Friday, January 9, 2015

Three different ways in which betting odds and betting markets have proved to be hopelessly unreliable predictors of election results

You've probably heard the claim that "democracies never go to war with each other".  Ironically, it's often used as an excuse for warmongering, because the theory is that democracies can eventually bring lasting peace to the world by forcibly "liberating" non-democratic countries.  And note that it's an absolutist claim, or a "golden rule" - there is supposed to be literally no example in history of an armed conflict between democracies.

That's utter garbage, of course - in reality there's a list as long as your arm of conflicts between democracies.  But in each and every case, the true believers in the golden rule have a get-out clause up their sleeve.  "That wasn't really a war, was it?  And that country wasn't a proper democracy."  Sometimes the logic becomes laughably circular - for example, a war causes instability, and no country can possibly call itself democratic if it isn't stable!

I'm reminded of that kind of nonsense when I hear people profess their absolute faith in the pseudoscience of 'the betting markets as an infallible predictor of election results'.  And I use the word 'infallible' advisedly, because you'll doubtless recall Betfair's breathtaking arrogance in feeling able to declare a few days before the referendum that No had already won, on the grounds that their exchange is "never wrong".  As a marketing stunt, they even paid out early on Sportsbook referendum bets.

One of the High Priests of this bizarre religion is self-styled superstar "risk assessor" Neil Edward Lovatt.  As regular readers may know, I had to mute Lovatt on Twitter a few months ago because he was wasting far too much of my time with his incessant trolling.  I eventually blocked him altogether when he started making some deeply unpleasant personal comments, and I also muted some of his more persistent risk assessor groupies.  However, they seem to multiply with water, and although I haven't had an exchange with Lovatt for months, my Twitter notifications timeline has recently become clogged up once again with people I've never heard of demanding that I worship at the altar of the betting markets, and warning me that if I don't, it will be proof that I'm a believer in homeopathic medicine (yes, really!).

To save myself the trouble of responding individually to these inane tweets for the remainder of time, here is a quick cut-out-and-keep guide to three examples that explode the myth of the infallibility of betting markets and betting odds.  In each case, there's a very different reason why they proved to be so embarrassingly unreliable.

Exhibit A : The 2006 Liberal Democrat leadership election.  There was a straightforward clash between conventional political wisdom, which identified Menzies Campbell as the obvious successor to Charles Kennedy, and the betting odds and markets, which counter-intuitively installed the little-known Chris Huhne as the favourite.  If the markets are infallible, or even if they're more reliable than other sources of information, then the money on Huhne should have been a very clear indication that something surprising was happening beneath the surface.  But there wasn't anything happening.  As it turned out, the conventional wisdom proved correct, and Campbell won at a canter.

Reason for the inaccuracy : It seems overwhelmingly likely that rich supporters of Huhne had placed a series of large bets to earn him the coveted 'favourite' tag, and to artificially generate momentum for his campaign.  This is the problem with the belief that the betting markets somehow reflect "the wisdom of crowds" - gamblers may be motivated by something other than the quest for a value bet, and in any case it only takes a few wealthy individuals (as opposed to crowds) to tip the balance.  We heard from the bookies during the closing stages of the referendum campaign that the vast majority of bets taken in Scotland were backing Yes to win, but the odds instead reflected a relatively small number of six-figure bets that had been placed on No, mostly south of the border.

Exhibit B : The 2010 UK general election.  A month before polling day, there was a clear divide between the polls, which were pointing to a hung parliament, and the betting markets, which pointed to a Conservative majority.  Again, if the markets reflect the "wisdom of crowds", they should have been able to easily outsmart the polls, which after all are snapshots and not predictions.  But for all their imperfections, you'd have been much better off believing the polls - the Tories fell twenty seats short of a majority.

Reason for the inaccuracy : Wishful thinking from wealthy Tory-supporting punters, who slipped into the groupthink of believing that the polls must be underestimating the Tories' chances, as they did in 1992.

Exhibit C : The 2007 Holyrood election.  This is my absolute favourite.  As you'll recall, because of difficulties with the counting machines, it took until 6pm on the day after polling for the final results to be declared, and for most of the intervening period, Labour were well ahead in the running seat tally.  The betting markets followed those numbers, and suggested that Labour had a 90%+ probability of winning.  However, as early as about 10 or 11am, BBC Scotland's political editor Brian Taylor had announced that it looked from intelligence on the ground that the SNP would end up marginally ahead by the end of the day.  There were a number of televised interviews throughout the afternoon with despondent Labour MSPs, who clearly had the same expectations of the final outcome.  And yet still - astonishingly - the "infallible" markets didn't budge.

