the only reasonable choice is to vote SNP Constituency and then, if your goal is Independence, Green on the list. It offers an additional 20 MSPs over Same Vote and can wipe out the Liberals and Labour, which is a worthwhile goal in itself.

    You can gerrymander the numbers as much as you want in your examples, James, but the reality is that in Holyrood, the SNP can expect 50% vote share in the vast majority of seats and, critically, over 45% of the vote share in almost every seat - and at 45% Tactical Voting is NOT viable to beat them.

    ReplyDelete
    Replies
    1. You seriously think you can predict, in May 2015, how the Scottish electorate are going to vote in an election in May 2016?

      Your other problem is that you're assuming the Green party leadership can be trusted to pursue independence as single-mindedly as the SNP leadership. I don't think that's a safe assumption. You want to sacrifice some SNP seats for (hopefully) rather more Green seats, but Green MSPs are not SNP MSPs and the party has different priorities.

      Delete
    2. Actual no.

      My primary goal is the elimination of the Liberals who have betrayed every single principle of liberalism and the continued dimunition of the Labour Party who have been by far the biggest obstacle to Scottish Independence going right back to 1979 and before.

      First eliminate the enemy. In an AMS system this requires intelligent voting. Every SNP List converted to Green List is a further squeeze on the Liberals and Labour.

      Delete
    3. As far as voting predictions go, yes you can predict and expected outcome most of the time. Scotland has had outcomes in line with polling for as long as I can remember. Obviously it is important that those intending to Split Vote be aware of what is going on and what might change towards polling day. But that's a given, no-one is going to Split Vote without being aware of the political reality.

      Delete
    4. Labour look set to remain at around 25 per cent for the foreseeable future, with the Tories on around 15 per cent. There may be a slight improvement in the Lib Dem vote given that they are no longer in a disastrous coalition with the Tories. They will likely be competing for the last 10-15 per cent or so of the vote with the Greens, SSP. Labour are not going to be either wiped out, or almost wiped out, in Scotland under a PR system.

      They have become one of the two parties to which a majority of the middle class in Scotland, who are staunchly opposed to independence (or even more Home Rule), have turned too in fear of the SNP. As we saw in Edinburgh South, they now have no hesitation in voting Labour, even if they are mostly natural Tories.

      Delete
    5. My primary goal is the elimination of the Liberals who have betrayed every single principle of liberalism and the continued diminution of the Labour Party who have been by far the biggest obstacle to Scottish Independence going right back to 1979 and before.

      First eliminate the enemy. In an AMS system this requires intelligent voting. Every SNP List converted to Green List is a further squeeze on the Liberals and Labour.


      If that's your primary goal then fair enough, but my primary goal is independence. It's dishonest to lure SNP supporters into a list vote that will damage the independence cause on the false prospectus that it will advance it.

      Only a strong SNP will take us closer to independence. If the smaller pro-independence parties also manage to perform well then that's a bonus, but what we must avoid at all costs are tactics which would weaken the SNP in the mistaken belief that this will boost the smaller parties and so the independence cause.

      First, the strategy is fraught with risk, in that there's a real possibility that it will let the unionist parties come in through the middle. Second, it's very unwise to assume that the Green party will support and pursue independence in the way the SNP does.

      So if your goal in life is to kill off the LibDems, fine, but don't do that by lying to independence supporters that your strategy is positive for the cause of independence as a whole.

      Delete
    6. Obviously it is important that those intending to Split Vote be aware of what is going on and what might change towards polling day. But that's a given, no-one is going to Split Vote without being aware of the political reality.

      Really? You'd hardly think it from the number of people running around right now insisting that the split vote is the only way to go.

      It's also very naïve to base the decision on country-wide polling figures, when the regions differ so much. I doubt if a split vote will ever be a sensible move in the South of Scotland, for example. The SNP vote was under 40% there in 2015.

      Delete
    7. Alasdair, you're talking absolute nonsense, and after your pearls of wisdom only a few months ago about Lord Thurso being unbeatable (would you like me to drag those quotes out?), I'd have thought you might have learned to be a little more cautious in the way you express yourself. Contrary to your bizarre claim, opinion polling in Scotland does not have a particularly outstanding record of extreme accuracy (and that is exactly what would be required to make tactical voting on the list viable) - I'm sure we can all recall Patrick Harvie on election results programmes talking confidently about 10 Greens MSPs at the start of the evening, and ending up with two.

      Delete
    8. Patrick Harvie has never quite got over that night. He expected to be in a position to form a majority government with the SNP, and to be offered the position of Deputy First Minister. By morning he was "leader" of a two-member party on the back benches.

      He's hitched his wagon to the independence star, as much for party advantage as anything. Party members were given freedom to choose their own position though, and bear in mind that Robin Harper came out for the union. So far so good, and it's undeniable that many ordinary Green members are very committed to independence, but as far as the party leadership is concerned, I'm not confident in the long term.

