Howdy, Stranger!

It looks like you're new here. Sign in or register to get started.

politicalbetting.com » Blog Archive » And before the big Question Time event here’s Marf

14567810»

Comments

  • Chris123Chris123 Posts: 174
    kle4 said:

    Danny565 said:

    Populus will be the test tomorrow. Even as stable as they usually are, surely there'll be a Tory lead or atleast a tie if there genuinely has been movement this week.

    YG static & dead as a floorboard. Can we all agree to bin YG? Regardless of opinions no polling 7 days a week should have this little movement. Somethings badly wrong there.
    Well, we feel like there should have been more movement, but if they end up calling the election better than anyone else, it will have been our expectations that were wrong, not them. I don't think anyone should be certain YouGov must be wrong.
    MikeK said:

    Sad story of the night.

    MissLeahMarie ‏@LittleMissUKIP 4h4 hours ago
    The BBC are disgusting. Decided to tell everyone WHEN WE GOT THERE that only some of us are allowed in for "balance" reasons. BBC bastards.

    I don't see how that is particularly sad. A little inconvenient, but parties and their supporters whinge incessantly about balance, surely worth an individual's inconvenience.
    I find YouGov to be the most credible. The underlying fact is that the race has been incredibly static. I don't think the dial has moved very much. And pollsters that are all over the place from week to week are just showing us that their overall methodology isn't as tight as that of YouGov.
  • oldpoliticsoldpolitics Posts: 455
    Tabman said:

    Tabman said:

    currystar said:

    Pulpstar said:

    currystar said:

    IOS said:

    Anyway - how many people have voted already - at least 15% of the people that are going to vote I reckon.

    Not sure there are many people to shift in the center - just about getting out your core.

    Step forward the ground game.

    As an Eastleigh resident i am stunned by the lack of the feeling of an election in this area. I always felt this was the lib dems safest seat, yet I have not received a single piece of election literature from them and neither have my work colleagues who also live in Eastleigh . It's very odd . The Tories seem to be the only ones who are making an effort. I'm very confused.
    Amazing. I'm seriously beginning to wonder if Ashcroft Question ONE might be the correct one.
    Eastleigh was a 3 way split at the by election. Since then then the LD vote has slipped further and I cannot see UKIP doing any where like the same. I think the Cons are value.
    Normally a week before an election Eastleigh turns orange with the number of lib dem banners and posters. Not this time.
    Surprised at that. LD ground troops there are rabidly busy normally. Keith House keeps them at it.
    OXWAB is a sea of Orange. A fair dash of blue, a little green. Labour is not obvious but helped by several estate agents with red and yellow.
    I happened to be in Jericho last weekend. There were more yellow diamonds than I expected, but on the other hand there were more Labour posters per property than I can remember seeing anywhere in the last decade and a half. Admittedly the urban Oxford bit of OXWAB is not the whole of it, but I'm pretty bullish on a Tory hold here given the amount of tactical LD support that appears to be giving up and going back to Labour, or the principled LD support that's now Green.
    I don't doubt it. But having never been in a tight contest before I find it interesting. I've not been off the main drag in Jericho. Have you been down Botley Road and up Cumnor Hill?
    No, just Jericho (Red with bits of yellow), Wolvercote (Green with bits of yellow) and some surprisingly Blue areas around the Woodstock Road. Well, surprising for Oxford, those kind of houses would be solid Tory anywhere else in the country. Mansion Tax issues?
  • RobDRobD Posts: 60,045
    916 comments? a taste of things to come on election night :D
  • PulpstarPulpstar Posts: 78,417
    RobD said:

    916 comments? a taste of things to come on election night :D

    :D I'll do my best.
  • RodCrosbyRodCrosby Posts: 7,737

    RodCrosby said:

    NoEasyDay said:

    Scott_P said:

    @PeteWishart: Important front page in the National tomorrow. This is only going to get bigger. All over for Lab, methinks. https://t.co/t3upVzfxGV

    Two things
    1. Milliband must be very worried about how the SNP issue is playing on the doorstep to keep going on about this.
    2. How does he explain getting a queens speech through, with what must be a deal with the SNP.
    Miliband was very clear about this tonight. Suppose that Cameron resigns, or loses a vote in the House, and so Miliband has a shot at being PM. What Miliband has said he will do is to put forward a Labour Queen Speech, without negotiating with the SNP to secure their support.
    Citation?
    This is what Miliband said tonight:

    "We do want to put forward a Labour Queen Speech, I want a majority Labour government, and it would be for parties in the House of Commons to vote for it."
    OK, thanks, I didn't hear that, and thought I paid close attention.