Reason for the inaccuracy : The punters piling in with the most money, firmly believing they were earning themselves free cash, were based south of the border. They were getting their "information" from the London media, and were missing what was right under their nose - it didn't even occur to them to do something as obvious as check the live blog of Scotland's leading TV political journalist.  There must have been people in Scotland who made an absolute killing that day, simply by switching on their TV.

Lesson - the "wisdom of crowds" isn't of much use if the crowds are in the wrong country.  If you wanted to predict an election result in Poland, you wouldn't ask a crowd in Denmark, would you?

41 comments:

  1. Some reasonable points here although, as Ladbrokes' political odds compiler, I have a few observations:

    1. I agree with this. A lot of political betting markets don't attract that much volume and it wouldn't take a lot of money to move the odds in whatever direction you wanted. Much less likely to happen in something like the overall result of a UK general election.

    2. I think there is a small pro-Tory bias in UK general election markets which probably skews the odds a little bit in some circumstances.

    3. I don't know enough about this particular case, but wouldn't the indyref be a valid counter-example? It looks as if non-Scots did a better job of "predicting" the result of that event; I think that may have partly been because the more energetic and visible YES campaign misled those "on the ground" into thinking that was a good representation of the state of public opinion, and to assume the polls must have been inaccurate.

    I doubt anyone is actually arguing that the political betting markets are"infallible" in the sense that nobody should expect that the favourite would always win. If the markets say there is a 90% chance of a particular outcome then it should fail to happen 10% of the time. When it does fail to happen, that doesn't mean the market was "wrong".


    ReplyDelete
    Replies
    1. Non-Scots didn't get it right since Yes was winning without THE VOW. 55-45 according to BT's own polling. And yet the massive english bets sponsored some may say by HM Government were used as propaganda against the Yes vote.

      So stick it up your arse racist. PB Herd member. All round Twat and gormless nork.


      Delete
    2. "Non-Scots didn't get it right since Yes was winning without THE VOW. 55-45 according to BT's own polling."

      Really? I must have missed that one. Do you have a reliable source? (I really don't know why I'm asking)

      Delete
    3. 55-45 isn't quite correct, but it's fairly well-documented there was a pre-Vow internal poll for Better Together that had Yes ahead by 53% to 47%.

      Delete
    4. I've genuinely never heard of that. The only internal polls I've really heard of were those large British government commissioned Ipsos Mori ones, and the internet rumour mill certainly went into overdrive over that. Having searched on Google, I've not found anything of it. I'm not saying that such a poll definitely didn't exist, but given the Chinese whispers effect you always get, you need to be sceptical.

      Delete
    5. Several high-profile journalists have reported it, including Laura Kuenssberg. I'd be very surprised if it wasn't true.

      Delete
    6. It is utterly inconeivable that the No campaign didn't have private polling (and likely more than one source) at that time which spooked them badly. Their actions with the VOW speak for themselves. The Yes campaign certainly had private polling and we know from the AV referendum (which had some of the exact same team working for No) that private polling was commonplace and indeed one of things which resulted in their change of strategy over AV.


      For those who might forget there was a time when Yes to AV was actually doing surprisingly well in the polls which caused a pretty vocal backlash among some very upset tory backbenchers who made some extremely ominous noises about what might happen to Cameron should AV win. That resulted in a change in strategy backed up by the private polling in which Cameron took a far more vocal and central role in that campaign as well as making Clegg the face of Yes to AV. Something which caused the first real fracture in the tory lib dem coalition with calamity Clegg upset over what he thought he had as a 'gentleman's agreement' with Cameron that neither of them them would be on the front line for AV. Not all of Clegg the the lib dem's 'differentiation' was mere posturing and you could probably trace back the start of the death of Lords reform and the failed boundary changes to that time. But again, that was prompted by not just public but private polling as it is utterly commonplace for political parties to use it not just for elections but referenda. They can make massive decisions based on it so it is their 'fail safe' against volatile public poling if you like. Course if they ask the wrong questions in that private polling then it is worse than useless and actively counterproductive. However, they are the clients rather than newspapers or other sources which is why they often put so much stock in them. If you have to pay for it and you set the questions and parameters then you naturally tend to give it more credence than other polling.