      If Harvie saw irresistible party political advantage in coming out against another referendum for example, I think he'd do it. We hand him a veto over that at our peril.

      Delete
    9. muttley, Labour were on 26% of the List vote in 2011 since when, surely you agree, their position has significantly worsened. Labour can quite easily slip below 20% and into the range where higher list votes for the Greens crucify their seat numbers.

      Rolfe, you misunderstand my point in wanting to end the Liberals and Labour. It is because they are the true enemies of Independence that this is needed. They have done far more than the Tories every will to stop Independence, particularly the Labour party. Eliminating them as a force, creating a Tory based Unionist opposition associated with the baggage the conservatives already hold is the ideal scenario to Independence. Especially if the SNP bottle it on a 2018 Referendum.

      James, I already accepted I was a little drunk in that posting and in any case the premise that Thurso might have some strong personal vote was borne out by his drop of just 9ppts.

      Delete
    10. The single factor most likely to prevent the SNP from calling a referendum in 2018, if conditions are favourable, is having lost so many seats to pro-Green tactical voting that Patrick Harvie has an effective veto.

      Delete
    11. "creating a Tory based Unionist opposition associated with the baggage the conservatives already hold is the ideal scenario to Independence. Especially if the SNP bottle it on a 2018 Referendum."

      Says it all.

      Completely out of touch with the real world and as nutty as the Pouters.

      Anyone even considering risking the list had best ask themselves if they think that's the kind of strategic 'genius' they agree with or think remotely likely to happen.

      Delete
  11. I've seen comments elsewhere suggesting a cartel on the regional list (e.g. an umbrella "anti-austerity coalition" whose list comprised a mixture of Green, SSP, maybe even SNP). I know such cartels have been a big thing in other proportional votes, but have you tried playing through that scenario here? Are they even permitted in Scotland? Of course an anti-austerity cartel could be countered by a pro-austerity one, but the fallout from the austerity parties' past partnerships might make that less likely.

    ReplyDelete
  12. All tactical voting is completely irrational. It presupposes that your rationale for action is to bring about the best, or at any rate least worse, consequence. In this case, the goal is to achieve one particular result among alternatives; if the election of your favoured candidate looks unlikely then the election of the second best etc. But the probability of any single person's vote making a difference to the result is so close to zero, that, if your rationale is the best overall consequence, you are better, weighing probabilities with the value of the possible outcomes, just sitting on your arse at home not bothering to go out nor to fill in a postal vote form.

    We have a duty as citizens to vote in such a way as to express our political will, by voting for the party whose manifesto is closest to our political views. And that is what each private individual should do.

    Organisations which know they can persuade their supporters to vote a particular way, that is different. They can make a non-negligible difference by e.g. refusing to stand, if it is a political party we are talking about, and telling supporters to express their support for that party. But for individuals the situation is very close to what is known as a prisoner's dilemma. The rational thing, for people who reject the idea that we have duties as citizens and instead act so as to bring about this consequence or that, is not to vote at all. So tactical voting: GRUY!

    ReplyDelete
  13. It's correct to describe the Scottish d'Hondt system as semi-proportional in the sense that there are fewer List MSPs than constituency. Thus the party that dominates in the constituencies still has part of the inbuilt advantage of FPTP. Labour deliberately rigged it that way because they thought they would always have dominance in the constituency vote. Oops.

    This isn't really the bug that the disputed idea is trying to capitalise on though. One decision that had to be taken with the list seats was whether to apportion them on the basis of party votes for the constituencies, or have a separate vote for the list. There are advantages to that, and it's what was chosen. But it does mean that massive splitting of the vote can theoretically lead to a wildly unfair result.

    Suppose the SNP have 50%, Labour 25%, Tories 15%, LibDems 5% and Green 5% in basic voter support, just for convenience. That spread of votes would be enough to get the SNP virtually 100% of the constituency seats, however the Greens chose to vote in the constituencies.

    But then, if the SNP support (mind control ray, remember?) split off en masse and voted Green on the list, the list calculations would be trying to give the Greens 55% of the parilament! Of course that can't be done, and the SNP already have more than 55% of the parilamentary seats, the constituency ones. Let alone whatever the other three parties are entitled to.

    You'd end up with a party which only commanded 5% of basic voter support having maybe 30% of the parliamentary seats (sorry, I haven't done the actual sums). This isn't proportional and it isn't equitable, and people would quite rightly be outraged.

    That's the bug. The system is not designed to accommodate massive vote-switching like this. And if that did actually happen, it would be a huge incentive to change the system, maybe to have the list seats allocated in proportion to the spread of the vote for the parties in the constituencies.

    ReplyDelete
    Replies
    1. 'Semi-proportional' implies 'half-proportional'. It's clearly much more proportional than that. A few months ago, I had a look through voting systems of the world, in the hope of finding a pure proportional system. I couldn't find a single one. There's always some sort of quirk attached to the system that reduces the proportionality (most typically an artificial minimum threshold for representation). So to contrast Holyrood as a 'semi-proportional' system with other 'fully proportional' systems makes no sense at all.

