Friday, September 20, 2019

It's quiz time!

Somebody took me to task about an hour ago in the comments section, sternly telling me "James, you don't understand the electoral system".  He then proceeded to 'explain' the electoral system to me - but there was just one snag.  His description bore absolutely no resemblance whatsoever to the Additional Member System, or indeed to any other voting system that is used in this country!  I think possibly he'd "reverse-engineered" some of the modelling he'd seen on Wings and assumed there must be a fiendishly complicated formula involved.

It does seem to me that there's still an awful lot of confusion out there about how AMS works, so to put that theory to the test, here's a little quiz.  There are only three questions, and the answers are at the bottom of the post.  (No peeking in advance.)  If you get all three right you are officially an AMS Grand Master.

Question 1....

"There is a cap on the total number of list seats any large party can win, regardless of how many list votes they take."  Is this statement TRUE or FALSE?

Question 2....

"Small parties with enough votes are awarded a proportion of seats that is much larger than their proportion of the vote."  Is this statement TRUE or FALSE?

Question 3....

If it was possible to successfully game the system by voting tactically for a small pro-indy party on the list, would that party be more likely to take a LARGER or SMALLER percentage of seats than its list vote would normally justify?

*  *  *

*  *  *

*  *  *

ANSWERS:

Question 1.  The statement is FALSE.  If any party was to theoretically take 100% of the list vote, they would take every single list seat, and if they had swept the board in the constituencies as well, they would take 100% of the seats in the entire parliament.  As with any proportional representation system, it's much harder to win a large number of seats - you're not going to win a landslide majority of seats on 35% of the vote, which is something that can easily happen under first-past-the-post.  But there's no cap to directly prevent big landslides from occurring - you can win any number of seats providing you have enough list votes.

Question 2.  The statement is FALSE.  It's possible for a small party to get a slightly bigger proportion of seats than its proportion of votes, but the emphasis is on the word 'slightly', and there's certainly no in-built advantage for small parties.  Quite the contrary, in fact - it's the largest single party that is most likely to be significantly over-represented, due to winning an excess number of constituency seats.

Question 3.  The answer is SMALLER.  This may seem counter-intuitive, but it's actually quite logical when you think about it.  The whole game-the-system theory (which as you know I don't think is workable in practice) depends on the largest pro-indy party doing exceptionally well on constituency seats, but not being competitive at all on the list ballot.  If that happens, there probably wouldn't be enough list seats available to bring any of the other parties up to quite the level of overall representation that their share of the list vote would warrant - and that applies just as much to the smaller pro-indy party that is the recipient of tactical votes as it does to all the other parties.  So although the pro-indy parties in combination might end up being over-represented, the party that receives tactical votes might paradoxically end up being under-represented.

63 comments:

  1. LOL at Wingnuts who think this post is about the Rev.

    ReplyDelete
    Replies
    1. I know, I'm starting to think they're losing the plot. This is a post about the voting system, it's got nothing to do with Stuart Campbell, but for some reason it still seems to be 'triggering' them. Oh well, sometimes all you can do is shrug.

      Delete
    2. It's a bit strange that when the SNP is riding high, we get people complaining that they're slacking, ditching independence, needing to be lose votes to another party ...

      Delete
    3. SNP votes should be for both constituance and list votes. After independence vote for whoever you want. Don't bugger independence.

      Delete
  2. Given that the total number of seats is fixed there *is* a cap on the number of seats any party can win. But I know that's not what you meant.

    ReplyDelete
    Replies
    1. I thought that too. Of course, anyone with a decent memory might recall that, in 2011 the SNP won all ten constituency seats in the North East and followed that up by taking a regional seat there as well. Not much evidence of a cap there.