      Delete
    7. "Several high-profile journalists have reported it, including Laura Kuenssberg. I'd be very surprised if it wasn't true."

      Looks like you're right. Interesting...

      Delete
    8. Shadsy
      It wouldn't surprise me if for elections, people were puttng bets on to pay for the celebrations if their "team" won. But the referendum was a very different affair and it was obvious to me from even before the campaigns started that for such a momentous affair in Scotland's history, people would want to be able to say "I was there, I voted X" with X being the winner.

      From that point of view the odds on a YES / NO could be very influential in themselves (in theory) to the way people would vote on the day. They would want to be on the winning side. So if it was 1/7 on NO, then NO would appear to be the almost certain winner. 1/2? Well, that's more open.

      In that scenario, an early bet of £500,000 on NO would set the odds on a NO win very much odds-on and hey, that's what happened. That of course negates a lot of small YES bets. It means, basically, that anyone willing to spend a large sum of money to try to influence the result, would have a high chance.

      Add to that a research project in August from I believe Edinburgh Uni "showing" that betting odds were a far better predictor of results than opinion polls, seized on by the media and given a lot of publicity, and, job's a good 'un.

      Delete
    9. On Betfair alone the figure of money supposedly placed on the referendum was almost £9 million. Given the amount of money you'd have to invest to shift the overall odds it would be utterly pointless to even bother going down that road when you could spend the same money on actual campaign relevant activities (staff, research, billboards, etc.).

      The latter has a direct and proven impact on voting, the former constitutes a ridiculous investment of money simply to shift something that there's no evidence (as far as I'm aware) has any impact on voters. If there was a conspiracy by the No campaign to manipulate betting markets then it was an utterly woeful use of resources.

      Delete
    10. "Given the amount of money you'd have to invest to shift the overall odds..."

      And yet we know that bets of the required scale (six figures) were indeed placed. Given that elsewhere on this thread you demand absolute proof from me, as opposed to just a theory, I look forward to you meeting the same standard. What hard proof do you have that those large bets were not intended to shift the odds?

      Delete
    11. "What hard proof do you have that those large bets were not intended to shift the odds?"

      The old "prove my conspiracy theory isn't true" line of argument - can you prove the Queen isn't a shape shifting lizard?

      Delete
    12. In which case I eagerly await your apology for fatuously demanding similar proof from me elsewhere on this thread.

      I believe the word in your part of the world is "gotcha".

      Delete
  2. WeSaidNoToYesMen :-)January 9, 2015 at 11:12 PM

    Did they get it as wrong as the SNP's hopelessly unreliable predictors of oil prices? B-)

    ReplyDelete
  3. Edna Welthorpe (Mrs)January 9, 2015 at 11:27 PM

    Nope. Oil Industry forecasts. How is George going to make up the shortfall. For Scotland it is a bonus for Westminster it is a necessity.

    ReplyDelete
    Replies
    1. "For Scotland it is a bonus for Westminster it is a necessity."

      It's frightening how often you hear people say this. Without North Sea revenue we generated roughly 8.2% of UK taxation revenue in the last GERS report. We received 9.3% of UK public spending. If you're telling us that oil is a "bonus" for the Scottish economy (and not a fundamental part of being able to maintain our current levels of spending) then you simply haven't read the figures.

      This is basic stuff you can read just by going on the Scottish government's website - yet supposedly it's those of us on the No side who are "gullible"...

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

      Delete
    3. @GrimshardJanuary 10, 2015 at 10:12 PM

      Do you understand that cherry picking only one side without doing the sums on the other gives one a very misrepresented scenario.

      Where does Scotland’s wealth go?

      Delete
    4. Edna Welthorpe (Mrs)January 10, 2015 at 10:35 PM

      Have a look here for more info for dodgy UK stats;

      http://wingsoverscotland.com/a-few-not-very-good-men/

      Delete
    5. If you can explain to me why it's cherry picking to quote the amount we generate and the amount we spend (i.e. our basic fiscal position) then go ahead. Those are the Scottish government's figures incidentally.