      Delete
    2. If you take "semi" as meaning "half" then you're correct of course. It's not fully proportional either though. And if there's a massive difference between the way people vote in the constituencies and on the list it may not be proportional at all!

      Delete
  14. People have pointed out ad nauseam the danger of deciding, now, how you'll vote on the list in 11 months time. The assumption that the SNP will undoubtedly be polling at over 50% next May is completely unfounded. Anything could happen.

    The other serious snag is that people have to consider which region they're in. I'm in South of Scotland, where the SNP only won four of the nine constituency seats in 2011, and that was when the party was standing at 45% nationally. We then got another four list seats on the strength of our list vote.

    It's not going to be all that different even with 50% of the vote, country-wide. But this mad notion wants us to throw away our four list seats in the hope of gettin the Greens one or two? What are these people smoking?

    In practice, looking at the spread of the vote last time, the first beneficiaries of a drop in the SNP list vote would almost certainly be the Tories. Way to go chaps, throw away our list seats hoping to gift some to a party that's only lukewarm about independence anyway, and actually end up giving some to the Tories.

    ReplyDelete
  15. This whole idea of treating SNP and Green seats as equally supportive of independence is fatally flawed in any case.

    I know, personally, quite a few Green party members and supporters who are just as committed to independence as I am. I campaigned with them last year. Even they admit that their party leadership is not so committed.

    Even if a plan to sacrifice some SNP list seats in the hope of gaining rather more Green list seats succeeded initially, it might backfire spectacularly further down the line. If circumstances change suddenly, if there is a sudden swing to Yes among the Scottish electorate, we need a strong SNP to take advantage of that and go for a second referendum. I simply do not trust Patrick Harvie's party sufficiently to want to leave him with a veto on that.

    ReplyDelete
  16. We will have countless national polls as it gets closer. If SNP is around 50% they'll take virtually every seat. We'll have ashcroft on the marginals which will provide detail on the difficult ones. We can infer the other seats. We will have certainly national list information which will likely show SNP around low 40s and in this scenario will pick up maximum 1 or 2 seats. What we may or may not have (wings maybe could step in if ashcroft or large scottish polls don't) is regional list seat information. If we can get the last one we have a pretty solid basis to estimate whether we are in landslide constituency seat scenario. If we are confident we are then we need to vote Green on the list. I've always supported SNP but will have no hesitation in doing so and recommending other like-minded people do so if I am beyond reasonable doubt that we are there. At this point I'm informing people of its possibility, saying its only viable in the landslide scenario and explaining the information that we need and hopefully will get to assess whether this is the case. My question to James is do you think the information that I have outlined above will provide us with a reasonable basis to make such an assessment (and I'm erring on the side of caution i.e. H0: Non-landslide H1: Landslide)? Jam

    ReplyDelete
    Replies
    1. If this month's election hasn't demonstrated why opinion polls cannot be used in the way you're suggesting, I don't know what would. In particular, the accuracy level of Ashcroft's constituency polls was appallingly low - he's going to have to go right back to the drawing board. He's doing his best to put a positive gloss on it, but he must know in his heart of hearts that the whole exercise was a flop.

      Delete
    2. I doubt very much whether Ashcroft will plough his cash into Holyrood constiutency polls - they're bloody expensive (you don't get to a billionaire from chucking your money away) and I suspect he won't see the contest as important enough to warrant it. I reckon he'll probably do national polls though.

      Delete
    3. The "beyond reasonable doubt" thing simply isn't going to happen. We've all sat through election nights in either horror or delight as events panned out very differently from all predictions.

      The reason the SNP did so well in 2011 is that the party has always emphasised the importance of voting SNP on the list. It only achieved 53 constituency seats out of 72, well short of the 65 needed for an overall majority. The 16 list seats were vital to delivering the majority and the independence referendum which has so transformed Scottish politics.

      Now people want to abandon this winning strategy and throw away the list votes? It's madness.

      Delete
    4. James

      I take the point on the percentages (and size of the implied standard errors) but the direction i.e. who would win was pretty much correct in most cases? And if not where not? Berwickshire was too close to call, Orkney and Shetland he didn't do, Edinburgh South he had SNP slightly ahead (within margins of error) were late in the day extenuating circumstances and I think Dumfriesshire he had roughly neck and neck too...I suppose then the taking this into account as I accept your point, then the criteria for landslide becomes that little bit harder to gauge on the constituency vote - something approaching 55% for me then.. Jam

      Delete
    5. He had the SNP 11% ahead in Dumfriesshire, Clydesdale and Tweeddale. Look how that panned out.

      Delete
    6. not according to this, well within margins of error and thus not landslide at laest within that seat:

      http://lordashcroftpolls.com/2015/04/latest-scottish-constituency-polling/

      Jam

      Delete
    7. No, you're looking at the wrong poll. There was another one in the same constituency after that.