    But still, a bit muddled. Why would you want others to vote for it, if you already have a majority? You may want to put forward a QS, but in certain circumstances, discretion could be the better part of valour?

    I do detect a slight equivocation from "We will vote down a Tory Queen's Speech in a hung Parliament."
  • Sunil_PrasannanSunil_Prasannan Posts: 52,150

    Monthly "Super-ELBOW" for April - hope to be published by this time tomorrow (assuming tonight's YG is the last poll with field-work during April).

    Five different measures this time:

    The official ELBOW inc. every single poll with field-work end-dates between 1st and 30th April
    The YouGov only figures
    The Non-YouGov figures

    And, by popular demand, the figures for the online polls and the phone polls!

  • IOSIOS Posts: 1,450
    Rod

    nah. Miliband is flat out gonna shaft Cameron with the SNP. Basically unless Clegg is gone from the Libs he will sit as PM propose a bill and dare other parties to vote it down.

  • PulpstarPulpstar Posts: 78,417
    RodCrosby said:

    RodCrosby said:

    NoEasyDay said:

    Scott_P said:

    @PeteWishart: Important front page in the National tomorrow. This is only going to get bigger. All over for Lab, methinks. https://t.co/t3upVzfxGV

    Two things
    1. Milliband must be very worried about how the SNP issue is playing on the doorstep to keep going on about this.
    2. How does he explain getting a queens speech through, with what must be a deal with the SNP.
    Miliband was very clear about this tonight. Suppose that Cameron resigns, or loses a vote in the House, and so Miliband has a shot at being PM. What Miliband has said he will do is to put forward a Labour Queen Speech, without negotiating with the SNP to secure their support.
    Citation?
    This is what Miliband said tonight:

    "We do want to put forward a Labour Queen Speech, I want a majority Labour government, and it would be for parties in the House of Commons to vote for it."
    OK, thanks, I didn't hear that, and thought I paid close attention.

    But still, a bit muddled. Why would you want others to vote for it, if you already have a majority? You may want to put forward a QS, but in certain circumstances, discretion could be the better part of valour?

    I do detect a slight equivocation from "We will vote down a Tory Queen's Speech in a hung Parliament."
    That's what I heard too, Con minority not bad at 5.5 on Betfair given I doubt the yellow peril will have an appetite for Gov't with the shellacking they're going to get.

    I think Ed ends up as PM though is most likely now.
  • Y0kelY0kel Posts: 2,307
    Most great fights don't even reach the public consciousness, not even to many people who would be boxing fans. Last years Gamboa vs Crawford fight was as brilliant as some of the four kings' fights against each other for example.

    Katsidis/Armonsot, one the best dust ups in a decade, rivals Hearns Hagler except the former pair did it for 12 rounds of chaos not 3.

    Neither of those examples were some punch-ups in a barn, they were top-end boxers of their time up against each other.
  • Philip_ThompsonPhilip_Thompson Posts: 65,826
    Chris123 said:

    kle4 said:

    Danny565 said:

    Populus will be the test tomorrow. Even as stable as they usually are, surely there'll be a Tory lead or atleast a tie if there genuinely has been movement this week.

    YG static & dead as a floorboard. Can we all agree to bin YG? Regardless of opinions no polling 7 days a week should have this little movement. Somethings badly wrong there.
    Well, we feel like there should have been more movement, but if they end up calling the election better than anyone else, it will have been our expectations that were wrong, not them. I don't think anyone should be certain YouGov must be wrong.
    MikeK said:

    Sad story of the night.

    MissLeahMarie ‏@LittleMissUKIP 4h4 hours ago
    The BBC are disgusting. Decided to tell everyone WHEN WE GOT THERE that only some of us are allowed in for "balance" reasons. BBC bastards.

    I don't see how that is particularly sad. A little inconvenient, but parties and their supporters whinge incessantly about balance, surely worth an individual's inconvenience.
    I find YouGov to be the most credible. The underlying fact is that the race has been incredibly static. I don't think the dial has moved very much. And pollsters that are all over the place from week to week are just showing us that their overall methodology isn't as tight as that of YouGov.
    Random sampling should have random errors. Outliers aren't a flaw of random polls there part and parcel of how it works. In a healthy sample of polls a certain proportion should be outside the margins of error. At a 5% confidence interval one in twenty will be.

    YouGov aren't acting randomly. That means something is wrong. YouGov clearly aren't randomly selecting a sample of the UK under standard distributions.
  • PongPong Posts: 4,693
    http://www.mirror.co.uk/news/uk-news/royal-baby-news-live-updates-5606528

    What should the royal baby be called?