      The link you've posted isn't rebutting anything I've said.

      Delete
    6. @Edna "Have a look here for more info for dodgy UK stats"

      I haven't quoted any UK stats. I've quoted the Scottish government's stats which show quite clearly, with no spin whatsoever, that without oil we'd be generating far less in revenue than we'd need to maintain spending at our current level if we were independent. How in that case can you regard oil as a "bonus" for the Scottish economy?

      I'm not asking for what Business for Scotland, Wings Over Scotland, or anyone else has to say on the subject, I'm speaking to you personally. You made the claim and I've quoted the direct figures from the Scottish government (here: http://www.scotland.gov.uk/Resource/0044/00446179.pdf) showing that you're wrong. The fact that we're even arguing about this, when it's our basic fiscal position, is quite frightening.

      Delete
    7. @GrimshardJanuary

      Where is the comparative UK stats? GERS was a stitch up by Ian Lang just look at Hansard. Who audits the UK numbers other than themselves?

      Back in 1922-3 Scotland sent c£116,000 to the treasury and got back a measly c£35 grand as they have always seen Scotland as a cash cow.

      Can an Independent Scotland stand on its own two feet?

      Delete

    8. 'Without North Sea revenue we generated roughly 8.2% of UK taxation revenue in the last GERS report. We received 9.3% of UK public spending.'

      But even with a falling oil price we won't have no revenue we will have some i.e. enough to make up to 9.3% we need. Also if we are running our own economy we can organise to be less oil dependent, create an oil fund and so on.

      Delete
    9. It's hilarious how often the same witless PB troll comes on here get's comprehensively rebutted and then shrieks petulanty that he wasn't interested in the sources that disprove his sub-Daily Mail style assumptions and amusing cherry picking of quotes or figures.

      He usually then goes on to hold up as actual proof the fact that people disagree with him (or as he spins it, arguing) as some sort of vindication for his trolling tells you all you need to know about the vapid nature of his utterly clueless and out of touch westminster bubble thinking. :-)

      Delete
    10. @Denise "But even with a falling oil price we won't have no revenue we will have some i.e. enough to make up to 9.3% we need. Also if we are running our own economy we can organise to be less oil dependent, create an oil fund and so on."

      Of course we won't have no oil. We're talking about the argument that oil is a "bonus". If you want to have a general discussion about whether independence is or isn't a good thing then that's completely fine, but it's a different discussion. I'm making a very limited point here that if you took oil away from the Scottish economy we'd be in a much worse position that we are now. That's not a controversial point, I'd suggest it would be better to just accept it rather than arguing about it for no particular reason.

      Delete
    11. @Mick

      I have no idea what on earth any of that is supposed to mean, but if you disagree with anything I've said above feel free to try posting a response that explains precisely why you disagree with it. You seem to be implying what I've written is "utterly clueless", "out of touch" and "trolling" when I've simply posted the Scottish government's own statistics to show that oil isn't a "bonus" to the Scottish economy.

      Do you actually dispute that?

      Delete
  4. "It seems overwhelmingly likely that rich supporters of Huhne had placed a series of large bets to earn him the coveted 'favourite' tag, and to artificially generate momentum for his campaign."

    And your evidence for this is what? This sort of thing is said constantly about betting markets "such and such has probably just ploughed money into it to make it seem like they're the favourite" yet there's almost never any evidence offered to back it up. It was said constantly during the referendum, yet the betting markets gave a far more accurate reflection of the state of play than the polling (which all exaggerated the Yes vote).

    Anyone who genuinely thinks betting markets are infallible is an idiot - if they were infallible then making a fortune would be as simple as backing the favourite any time an election was held. However proving they aren't infallible isn't the same thing as proving they're useless.

    ReplyDelete
    Replies
    1. "yet there's almost never any evidence offered to back it up. It was said constantly during the referendum, yet the betting markets gave a far more accurate reflection of the state of play than the polling (which all exaggerated the Yes vote)."

      Wonderfully oblivious without even a hint of irony. Superb comedy gold! :-D

      Delete
    2. "And your evidence for this is what?"

      Given that demand, I'll look forward to you providing proof for the speculative claim that you made earlier in this thread.