      Delete
    8. Got us quite cheerful, that did. But I think the unionist rump of the LibDem vote saw it and switched.

      Tactical voting in a FPTP election is alive and well.

      Delete
  17. I am quite dissapointed having read the erudite stuff up above. I had intended, still intend, to vote SNP in my consituency and for Patrick Harvie on the list.

    My priority is, obviously independence, which is where my first vote will lie. But the broader perspective, from my point of view, is that we absolutely must deal with anthropogenic global warming.

    I am now told that my split vote is likely to limit the opportunity to win my heartfelt desire, independence.

    There must be something wrongwith a voting system that mis-represents me to the extent that our good host says it does.

    Or am I missing something?

    ReplyDelete
    Replies
    1. You're overlooking the fact that any electoral system will struggle to allow you to express two deeply-felt desires, unless the same party is the main proponent of both of them.

      If you think it's possible to achieve meaningful action on anthropogenic global warming while Scotland remains in the union, you might want to prioritise your Green preference. However, if you believe as I do that we're powerless to influence this within the UK, you might choose to prioritise your SNP preference. The reason being first that there is some actual hope of success on that front, and secondly that one this is achieved, success on the climate change front will hopefully become more attainable.

      Delete
    2. Rolfe,

      Thanks. I will cast my votes carefully.

      To answer your points, if I may:

      If you think it's possible to achieve meaningful action on anthropogenic global warming while Scotland remains in the union, you might want to prioritise your Green preference.

      Well, obviously I don't. That is why the SNP was my first vote.

      However, if you believe as I do that we're powerless to influence this within the UK, you might choose to prioritise your SNP preference. The reason being first that there is some actual hope of success on that front, and secondly that one this is achieved, success on the climate change front will hopefully become more attainable.

      I hope so.

      Thanks for your contribution, frankly, I am no further forward.

      The SNP are about the most positive folk on GHG emmissions that you could shake a stick at. I will vote for that.

      I will also vote for Patrick Harvie on the list

      Nichola Sturgeon needs opposition worthy of being an opposition....



      Delete
    3. If the SNP fail to reach 65 seats because of this vote-splitting malarkey, and it does in fact benefit the Greens, then Patrick Harvie will not be in opposition, he'll be in coalition, or at least confidence-and-supply. You will have handed him a veto over a second independence referendum.

      Is this wise? Do you trust him to pursue independence give the chance, and not to play politics with it for some perceived party advantage? To be absolutely honest, I don't.

      Delete
    4. To go further with the question of Greens-as-opposition, it cannot be emphasised strongly enough that this is not possible as things stand. Labour have about 25% of the popular vote to the Greens 5% or so. You cannot transform a party with 5% of the popular vote into the official opposition, over the party with 25% of the vote, simply by repackaging SNP list votes.

      Forget hypothetical numbers of what could happen if hundreds of thousands of SNP supporters were suddenly seized by some sort of mind control and voted Green on the list. It's not going to happen. All that might be achieved by this misbegotten campaign is to switch some SNP list seats for some Green seats. Thus potentially swapping a majority SNP government for an SNP/Green coalition.

      This at the risk of weakening the SNP without significantly benefiting the Greens, and allowing Labour or the Tories to come in up the middle.

      Do it if you want to, but be aware of what it is that you're actually doing.

      Delete
    5. Douglas I think what you are doing is fair enough, you genuinely have a reason to want a strong Green showing. I could also imagine a ROC independence supporter splitting their vote SNP\Conservative.

      James is warning against SNP supporters being persuaded to vote Green\SSP on the list on the grounds that it will further the independence cause - it won't.

      It would be great to destroy the Labour Party at Holyrood as they are our biggest obstacle to independence but our best hope is they lose a further 5-10% support as UK Labour lurches ever right-ward.

      Delete
    6. Nobody is suggesting that genuine Green supporters shouldn't vote Green on the list. Of course they want their party to get its fair share of list votes. What is objectionable is when they pretend, falsely, that this is a better way to secure independence than voting for a strong SNP, and so seek to persuade SNP supporters to weaken their own party and strengthen the Greens.

      It's dishonest, and it could cost us independence of enough SNP supporters are unwise enough to buy into it.

      Delete
    7. Douglas, there's definitely a role which the Scottish Government can play (and indeed is already playing) in responding to the challenges of climate change. I'd suggest that the a vote for the Scottish Greens would demonstrate there's public support for stronger action on climate change and that the Scottish Greens would be a strong voice in the Parliament pressing for more significant action to be taken. For example, the SNP announced a moratorium on fracking, but as Alison Johnstone noted:

      “A year ago the First Minister said shale gas was an undoubted opportunity; today the Energy Minister announced a pause but asks us not to rule it out. It is clear that the sustained pressure we’ve been putting on the Scottish Government has paid off, but we do not intend to rest here. - See more at: http://www.scottishgreens.org.uk/news/greens-welcome-government-caution-on-fracking-insist-on-ban/#sthash.bgR7YfhT.dpuf

      Similarly, as Patrick Harvie has written,

      the Climate Change Act requires the Scottish Government to develop an approach to measuring [...] “consumption-based emissions”.