      I remain of the view that a Wings party is only viable if the SNP forfeits its right to be Scotland's party of independence

      Delete
  3. For every single analysis which ends "In this scenario, Wings would have won "x" seats", you can replace the word "Wings" with "Solidarity" or "RISE", and it would be just as true, and just as imaginary. The ONLY advantage that a Wings party would have over RISE or Solidarity is the huge footprint that Wings has among yes voters, but there is no way of knowing whether that will translate into votes. Until Stu does his polling, nobody knows what the number of vote switchers is likely to be, and even then, as James has pointed out, it ALL depends on the question asked, as to how accurately that poll will translate into votes. PS I got two and a half out of three in the quiz. For question 3, I answered "no way of knowing", so, I'm claiming a half point.��
    Alex Birnie

    ReplyDelete
  4. Two points:

    First, our version of AMS is only *broadly* proportional. The devil is in the mechanism for allocating the regional seats. The method used is the d'Hondt method ("The total number of seats in the Parliament are allocated to parties proportionally to the number of votes received in the second vote of the ballot using the d'Hondt method." - Wikipedia), and the d'Hondt method skews the result. Here are the votes, vote share and seats for the 2016 Holyrood election:

    SNP: 1059897 votes, 46.5%, 63 seats
    Con: 501844 votes, 22.0%, 31 seats
    Lab: 514261 votes, 22.6%, 24 seats
    Grn: 13172 votes, 0.6%, 6 seats
    Lib: 178238 votes, 7.8%, 5 seats
    Others not shown
    Tot: 2279153 votes, 100.0%, 129 seats

    Con and Lab have the same vote share, but one has half as many seats again as the other: Lab has been promoted.
    Green should have less than 1 seat by vote share, but gets 6: it has been boosted massively.

    Here are the results of the same election in Central Scotland:

    SNP 47.7% vote share, 0 seats
    Lab 24.8% vote share, 3 seats
    Con 16.1% vote share, 3 seats

    The SNP's 48% vote share gets it no seats, so SNP is capped. Con has only two-thirds of Lab's vote-share, but gets the same number of seats, ans so is promoted.

    Second, your beliefs (I mean you, James) leave you with two self-contradictory views. If the system is perfectly proportional, then Wings standing cannot in any circumstances reduce the number of indy seats (if it takes votes from other indy parties, the number of indy votes and hence the number of indy seats remains the same). If the system is not perfecly proportional (and it fact it isn't), then Wings can do damage. You believe that the system is perfectly proportional, but Wings can do damage.

    I have to say I doubt your competence, James.

    ReplyDelete
    Replies
    1. "The SNP's 48% vote share gets it no seats, so SNP is capped"

      Sorry, but there are a total of 15 (9C+6) regional in the Central Scotland Region. These are allocated on the regional PR list basis taking into account who already won 9 of them via the constituency vote.

      This was the 2016 list vote % result minus all shares for parties who didn't make the 5% threshold:
      54% SNP
      28% Lab
      18% Con

      The 15 seats were allocated based on the above as follows:
      9 (60%) SNP (via constituency wins)
      3 (20%) Lab
      3 (20%) Con

      Which is about as PR as you can get, although you can see the SNP has done rather well. It got 60% of seats in the region on 48% of the total vote and 54% of the 'above threshold' vote.

      The only 'cap' on the number of seats won here was the number of people that voted SNP.

      Delete
    2. If you want the SNP to get more seats without more votes, you want unfairness / something that is anti-democratic.

      Delete
    3. "I have to say I doubt your competence, James."

      But luckily no-one needs to doubt your brass neck, Derek. You've now left three comments on this blog over the last 24 hours, all three alleging that I don't understand the voting system, and all three comically leaving no room for doubt that you don't understand the voting system yourself. It's not even that you have a 'hazy' understanding of it - there are huge fundamental aspects of it that you've got completely wrong. It looks like you've spent two minutes on Google search, and decided "aha, so *that's* the way it must work, I am now a world-leading expert!"

      First of all, for some bizarre reason you seem to think the Additional Member System is a pure d'Hondt list system. Er, no it's not. The clue is in the title. Your first example gives only the constituency vote shares and points out that the seat allocation isn't terribly proportional to those vote shares. Hmmm, that'll be because overall seat allocation is based on the *list* vote, not the constituency vote, Derek.