      By definition, "overwhelmingly likely" does not imply the existence of irrefutable proof. However, if you can offer an alternative explanation for the least well-known of the three candidates (in fact by some distance the least well-known) becoming the bookies' favourite out of nowhere, in spite of the fact that he plainly wasn't piling up the votes in the real world, I'd be interested to hear it.

      "yet the betting markets gave a far more accurate reflection of the state of play than the polling (which all exaggerated the Yes vote)."

      WHAT?! Until just a few weeks before polling day, the three No-friendly pollsters had never shown a higher Yes vote than 42%. Ipsos-Mori had never shown a higher Yes vote than 40%. The actual Yes vote, let me remind you, was 45%.

      As you're so keen on proof, let's hear your proof that the betting markets were more accurate than the polls, in spite of the fact that we know large-scale internal polling for the No side had Yes ahead in the penultimate week - at a time when the markets were still showing an overwhelming probability of a No win.

      Delete
    3. "By definition, "overwhelmingly likely" does not imply the existence of irrefutable proof. However, if you can offer an alternative explanation for the least well-known of the three candidates (in fact by some distance the least well-known) becoming the bookies' favourite out of nowhere, in spite of the fact that he plainly wasn't piling up the votes in the real world, I'd be interested to hear it."

      Pretty much your entire narrative of that leadership contest is misleading. You've implied that Huhne was installed as the favourite from the beginning, when in reality Campbell was actually the favourite at the outset. Huhne only cemented his place as favourite after several polls of Lib Dem members put him ahead (something you haven't even mentioned if you were aware of it). You've presented it as if there was literally no reason for anyone to think Huhne could have won, when in reality he was regularly topping members polls.

      -----

      "WHAT?! Until just a few weeks before polling day, the three No-friendly pollsters had never shown a higher Yes vote than 42%. Ipsos-Mori had never shown a higher Yes vote than 40%. The actual Yes vote, let me remind you, was 45%."

      That's a genuinely bizarre comment. You seem to be suggesting we should judge the polls not on what they produced right before the referendum (almost every company giving Yes a higher vote share than 45%) and instead judge them on what they were saying weeks beforehand.

      --------

      "As you're so keen on proof, let's hear your proof that the betting markets were more accurate than the polls, in spite of the fact that we know large-scale internal polling for the No side had Yes ahead in the penultimate week - at a time when the markets were still showing an overwhelming probability of a No win."

      That doesn't make the slightest bit of sense. You're claiming that the fact internal No polling showed Yes ahead (which it wasn't - you do remember the result of the referendum right?) is proof that the betting markets (which said No was ahead the entire time) got it wrong. So the betting markets get it right, the polling (even internal No polling at that) gets it wrong, and your conclusion is that the polling is more accurate than the betting markets?

      I'm going to be charitable and assume you didn't actually mean to type this because it makes no sense whatsoever.

      Delete
    4. Grimshard : My first and most important piece of advice to you is stop acting like a cretin. I'll try and make my way through your train-wreck of a response, but I think it would be better for all concerned (most especially yourself) if you just pack it in now.

      First of all, let's see some links to the Liberal Democrat leadership polls that you're praying in aid. You do of course have to establish the truth of your claim. My recollection of those polls is that they did not in fact show a clear (or at least not a consistent) lead for Huhne, and that there were severe doubts over their reliability. Certainly they didn't carry sufficient weight to negate the common sense assumption that the much better-known Campbell was the more likely winner, and if the markets are sensitive to information and wisdom that is not available elsewhere (which is the claim that is often made) then there is no way Huhne should have been the favourite - as indeed he was for a prolonged spell, in spite of your protestations. (If you want to go back to the beginning of the contest, Simon Hughes was an early favourite - how did that work out?)

      "That's a genuinely bizarre comment."

      Sneer as much as you like, but I am indeed suggesting that it's entirely reasonable to judge the polls on their accuracy at the time they were conducted. I'm quite sure that Peter Kellner would be just as keen as you are to erase from history the YouGov poll just a few weeks before referendum day putting Yes on 39% (at a time when other pollsters had Yes in the mid-to-high 40s), but I'm afraid it's there as a matter of record. Are you SERIOUSLY claiming that Yes were being "overestimated" at 39%?

      "You're claiming that the fact internal No polling showed Yes ahead (which it wasn't - you do remember the result of the referendum right?)"