      Just this week, slipping the news out quietly on Budget day, the Scottish Government confirmed that our overall carbon footprint, which includes these consumption-based emissions, has increased. http://www.patrickharviemsp.com/2015/03/the-national-column-on-our-carbon-footprint-this-vow-on-climate-change-is-just-hot-air/

      Delete
    8. That's a fine argument, directed at voters whose primary political allegiance is Green.

      What is not a fine argument is "vote Green to strengthen the independence cause", when aimed at voters whose primary political allegiance is SNP. That's either hopelessly naive, or a cynical, dishonest attempt to boost the Green party at the expense of the SNP and our chances of independence.

      Delete
    9. In this election just past, the Greens (in Scotland and in England and Wales) were arguing against tactical voting. Patrick Harvie, for example, wrote that:

      The problem is deeper than the confusion it creates for the voter, or the imbalance it creates in Parliament. It locks in an assumption which should be alien to a democratic system; the assumption that you can’t afford to vote for what you believe in, but that you must accept the ‘least worst’ option and just put up with it. http://www.patrickharviemsp.com/2015/02/the-national-column-why-tactical-voting-is-settling-for-least-worst/

      He was writing that in the context of FPTP, of course, but I think the underlying sentiments would also be applicable to voting for Holyrood.

      Delete
    10. To be quite honest, I think the underlying sentiment was "please support my part even if we don't have a hope in hell of winning your constituency".

      He seems to have changed his tune somewhat, in relation to tactical voting, now he thinks he's found a way to promote a tactical vote that will benefit the Greens.

      Delete
  18. In the unlikely event of opinion polls close up to the 2016 vote showing the SNP over 55% on a consistent basis, while I wouldn't split my own vote, I'd at least stop telling people the idea was madness.

    I wouldn't split my vote because I'm in South of Scotland, and there's no way this side of the Second Coming that the SNP won't need list seats there. And get them if people simply vote SNP for both constituency and list. Bear in mind that the SNP vote here this month was under 40%.

    Still, it might be a harmless enough ploy in certain regions, IF the SNP is riding impossibly high in 11 months time.

    What I object to is the promotion of the notion, NOW, when we have no idea whether the SNP will even be polling 45% next year, and without any consideration that some of the people listening to this message will live in regions where it will never be a sensible thing to do.

    ReplyDelete
    Replies
    1. I think this is part of the problem and the reason this argument gets so heated. I asked James the last time this came up (or the time before that) if there are circumstances he would consider a split vote and he said No. I take it from this that is also your view.

      Are you so certain about South of Scotland? Have a look at the 2011 results, remember the shift in position, especially the Labour and Liberal to SNP switching. Then consider. What seats will the SNP NOT win.

      The answer, currently, is Ettrick. I expect 8 SNP Constituencies and therefore they will require an additional 13% on their list vote over 2011 to get a single List seat this time. Statistically, I know where I would put my money.

      This indicates that the problem isn't with those advocating a split vote strategy as all I see is an understanding that there is an element of risk and that it should only be decided much closer to the time. In the meantime it is a potential strategy.

      Delete
    2. Rolfe,

      This is not theoretical for me. I live in Nichola Sturgeon's constituency and will vote for her. Indeed I am a member of the SNP.

      On my second vote, I was very impressed with Patrick Harvie over the whole referendum debate. I think the man deserves a place at Hollyrood. I will, other things being equal, give him my second vote.

      I think I recall doing that last time around. And they were both elected.

      If, between now and the vote, the Greens walk away from an independence stance, well, that would be different.

      I tend to view politics in a sort of holistic way. Anyone that hates Westminster, and that.

      Thanks for your comments.

      Delete
    3. "What I object to is the promotion of the notion, NOW, "

      Which is why some of those promoting this on social media are sounding as potty as the Pouters.

      We are supposed to believe Harvie will be well placed to be the SNP's opposition in Holyrood yet where is the evidence of that that focus from the Greens on winning over Labour voters?? Nowhere.

      Until I see real and effective signs of that it's just absurd rhetoric from those who want an unearned vote without having to do the legwork of policy and campaigning for it. To do so this early on is also seriously damaging to what they profess is their own cause.

      Lets also be clear, after the first Indyref we certainly aren't about to take the polling as gospel and anyone who thinks we didn't go into the westminster election worrying about the polling being wrong doesn't know what they are a talking about. Even on election day, despite having very good feedback from the ground we truly didn't believe we had won till we had won. And no wonder. The first Indyref taught us we can't rely 100% on the polling while the westminster election showed us, yet again, that polling can be very wrong indeed. Not only did the tories get back in to power (despite all the poling) but the farce of SNPouters proved beyond all doubt that those who place all their eggs in the tactical basket can end up looking like fools.