      Your second example does appear to use list vote shares, but somehow manages to completely ignore the existence of the constituency seats that the list seats are topping up on a compensatory basis, which is the whole basic principle of the Additional Member System!

      Thirdly, you appear to believe that the d'Hondt formula is "skewed" to favour smaller parties. No it isn't. If anything it favours larger parties at the margins, which is the reason why the Blair government selected it in the first place. It's the Sainte-Lague formula that favours smaller parties.

      Fourthly, and as I've now pointed out to you four times, AMS does not impose any "cap" on larger parties.

      My suggestion to you is that if you have no interest in educating yourself on how the voting system works, perhaps you should leave the discussion to those who understand it.

      Delete
  5. Wings has a large profile among YES campaigners, but much less so among voters.
    That could change of course, but the problem could be how the press would portray a party whose leader lives in England and has no democratic structure.
    Too many imponderables at the moment.
    Also any perceived criticism of SNP strategy by a new party could rapidly lose it support as most YESSERS are also SNP members or supporters.
    So many imponderables for me anyway.
    Events could pre-empt all this, such as Indyref2.

    ReplyDelete
    Replies
    1. Where does the leaders of the Tories and Labour live? London.

      Delete
    2. Only idiots would vote for a party that does not have a serious manifesto or stands in constituancies.

      Delete
  6. The main advantage that Wings has over RISE and Solidarity is its brand. But the brand involves a social media history longer and more colourful than the coast of Argyll. So a rebrand as the Concerned Absolutely Non-Bathist Party probably puts it back down amongst the minnows.
    I wonder if there's a case for various independent and 'straight-taking' pro-indy mavericks of different kinds, including Rev Stu, standing in different regions. Of course the Greens would be highly unlikely to step aside in all or most areas.
    Model that one.

    ReplyDelete
  7. Horses for courses in each region might be more effective.
    All under the banner of YES independence would provide clarity.
    Mind you Wings could be sponsored by the RSPB!

    ReplyDelete
  8. Why don't the SNP (with Green backing) just change the Holyrood electoral system to FPTP and they'd have 81% of seats?

    Given both the Greens (in the past) and the pro-Wings party both say the reason to vote for them is to 'maximise pro-indy seats', they should support this plan wholeheartedly.

    ReplyDelete
  9. This comment has been removed by the author.

    ReplyDelete
  10. Let's try again. The Wings proposition is this:

    - a party's final total seats should be in proportion to its list vote-share - that's what d'Hondt aims to do;
    - if a party gets its full quota from the constituency seats (a plausible result, given current polling), then d'Hondt won't give it any list seats;
    - those list seats have to go somewhere, so d'Hondt has a problem;
    - d'Hondt solves that problem by giving those seats to another party, producing a non-proportional result.

    I'll model that with some easy numbers - a parliament with 60 constituency seats and 40 list seats. There are two parties, Nats and Yoons. The Nats take 50 constituency seats (and their constituency vote-share doesn't matter, because it doesn't come into the calculations); the Yoons take the other 10. Their list vote-shares are 50% each. The discrepancy between vote-share and seats is a known feature of FPTP.

    d'Hondt now has to allocate 40 list seats, but it can't give any to the Nats (they alreasdy have their vote-share). So it gives them all to the Yoons. Final result: 50 Nats, 50 Yoons, perfect list-vote proportions. That's how AMS corrects FPTP.

    Now I'll model it again with Wings taking 10% of the list vote from the Nats (and not standing in the constituencies). The constituency seats are still 50 Nats, 0 Wings, 10 Yoons. The list vote-shares are now 40% Nats, 10% Wings, 50% Yoons.

    d'Hondt goes to work. It can't give any seats to the Nats, because they've already got their quota, so it divides the 40 seats according to the vote-share of Wings and Yoons. That vote-share is 10%-50%, so for every seat it gives Wings, it gives 5 to the Yoons; in other words, one-sixth of the seats go to Wings (7 seats) and five-sixths to the Yoons (33 seats). So there's a net transfer of 7 seats from Yoons to Indy. Result: Indy 57, Yoons 33, and an Indy majority of 24. Not proportional - it should be 40 Nats, 10 Wings and 50 Yoons according to vote-share - but you can't take away the Nats' 10 constituency seats by which they've exceeded their quota.