      That must be the stupidest of all the incredibly stupid things you've said this evening. The Yes vote was 45% on the 18th of September, therefore it must also have been 45% on the 11th of September? Yeah, sure, mate - that makes perfect sense. Nobody ever changes their mind about anything. It's the law.

      "I'm going to be charitable"

      No need to be charitable - just be kind to yourself and put a sock in it.

      Delete
    5. "First of all, let's see some links to the Liberal Democrat leadership polls"

      6/2/2006 (Times): Campbell (24.5%), Huhne (30.5%), Hughes (19%)

      7/2/2006-9/2/2006 (Yougov): Campbell (34%), Huhne (38%), Hughes (27%)

      23/2/2006 (Guardian): Campbell (29%), Huhne (36%), Hughes (21%)

      Wikipedia link for ease of reference: https://en.wikipedia.org/wiki/Liberal_Democrats_leadership_election,_2006#Opinion_polls_.28leadership_candidates.29

      ---------

      "My recollection of those polls is that they did not in fact show a clear (or at least not a consistent) lead for Huhne, and that there were severe doubts over their reliability. Certainly they didn't carry sufficient weight to negate the common sense assumption that the much better-known Campbell was the more likely winner"

      This is a completely different argument from what you'd originally written. You didn't even acknowledge the existence of any polls in favour of Huhne, now your argument is that the polls were flawed and nobody should have paid any attention to them. That's dubious on at least two grounds. First, the fact that you think people shouldn't have listened to the polling is irrelevant to an argument about whether they did actually listen to it. We're not arguing over whether the betting markets were right to make Huhne favourite (he didn't win, obviously they got it wrong) we're arguing about your claim that the support for Huhne was so utterly unreasonable that the "overwhelmingly likely" explanation is the betting markets were manipulated by his supporters.

      Second, members polls are always flawed, but the YouGov poll in particular was given great weight at the time because a similar poll had called Cameron's win in the Tory leadership contest fairly accurately. Again, whether you think they should have listened to the polling is neither here nor there. Nobody here is saying that betting markets are infallible, but the picture you originally gave is of a completely irrational betting market that put Huhne as the favourite against all evidence to the contrary. That simply isn't true.

      ---------

      "Sneer as much as you like, but I am indeed suggesting that it's entirely reasonable to judge the polls on their accuracy at the time they were conducted."

      If I'm "sneering" it's because what you're writing here is completely illogical. I said that the polling exaggerated the Yes vote. It's blindingly obvious that what I'm referring to is the fact that days before the referendum the polling was consistently putting Yes above 45%. You then proceeded to write "WHAT?" in capital letters and argue that they didn't overestimate the support because weeks earlier (when we have no earthly idea what the actual level of support was given opinions change over time) it was putting Yes below 45%.

      I'd suggest it might be more reasonable to simply acknowledge that the final polling did in fact overestimate the Yes support rather than arguing about that limited statement for no particular reason.

      ------

      "The Yes vote was 45% on the 18th of September, therefore it must also have been 45% on the 11th of September?"

      At no point, anywhere, did I say anything about what the Yes vote actually was on 11 September - nobody knows what the support was. What you're doing is attempting to argue that the betting markets were wrong on 11 September simply because they didn't match what the polling said on 11 September. You're then bizarrely trying to use that as evidence that polling is more accurate than the betting markets.

      So in other words your proof that polling is more accurate is that the results of the polling on 11 September more accurately matched the results of the polling on 11 September than the betting markets did. Can you honestly not see why that's a completely circular/absurd argument?

      Delete
    6. "This is a completely different argument from what you'd originally written. You didn't even acknowledge the existence of any polls in favour of Huhne, now your argument is that the polls were flawed and nobody should have paid any attention to them."

      You know how I said it would be a good idea if you stopped acting like a cretin? Well, this would be an excellent place to start. You're correctly noting that I didn't mention your polls at all - the reason for that is because they're not relevant to my argument. You'll recall, I trust, that I said they "didn't carry sufficient weight to negate the common sense assumption that the much better-known Campbell was the more likely winner". What you're in fact whingeing about is that I've remained entirely consistent in ignoring unreliable polls that weren't taken seriously at the time (and in spite of their bias towards more committed activists, they didn't show a significant lead for Huhne, as I correctly recalled).