      Delete
    4. I can vote, as a second preference, for Patrick Harvie, as a person. That is not to say that I think the Greens have a chance of being the main opposition. It just means that they should have a voice.

      What's so hard to understand about that?

      Delete
    5. Well said, Mick.

      There's also a degree of confusion about what all this is supposed to achieve.

      First, the idea is that we sacrifice some SNP list seats in favour of (hopefully) a larger number of Green MSPs. This is theoretically possible, but vulnerable to serious backfiring. It's entirely possible that the SNP will be weakened by the ploy, but the main beneficiaries turn out to be the unionist parties.

      It's also built on a questionable premise, that all "pro-indy" MSPs are of equal value to the independence movement. They're not. SNP members may be reasonably assumed to have independence as their top priority when push comes to shove, and our best chance of getting a second referendum is for Nicola Sturgeon to be in a strong and commanding position when and if the opportunity arises. The Green party has other priorities and some of its senior members aren't even in favour of independence! If we weaken Nicola's position (to, no doubt, the jeers of the unionists at the SNP's poor showing on the list vote) and force her to rely on a coalition partner with lukewarm support for independence, we could deal the entire indepedence movement a fatal blow.

      The other idea that's been mooted is to "vote for the opposition" on the list. This assumes that the SNP will comfortably top 65 constituency seats (something I will believe when I see it) and so will not require Green support to continue as a majority government. The idea seems to be to transfer so much of the SNP vote over to the Greens on the list, that they get more list seats than Labour and form the official opposition.

      If the Greens could achieve this on their own merits, by attracting support from former Labour (and LibDem etc.) voters, I'd be thrilled to bits. But they show no sign of campaigning for that. Instead they attempt to hoodwink votes from SNP supporters.

      Be very clear, the prospect of the SNP taking north of 65 constituency seats next year is by no means guaranteed. It's very dicey. But even if they did, the prospect of sufficient SNP voters falling for this ploy to give the Greens more seats than Labour is tooth-fairy land. You cannot turn a party with a basic 5% or so support in the general electorate into the official opposition, over a party with over 20% support, by repackaging list votes.

      You'd need a mind control ray. And if you somehow acquired one, the electoral system would be altered so that such a grotesque parody of democratic choice could never happen again.

      Delete
    6. Douglas, there's nothing hard to understand about that. Just so long as you realise what you're doing, and that is potentially weakening Nicola Sturgeon's position in Holyrood and leaving her dependent on the support of a party which isn't wholeheartedly committed to independence.

      If you prioritise the Greens having a voice, above having a strong SNP when push comes to shove (and no doubt many people will), then that's how you should vote.

      Delete
  19. I will certainly not split my own vote, no matter how high the SNP is riding. I understand people who are not specifically SNP supporters going for a different party on the list, but not the attempt to sell the idea to people whose primary political allegiance is SNP.

    If you think you can predict how constituency seats will go, 11 months in advance, you're a lot braver than I am. And even there, you don't expect a clean sweep. I want as many SNP MSPs as possible in Holyrood, to maximise the chance of another independence referendum. I expect the Greens to campaign on their own account, and I wish them well so long as they campaign honestly. I do not expect them to try to boost their own party at the expense of the SNP, which is what they currently seem to be doing.

    Patrick Harvie is not going to walk away from an independence stance before the 2016 election. His positioning on the issue has been shrewd and has resulted in a flood of new members and a higher profile. He's currently trying to capitalise on that by hoodwinking SNP supporters to desert their own party and boost his. That ploy would be killed stone dead if he changed his stance. He would also piss off most of his new members.

    My concern is for events after the election. If the SNP comes out of it badly because its support melted away on the list, and it finds itself reliant on Green support, I fear for the future. If Patrick Harvie sees advantage in withholding his party's support for a second independence referendum in that situation, we're screwed. And I don't trust him. His priority is his party, not independence for Scotland.

    ReplyDelete
    Replies
    1. Sorry, that was a sort of composite reply to Alasdair and Douglas.

      Delete
  20. I think the events of 7th May have gone to a lot of people's heads. It was heady stuff right enough, and if anything it demonstrated that the people pointing out that the opinion poll methodology being used was likely to be underestimating SNP support were right.

    That's two "transfigured nights" we've had now, one in 2011 and now this one. But remember 2012, when the SNP did OK but not as well as the preceding hype had suggested. And remember all the other nights we sat up watching disappointment after disappointment. 1992 and 2003 spring particularly to mind.

    There's no law of nature that mandates the transfigured nights will go on indefinitely. Eleven months is a long time in politics. Opinion polls can be misleading, and public opinion can swing at the last minite.

    The very last thing I want is to be sitting up during the night of 5th May next year, watching the SNP narrowly fail to take seats we'd been certain of winning, and then have our MSP numbers collapse because our list vote collapsed. That would spell absolute disaster for the independence movement, irrespective of how well the Greens performed.