    The logic is impeccable - in those two scenarios that's what the outcome will be, and the opportunity is big.

    Before you tell me that this is a model and only imaginary, let me remind you that we run our lives by models and imaginary numbers. They answer the question What if...?, and it's only because people have had the forethought to deal with that question that our houses don't fall down on us, our oncologists cure cancer and we take an umbrella when it might rain. Psephology is all about imagining the numbers in future elections, given what the real world tells us through polls; every time a psephologist tells us what the polls indicate, they're imagining some numbers and presenting a model.

    You're a psephologist, James, so you need to know about d'Hondt. And you need to do some numbers.

    ReplyDelete
    Replies
    1. Derek, my patience is not inexhaustible. I've just posted a detailed reply to your earlier comment in this thread, in which I've explained in considerable detail the multiple ways in which you've fundamentally misunderstood the Additional Member System. Please read that reply and stop making a fool of yourself. At the moment you're coming across as a six year old kid trying to explain to a professional snooker player how to play snooker.

      Delete
    2. The main problem is the blind faith that the FPTP part of the election isn't vulnerable to tactical voting. Easy enough to agree to stand a paper candidate in a constituency where there's little chance of winning and rely on making up that "loss" in the regional list.
      A scandal at the right time would help obviously. You'll know indyref2 is getting close when Salmond's day in court is scheduled. Although that might actually work to the SNPs advantage of he's tied to another party at th e time of a Holyrood election should it come to it.

      Delete
  11. To James Kelly
    I'm afraid it's not just you being attacked by the Wings soldiers anymore it's anybody who supports the SNP or who may have a different view or who just doesn't agree with Wings

    It's become an unedifying sight and the abuse is growing not lessening

    ReplyDelete
    Replies
    1. Quiz time for "Il Duce":

      "Readers! If the entire Solar System (for the sake of argument, defined here as everything inside the orbit of Pluto) was contained inside a pea 1cm across, how far away would the nearest neighbouring system (Alpha Centauri) be?"

      rapidly followed by:

      "Wings Over Scotland
      ‏Verified account @WingsScotland
      28m28 minutes ago

      Everyone who's replied to this so far is blocked."

      But imitation is the sincerest form of flattery.

      Delete
  12. Get back to first past the post and end the political gravy train. The Scottish system is robbing the hard working taxpayer.

    ReplyDelete
    Replies
    1. I suggested that earlier. 80% of MSPs would be SNP.

      Delete
    2. Fine with me Skier. Mps MSPs represent all their constituents. However the one million Scottish brexiteers have no representation at Holyrood.

      Delete
    3. That's their ain fault. They were free tae vote e.g. ukip.

      Delete
    4. Skier, Mps,MSPs represent all the people. UKIP are not required.

      Delete
    5. If Scotland stays in the EU, you can just identify as someone who lives in a brexited UK. Remember, a trans-noneuropean is a non-european, and don't let anyone tell you different.

      Delete
  13. Is Wings Over Scotland still on his 'I'll build a new political party' hobbyhorse? He is a good journalist. As a politician, he is still a good journalist.

    ReplyDelete
    Replies
    1. An excuse for a semi erection I would suggest.

      Delete
    2. She's pooping again. Party girl

      Delete
  14. Hello James - why not just let the Rev W Aubrey play with his trains and tables and ignore him? Even those that go to his brand website do not agree with him on this. Lets get back to pushing for independence - the easiest way to do that is to convince 50+% to vote for it. The trouble we have is that all our ndependence websites are read by independence supporters and of course GWC. Finding the way to communicate and convince the wider public is our challenge.