      "I said that the polling exaggerated the Yes vote. It's blindingly obvious that what I'm referring to is the fact that days before the referendum the polling was consistently putting Yes above 45%."

      Er, no. It's not blindingly obvious. What is blindingly obvious, of course, is your attempt to backtrack because you've been caught saying something that is at complete variance with the facts.

      "At no point, anywhere, did I say anything about what the Yes vote actually was on 11 September"

      Yes, you did. I helpfully quoted you doing precisely that (not on that exact date, if you want to quibble, but that's of no consequence).

      Please go away now. You've had your say, and I've got better things to do with my time than endlessly correct the lies, distortions and half-truths in your bile.

      Delete
    7. "You're correctly noting that I didn't mention your polls at all - the reason for that is because they're not relevant to my argument. You'll recall, I trust, that I said they "didn't carry sufficient weight to negate the common sense assumption that the much better-known Campbell was the more likely winner"."

      That's brilliant. You've said they didn't carry sufficient weight so that's it. Case closed. I mean who in their right mind would think that professional polling of Lib Dem members might have some impact on who people thought would win a leadership contest voted on by Lib Dem members?

      Your argument, to remind you, is that it was "overwhelmingly likely" Huhne's supporters were manipulating the betting markets. I've pointed out that Huhne was actually ahead in most of the major polling of Lib Dem members and your reaction to that appears to be "that doesn't matter, I say so, go away and stop being a cretin".

      Honestly, I'm trying hard to be respectful to you here, but as an argument that's about as weak as it gets.

      --------------

      "Er, no. It's not blindingly obvious. What is blindingly obvious, of course, is your attempt to backtrack because you've been caught saying something that is at complete variance with the facts."

      Yes, clearly when someone says the polling overstated the Yes vote what they really mean isn't the polling immediately before the vote (that would be too rational and straightforward) but polling arbitrarily selected from several weeks previously (which may well have been accurate at the time for all you or anyone else knows).

      Let's keep things very simple. Do you or do you not accept that the final polling overstated the size of the Yes vote? A yes or a no will suffice.

      ----------

      "Yes, you did. I helpfully quoted you doing precisely that (not on that exact date, if you want to quibble, but that's of no consequence)."

      Ironically you're the one doing the exact thing you're accusing me of - namely taking an opinion poll as gospel and trying to argue that the mere fact the betting markets disagreed with the polling means the betting markets were wrong (I'll take it as a given that the fact you're no longer trying to defend this means you've realised the absurd nature of that argument).

      -----------
      "Please go away now. You've had your say, and I've got better things to do with my time than endlessly correct the lies, distortions and half-truths in your bile."

      You put your arguments out to a public audience, if you can't defend them properly then that's your problem.

      Delete
    8. Actually, I'm the owner of this blog, so I do have a fairly straightforward recourse when someone like you abuses their posting rights. You seemingly don't have the self-awareness to realise that I'm the one that's allowed you to spew your ill-mannered guff here - in contrast to virtually every unionist blogger I can think of, who would have simply deleted your comments in identical circumstances.

      I'm on a bus at the moment, but just for fun, I'll have one more detailed wade through your tripe when I get home. In tribute to your new number one fan Neil "Alligators" Lovatt (you've "owned" me, apparently!), I might even conclude by screaming "Eet vos a pleasure to shhhhhred eet!!!!!!", followed by prolonged maniacal cackling.

      Delete
    9. OK, a bit later than expected, here we go -

      "I mean who in their right mind would think that professional polling of Lib Dem members might have some impact on who people thought would win a leadership contest voted on by Lib Dem members? "

      Certainly nobody who understands that credible polling of the members of a party as small as the Lib Dems simply isn't possible. I know you don't want to hear this, but the majority of serious commentators did grasp that point at the time. A firm like YouGov simply didn't have enough Lib Dem members on their books, and the ones they did have were disproportionately likely to be highly-engaged, and therefore Huhne supporters. How do you weight a sample of Lib Dem members correctly? You can't. Taking all those factors together, it was fully understood that a tiny Huhne lead was pretty meaningless. The fact that you apparently failed to understand that is your own affair.

      "Honestly, I'm trying hard to be respectful to you here, but as an argument that's about as weak as it gets."

      I'm afraid I can't follow your lead by "honestly" pretending to try to be "respectful" - your behaviour here has been a disgrace, and warrants no respect whatever.