    ReplyDelete
  21. A wasted vote is any vote that did not contribute to a candidate being elected, or to put it another way, the party would have won the same number of seats with this many fewer votes. Here are the number of wasted SNP and Green votes for each region in 2011.

    Central Scotland
    Wasted Green votes: 5,635
    Wasted SNP votes: 2,241

    Glasgow.
    Wasted Green votes: 2,065
    Wasted SNP votes: 19,208

    Highlands and Islands.
    Wasted Green votes: 9,402
    Wasted SNP votes: 3,402

    Lothian.
    Wasted Green votes: 3,144
    Wasted SNP votes: 110,953

    Mid Scotland and Fife.
    Wasted Green votes: 10,914
    Wasted SNP votes: 369

    North East Scotland.
    Wasted Green votes: 10,407
    Wasted SNP votes: 2,585

    South of Scotland.
    Wasted Green votes: 8,656
    Wasted SNP votes: 1,352

    West of Scotland.
    Wasted Green votes: 8,414
    Wasted SNP votes: 21,320


    And finally, the Scotland total.
    Wasted Green votes: 58,310
    Wasted SNP votes: 161,430

    ReplyDelete
    Replies
    1. And these are the number of additional votes the Greens and the SNP would have required to win one additional seat in each region in 2011.

      Central Scotland
      Extra Green votes needed: 6,395
      Extra SNP votes needed: 29,169

      Glasgow.
      Extra Green votes needed: 8,412
      Extra SNP votes needed: 355

      Highlands and Islands.
      Extra Green votes needed: 378
      Extra SNP votes needed: 19,138

      Lothian.
      Extra Green votes needed: 11,515
      Extra SNP votes needed: 37,637

      Mid Scotland and Fife.
      Extra Green votes needed: 2,052
      Extra SNP votes needed: 34,339

      North East Scotland.
      Extra Green votes needed: 2,388
      Extra SNP votes needed: 77,387

      South of Scotland.
      Extra Green votes needed: 5,628
      Extra SNP votes needed: 21,594

      West of Scotland.
      Extra Green votes needed: 4,805
      Extra SNP votes needed: 1,665

      In retrospect, the best strategy would to vote for the party with the lowest number of extra votes needed, but it is impossible to know which that it in advance.

      Delete
    2. We don't have a retrospectoscope.

      The system as it is set up is intended to allow voters to vote for the individual candidate they favour to represent them at constituency level, and then, in the possible event that their main party allegiance is to a different party (maybe they hate the candidate their preferred party has fielded), to vote for their preferred party to be the party of government.

      Do it that way, and it works out fine. Try to second-guess it by gaming complicated scenarios about what you think other people are doing in the polling booth that day, and it will come and bite you on the backside.

      Delete
  22. Rolfe,

    I will have prioritised my vote by giving Nichola Sturgeon my undying loyalty which she undoubtably deserves. There is no way anyone else is getting my first preference vote. That, as they say, is in the bag.

    Unless polling reverses dramatically, we will win the vast majority of constituency level seats, and perhaps some others.

    I agree with what you say:

    There's no law of nature that mandates the transfigured nights will go on indefinitely. Eleven months is a long time in politics. Opinion polls can be misleading, and public opinion can swing at the last minite.

    But, assume they do?

    I'll stand corrected, but the present voting system gave both Nichola Sturgeon a seat and Patrick Harvie a seat, IOt over-represented the Tories, and, heavens is likely to over-represent the Party that was once Labour.

    It sems to me that the good folk that gave us this voting system intended us to use it on first , or second principles, that we would vote for who or what we believed in, and that tactical voting wouldn't work.

    I may be wrong,but telling people which party or which candidate they should vote for is, well, just not what it was set up to do.

    ReplyDelete
    Replies
    1. It sems to me that the good folk that gave us this voting system intended us to use it on first , or second principles, that we would vote for who or what we believed in, and that tactical voting wouldn't work.

      I may be wrong,but telling people which party or which candidate they should vote for is, well, just not what it was set up to do.


      Exactly. The good folk who designed the system intended it to allow voters to select the individual candidate they most wanted to represent them at constituency level, but then still to be able to vote for their preferred party of government even if it wasn't the same as the candidate they chose for the constituency. That's how people understand it, and that's how they use it.

      Telling people to do something different, that is to vote for the party they don't primarily support, is absolutely not what it was set up to do.

      It is because people are telling others to use the system in the way it wasn't set up to be used, that we're having this conversation in the first place.

      Delete
  23. Sorry but it's impossible to take seriously anyone who says, "giving Nicola Sturgeon my undying loyalty which she undoubtably deserves" while using the phrase "Nichola Sturgeon needs opposition worthy of being an opposition...." and then "That is not to say that I think the Greens have a chance of being the main opposition."

    This is beginning to sound very like we are being trolled here and trolling worthy of the Pouters.

    Hence..

    "I may be wrong,but telling people which party or which candidate they should vote for is, well, just not what it was set up to do."