    ReplyDelete
    Replies
    1. All I can do is reiterate what I pointed out at the start of the thread: this post is not about the Wings party, and as it happens neither was the last one. Some people asked me to talk about a different subject, and now that I've done that they're still unhappy! "Confusion about the electoral system" is not a synonym for "the Reverend Stuart Campbell". However, I may well be posting about the Wings party again in the days, weeks or months to come - if he perserveres with the plan, there'll be plenty to say about it.

      Delete
  15. ""There is a cap on the total number of list seats any large party can win, regardless of how many list votes they take." Is this statement TRUE or FALSE?"

    TRUE. There are 8 regions with 7 list seats for a total of 56 list seats, so the cap overall is 56, and the cap per region is 7 - even if a party gets 100% of the list votes. See 2 if this is thought to be too literal.

    ""Small parties with enough votes are awarded a proportion of seats that is much larger than their proportion of the vote." Is this statement TRUE or FALSE?"

    TRUE if it relates to the LIST, but since this is a trick question fo catch the unawares who presumes list and doesn't think about overall including constituency, FALSE - but see last sentence in 1 above.

    "If it was possible to successfully game the system by voting tactically for a small pro-indy party on the list, would that party be more likely to take a LARGER or SMALLER percentage of seats than its list vote would normally justify?"

    See 1 and 2 above, the question admits of two or more answers, depending on the missing detail in the question itself. For instance, if a LARGER or SMALLER percentage of seats means on the list, then it's Larger, but if it's overall, then it could be anything basically - and the reason is that "small pro-indy party" is indeterminate, and doesn't indicate whether it stands candidates in constituences or not, nor whether it can do so successfully.

    In summary, the 3 Answers given in the blog itself are all incorrect.

    ReplyDelete
    Replies
    1. 0 out of 3, I'm afraid. The correct answers can be found in the blogpost. But thanks for playing, and better luck next time.

      Actually, you remind me of the contestant on University Challenge who was asked to identify a picture, and she said "an oak leaf". The correct answer was the National Trust logo, but she insisted she was correct, because the National Trust logo is an oak leaf. Jeremy Paxman said: "On a point of pedantry you're correct but in no other sense". I'm happy to confirm that when I said there was no cap on the number of list seats a party can win, I did not mean to imply that they could win more than 100% of the list seats available.

      But on questions 2 and 3, you're not even right on a point of pedantry.

      Delete
  16. James - I would appreciate it if you could devote one of your articles to the issue of independent candidates being elected to the Scottish Parilament. The only one who comes to mind is Margo MacDonald. It seems to me that it could enhance the quality of Scottish democracy if there were more independent MSPs. For example, the Holyrood system is set up to favour coalitions - with more 'free' members, a wider range of coalition options are possible, as well as productive collaboration around specific issues such as sectarian violence at football matches. Party-based MSPs seem to be restricted by party discipline, and a lot of the time do not seem to be allowed to say what they really believe. The list system makes it possible for list MSPs to define themselves as having broad constituencies (e.g., older people, citizens who do not have UK rights, etc). Margo MacDonald, for example, led the national debate on euthanasia. It could also be valuable to have broad non-party-based representation in Holyrood during the post-Indy phase of agreeing a constitution and setting up new institutions.

    ReplyDelete
    Replies
    1. There have been a few others - Dennis Canavan, Jean Turner and some who ended up as independents by losing or giving up the whip.

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

      Delete
    3. I used to vote for Margo. I did so knowing I was giving her my most important, 'always counted', regional list vote (obviously), appreciating that I was taking a chance my lesser/more risky constituency (SNP) vote might not count (if it was not for the winning candidate).

      Delete
    4. The problem with the current list system at Holyrood is it's ridiculously biased towards parties rather than individuals even within them.
      It really needs a form of STV so you could vote for a Margo safe in the knowledge that if they got an over abundance of votes then they could transfer on to another candidate.
      At the same time you could nominate the path your vote took even in the list so if, for example, voters really wanted to vote for a Murdo then they could do so and pick a party list.
      I'd at least keep regional lists to offset low turn outs caused by external factors in some areas skewing results. The FPTP might go but could still be incorporated.