      "Yes, clearly when someone says the polling overstated the Yes vote what they really mean isn't the polling immediately before the vote"

      Clearly what they mean is the polling. If in fact you only meant a very tiny percentage of the polling (the bit that was published on the final day), here's a gentle suggestion - perhaps you should have specified that.

      "Let's keep things very simple. Do you or do you not accept that the final polling overstated the size of the Yes vote? A yes or a no will suffice."

      Yes. But as I've already spontaneously volunteered that observation on this blog about seventeen billion times before, I'm not really seeing how this represents much of a triumph for you. It's also not relevant to the point you originally raised, which was about the polling in general, rather than just a tiny percentage of the polls (all conveniently after the late convergence between Yes-friendly and No-friendly firms).

      "Ironically you're the one doing the exact thing you're accusing me of - namely taking an opinion poll as gospel and trying to argue that the mere fact the betting markets disagreed with the polling means the betting markets were wrong (I'll take it as a given that the fact you're no longer trying to defend this means you've realised the absurd nature of that argument)."

      Is that paragraph even supposed to make sense? Maybe it's too early in the morning.

      Now, I'm guessing your response is likely to be something along the lines of -

      Oh, that's brilliant. Ah-ha-ha. Oh that's a classic. Tee-hee-hee. That's priceless. Oh-ho-ho. Ah-ha-ha-ha. Hee-hee-hee-ho-ho-ho-ho-ho.

      After all, who needs a defensible argument when they've got a Laughing Policeman impersonation?

      Delete
  5. That is totally not true Mick, the polling was spookily accurate. It picked up the Yes surge and the swingback almost perfectly with the final YouGov 54-46 just missing the final bit of swing back.

    ReplyDelete
    Replies
    1. Denise, there's no way that all of the polling can have been accurate. The No-friendly pollsters showed a huge swing over the closing weeks, while the Yes-friendly pollsters showed almost no swing at all. There are two possibilities - either one camp was broadly right and the other was broadly wrong, or else the truth was somewhere in between the two extremes (in which case all of the pollsters were wrong to some extent).

      Delete
  6. The final polls from 16th Sept: 52-48, then on the 17/9 it moved to 53-47, then 18/9 final YouGov 54-46. The swing to No was very clear. Apart from the ICM\Sunday Telegraph with low sample size which they immediately said was an outlier and not to be relied on, it is clear that Yes was behind for the final 10 days and in the final few days the waivers went to No.

    Polls are a snapshot so they would not 10 days out predict the result but they accurately predicted the state of play 10 days out. That is why there was the panic in the Better Together ranks because their private polling was telling them Yes was level or ahead and momentum at the point was with Yes.

    IMO YouGov probably had it right through out the campaign from their early 60-40 splits to the 51-49 poll and to their final 54-46 poll.

    Panelbase (3) 17/09/14 45 50 n/a 5 -5
    Ipsos MORI/STV (1) 16/08/14 47 49 n/a 5 -2
    Survation/Daily Mail 16/09/14 44 48 n/a 8 -4
    Opinium/Telegraph 15/09/14 43 47 1 8 -4
    Survation/Better Together (T) 12/09/14 42 49 n/a 9 -7
    Panelbase/Sunday Times (3) 12/09/14 46 47 n/a 7 -1
    ICM/Sunday Telegraph (O) 11/09/14 48 42 n/a 10 +6
    ICM/Guardian 11/09/14 40 42 n/a 17 -2
    Opinium/Observer 11/09/14 43 47 2 8 -4
    YouGov/Sun/Times (3) 11/09/14 45 50 2 4 -5
    Survation/Daily Record 09/09/14 42 48 n/a 10 -6
    YouGov/Sunday Times (3) 05/09/14 47 45 1 6 +2

    ReplyDelete
    Replies
    1. "The swing to No was very clear."

      It was clear with YouGov. There just isn't enough information from the other pollsters to be able to say that more generally, although it seems likely there was a late swing to No (differential turnout in favour of No probably also played a part).

      I personally think it's extremely unlikely that YouGov had it right all along (the Kellner Correction probably magnified the swing), but it's possible. It's also possible that Panelbase and Survation were right all along, in which case the swing to Yes in the closing months was minimal. We'll just never know for sure.

      Delete