    Yet this is precisely what these fringe people on social media are doing by demanding an unearned Green vote from SNP members to the Greens who self-evidently do not prioritise Independence like the SNP do.

    Bottom line is still if you take a risk on the list you will almost certainly end up looking as stupid as the Pouters.

    Vote for the party that earns your votes and is closest to your priorities and policies. Be it Independence or the Environment.

    Don't waste your time trying to second guess a Labour voting system that will backfire on you.

    If you want the Greens or the SSP then vote for them using both votes. If you want the SNP then vote for them using both votes. Simple as that.

    ReplyDelete
  24. "the first vote chooses your MSP, the second vote chooses your government"

    James says the SNP were criticised for saying that, but it's exactly right.

    You can play around with the constituency vote if you like. It's FPTP and like any other FPTP vote it's susceptible to tactical voting. If your preferred party (say Green or SSP) doesn't have an earthly in the constituency (or isn't standing), you can give your vote to the candidate with a realistic chance whom you'd prefer to see elected. Or if you can't abide the candidate your preferred party has put up for the seat, you can withhold your vote from them personally but still support the party on the list. That's fine. The system allows that.

    What you cannot, mustn't ever even dream about, is not voting for the party you really support, on the list. The system isn't designed for that, and it's capable of biting you hard on the backside if you try.

    I remember 1999. I knew about the voting system and was prepared to go in and vote SNP twice. I was a little concerned about the rhetoric about the "second vote" which seemed to give the list vote a subordinate status when it was arguably the more important one, and the possible implication that is was a second choice vote. I was worried SNP supporters might vote SNP in the constituency then believe they had to vote for someone else on the list.

    In the polling booth, a new anxiety arose. Dennis Canavan's name was on the list paper. I hadn't realised, but my constituency was in the same region as the one he was pitching for. I liked the guy. I had this sudden thought, "I've voted SNP on the other paper, I could vote for Dennis now!"

    The thought itself scared me. I knew the SNP candidate hadn't a hope in hell in my constituency (let's just say that the Labour candidate was one Jack McConnell), and that the sole purpose of my presence in that polling booth was to vote SNP on the list. I was appalled by my momentary attraction to a list vote for Canavan.

    I understood the system though, and put my cross in the right box. I was simply terrified that the SNP would lose vital list votes either through people thinking they had to express a second preference, or through people thinking they'd give Canavan a shout out.

    I needn't have worried. Most people got it right. I can't remember the exact figures, but there wasn't a lot of slippage between the constituency and list votes. The SNP did OK, and Canavan got in anyway. Phew!

    The SNP has always made a very strong point of urging its voters to vote SNP on the list. That was what won the 2011 election. The party's list vote was only 2.9% down on its constituency vote. Sixteen list seats made the difference between another minority government and an overall majority. (Labour on the other hand shed 17% of its constituency vote, and as a result wasn't fully compensated for all the narrowly lost constituencies.)

    The idea that this winning formula should be, not just abandoned, but turned on its head, is ridiculous. It makes Russian Roulette look sensible. The idea that SNP supporters should deliberately weaken their own party to obtain seats for a different party, thus leaving a weakened Nicola Sturgeon potentially at the mercy of Patrick Harvie when we need a swift and decisive move for another referendum, is frankly terrifying.

    The SNP got 69 seats last time. There are only 73 constituency seats. We're not going to get Orkney, or one or two in the South of Scotland. Do you feel lucky, punk? Just imagine the jeers and triumphalism of the unionists if we come out of 2016 with fewer seats than in 2011.

    If you want an SNP Scottish government and a shot at another referendum if favourable circumstances present themselves, you can't possibly take the risk of not voting SNP on the list. The very act says you don't want an SNP government!

    ReplyDelete
  25. Sorry to make your schema work you have to go from one extreme, all the seats, to another, only 1/3 of the seats. In science we say that such examples are 'not physiological'. And all your assumptions that no big campaign is possible for list votes to go elsewhere, well RIC might just run one and persuade a significant number of the apathetic and disaffected in the schemes to take part for the first time. What then James?

    Note 24% of SNP voters in the GE were first time voters. You and I both know they were not all 18yo's. What if something similar happens next May? What if we get another tsunami? That seems to me to be a very likely, physiological possibility. Basically the organism gets a decent meal or breeds with a success towards the upper bound of the normal distribution.

    Your examples assume business as usual but we are not operating in business as usual anymore in Scotland and that opens up avenues that were not previously possible using new voters. And besides, if they are not SNP members they can give their list vote to whoever they fucking well please. Since the vote is secret even SNP members can do that and I know several who are determined to do exactly that. And I will not be telling on them as I'm not in the SNP.

    ReplyDelete
  26. Of course anyone can give either of their votes to whomsoever they please. Bear in mind though that when people advocate batty and self-destructive "tactical voting" proposals, others are just as free to point out how batty and self-destructive these proposals are.

    ReplyDelete