      Delete
  17. This is what I've trying to suggest on or two different sites and @20/9/19, 11.11 on here.
    We need to bring out all the disappointed, disgruntled, marginalised Indy and potential Indy voters. We may not be able to bring them all out but if we use the "horses for courses" approach then we can cut into the vote that's liable to be left "at the side of the plate".
    I think there might also be intangible psychological factors that come into play if there's the idea of an army of people converging on independence from different directions. I also think that Soft Unionists will be more likely to reconsider and co-operate if they can see a greater Yes total support than if every independence vote is sweated and "doped" to achieve a result.

    "PictAtRandom"

    ReplyDelete
  18. Thanks James for your continued good common sense and analysis in the face of Idiocy.

    If floating the idea of a new Independence Party to game the system has any value it is only because it is so very dangerous and potentially destructive.

    Mutually Assured Destruction (MAD) is a doctrine that springs back to mind.

    This may simply be a ploy to force the SNP to action.

    For this to work it needs:

    1. A credible threat of very serious damage if SNP don’t act -check

    2. The ability of the person floating the idea to back off from the idea if it has served it’s purpose

    . ... (embarrassed silence)

    I’m really worried about this.

    Those whom the gods wish to destroy they first make MAD.

    P.S. Responding at 3am? Please look after yourself.

    ReplyDelete
    Replies
    1. Thanks, Doc. I was on an overnight bus going daan saaf (although admittedly I'm often awake at 3am anyway!).

      Delete
  19. I seen the mad mob Jocko EU fascists in George Square today doing the flag waving act. Seems the jockos are replacing the saltire with EU flags. What happened to independence! Seems we Unionists will have to salvage our saltire.

    ReplyDelete
    Replies
    1. Cordelia likes to go greenface on St. Paddy's Day. FACT.

      Delete
    2. "Seems we Unionists will have to salvage our saltire."

      So you can let England put a St. George's cross over the top of it?

      Delete
    3. Cordelia likes getting covered with Georgie. She likes to union jack.

      Delete
    4. We call her Pagoda Lil.

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

    ReplyDelete
    Replies
    1. Quite right, you replied to the wrong post. But no, I will not "just stop". You really do have an attitude problem, but you ought to have noticed by now that I don't take instructions from you.

      Delete
  21. As you like quizzes, James.

    So James, what does your own modelling of the Holyrood voting system indicate?
    Have you worked out the optimum number of constituency seats that would return the maximum regional seats and hence maximum SNP control of Holyrood?
    Because, that might just mean, perhaps, we could do all of this without standing another party, but it would require a feasibility analysis and you know, a detailed voting model, from someone who understands the Holyrood voting mechanism, the likely voting patterns across regions based on historical analysis and a knowledge of demographic groups that live there...

    I've done it. In your own time, matey boy.

    ReplyDelete
    Replies
    1. From "just stop" to "I demand you answer yet more questions about this subject, matey boy" in the space of nineteen minutes. Bizzare, but I'll go along with it, because who can resist a late night quiz about the What The F***'s It Got To Do With You, You C**t Party. I'll answer your questions in a fresh blogpost.

      Delete
  22. I merely asked a Q, to which one possible and perfectly acceptable answer was 'No, I haven't done it'. Which you just admitted to.

    I didn't demand anything. Touchy much?

    ReplyDelete
    Replies
    1. Gavin, let me just gently say to you that the "touchy" person in any conversation is probably the one who cannot cope with polite disagreement and reacts to it by calling the other person names like "c**t" and "matey boy". As you accepted, you asked questions and I answered them directly, and yet for some reason you still seem to have a problem. That's something you're going to have to work out for yourself. But perhaps start by considering this: in a discussion of this sort, what won't get you anywhere is stamping your feet and insisting that any reply to you has to buy into your basic premise (the premise in this case being that modelling based on silly assumptions tells us something interesting or important, and indeed that only modelling can tell us something interesting or important). If people bought into your premise, they wouldn't disagree with you in the first place. Deal with it.

      Delete
  23. I'm a bit concerned about where this blog is going. What's being said here about Gavin Barrie's data sounds like Michael Gove's "people have had enough of experts".

    ReplyDelete
    Replies
    1. Three scientists have commented on this blog over the last 24 hours to say that Gavin Barrie's analysis is deeply flawed and does not stand up to scrutiny. Are you going to dismiss their criticisms out of hand? If so, perhaps it's you that's had enough of experts.

      Don't worry about the direction of this blog - I'm determined it'll remain firmly grounded in reality, regardless of the flights of fancy that others may embark upon.

      Delete
  24. I'm Anynomous above. I'm sorry if I upset you, but I really would like to know more about this. (I'm a scientist myself, though not a very accomplished one.) I *think* the Wings proposal is this: if the SNP gets more constituency seats than its share of the regional votes would grant it, and Wings gets more than 5%, then the AMS procedure means that the indy seats *must* go up (if my understanding of AMS is right - it may not be!).

    If those conditions *don't* necessarily increase the indy seats, could you please give me an example with numbers - fictitious ones will do - where they don't? You see, I think Wings is right on this: there's no possible case where those conditions don't increase the indy seats, because it's in the logic of the AMS calculations.

    I agree there's a problem of knowing whether those conditions are going to be true when you start putting up election candidates, but that's what election strategists are for.

    Sorry to go on. Please don't sneer back.

    ReplyDelete
    Replies
    1. "I *think* the Wings proposal is this: if the SNP gets more constituency seats than its share of the regional votes would grant it, and Wings gets more than 5%, then the AMS procedure means that the indy seats *must* go up"

      Two points:

      1) No that's not true. To get to the point where indy seats "must" go up, you'd need significantly more than 5%.

      2) Even if you didn't need significantly more than 5%, it wouldn't be much comfort, because the whole point I've been making is that the chances of a pop-up party breaking the 5% barrier are extremely small - hence the risk of doing damage.

      I'm travelling at the moment, so not ideally placed to punch up fictitious examples. I did provide one a few weeks back where Wings was causing harm at 3%. I know you want one at a higher level than that, but if you play aroumd with the numbers yourself you'll quickly find scenarios where Wings at 6% or whatever doesn't necessarily increase pro-indy representation.

      Delete
    2. There was a mistake in my earlier post. I should have saaid, "...if the SNP's constituency seats are *a greater percentage of the total parliament* than its regional vote share would grant it...".

      On 5% or more than 5%, I can't either agree with you or disagree with you, because I don't understand how that figure is arrived at. Perhaps someone can help me with that.

      Your point that a new party won't achieve 5% is a good one, and carries a lot of weight. But it's not part of the Wings proposal. The proposal is that *if it gets more than 5%* and all the other conditions are true, then it will increase the indy seats. That's the hypothesis we need to test, and I still can't fault it. If Wings didn't expect to get 5% in a region, I guess they wouldn't stand there.

      I'll play around with numbers, as you suggest, and post again. I'd very much like to see your example of Wings losing indy seats at 3% - can you post the link here?

      Delete
    3. "If Wings didn't expect to get 5% in a region, I guess they wouldn't stand there."

      I don't see how they would have enough information to know that they're going to do better in some regions than others, unless they somehow put together a formidable canvassing operaton out of nothing (highly unlikely).

      Here's the link you asked for -

      https://scotgoespop.blogspot.com/2019/09/the-snp-must-make-this-independence.html

      Delete
    4. Thank you for posting the link. Yes, it really does look like Wings would lose indy some seats in that scenario. Maybe the Wings analysis (or Gavin Barrie, I should say - we're both scientists!) is not as accurate as they like to think. But your figures there do seem to be hard to argue with. I don't really understand the AMS procedure - I'm sure you could help me there - but there is one thing that bothers me. Your figures don't quote the constituency seats, so we don't know if your example matches the Wings initial conditions or not - maybe you could post the constituency seats here as well.

      It's a good example, and I'd like to replicate your AMS calculations (I was told that we should always do that). Can you show them to us?

      I'll look at this again in the morning - time for beddie-byes now.

      Delete