    Ed Miliband - 23%
    Charlotte - 21%
    Alice - 17%
    Alexandra - 16%
    James - 10%
    Arthur - 8%
    Richard - 5%
  • NemtynakhtNemtynakht Posts: 2,329
    Chris123 said:

    kle4 said:

    Danny565 said:

    Populus will be the test tomorrow. Even as stable as they usually are, surely there'll be a Tory lead or atleast a tie if there genuinely has been movement this week.

    YG static & dead as a floorboard. Can we all agree to bin YG? Regardless of opinions no polling 7 days a week should have this little movement. Somethings badly wrong there.
    Well, we feel like there should have been more movement, but if they end up calling the election better than anyone else, it will have been our expectations that were wrong, not them. I don't think anyone should be certain YouGov must be wrong.
    MikeK said:

    Sad story of the night.

    MissLeahMarie ‏@LittleMissUKIP 4h4 hours ago
    The BBC are disgusting. Decided to tell everyone WHEN WE GOT THERE that only some of us are allowed in for "balance" reasons. BBC bastards.

    I don't see how that is particularly sad. A little inconvenient, but parties and their supporters whinge incessantly about balance, surely worth an individual's inconvenience.
    I find YouGov to be the most credible. The underlying fact is that the race has been incredibly static. I don't think the dial has moved very much. And pollsters that are all over the place from week to week are just showing us that their overall methodology isn't as tight as that of YouGov.
    I find yougov least credible for the same reason. There should be outliers fairly frequently due to the number of polls undertaken, but instead nothing moves. If you add to that the people on this forum who have been interviewed multiple times, then it could be the panel is too stale.
  • Chris123Chris123 Posts: 174

    Chris123 said:

    kle4 said:

    Danny565 said:

    Populus will be the test tomorrow. Even as stable as they usually are, surely there'll be a Tory lead or atleast a tie if there genuinely has been movement this week.

    YG static & dead as a floorboard. Can we all agree to bin YG? Regardless of opinions no polling 7 days a week should have this little movement. Somethings badly wrong there.
    Well, we feel like there should have been more movement, but if they end up calling the election better than anyone else, it will have been our expectations that were wrong, not them. I don't think anyone should be certain YouGov must be wrong.
    MikeK said:

    Sad story of the night.

    MissLeahMarie ‏@LittleMissUKIP 4h4 hours ago
    The BBC are disgusting. Decided to tell everyone WHEN WE GOT THERE that only some of us are allowed in for "balance" reasons. BBC bastards.

    I don't see how that is particularly sad. A little inconvenient, but parties and their supporters whinge incessantly about balance, surely worth an individual's inconvenience.
    I find YouGov to be the most credible. The underlying fact is that the race has been incredibly static. I don't think the dial has moved very much. And pollsters that are all over the place from week to week are just showing us that their overall methodology isn't as tight as that of YouGov.
    Random sampling should have random errors. Outliers aren't a flaw of random polls there part and parcel of how it works. In a healthy sample of polls a certain proportion should be outside the margins of error. At a 5% confidence interval one in twenty will be.

    YouGov aren't acting randomly. That means something is wrong. YouGov clearly aren't randomly selecting a sample of the UK under standard distributions.
    Good point. But firstly their sample size is larger than other polls which reduces the margin of error. Secondly, they have also reported swings from day to day just not as large as some of the other polls.
  • Tabman said:

    Tabman said:

    currystar said:

    Pulpstar said:

    currystar said:

    IOS said:

    Anyway - how many people have voted already - at least 15% of the people that are going to vote I reckon.

    Not sure there are many people to shift in the center - just about getting out your core.

    Step forward the ground game.

    As an Eastleigh resident i am stunned by the lack of the feeling of an election in this area. I always felt this was the lib dems safest seat, yet I have not received a single piece of election literature from them and neither have my work colleagues who also live in Eastleigh . It's very odd . The Tories seem to be the only ones who are making an effort. I'm very confused.
    Amazing. I'm seriously beginning to wonder if Ashcroft Question ONE might be the correct one.
    Eastleigh was a 3 way split at the by election. Since then then the LD vote has slipped further and I cannot see UKIP doing any where like the same. I think the Cons are value.
    Normally a week before an election Eastleigh turns orange with the number of lib dem banners and posters. Not this time.
    Surprised at that. LD ground troops there are rabidly busy normally. Keith House keeps them at it.
    OXWAB is a sea of Orange. A fair dash of blue, a little green. Labour is not obvious but helped by several estate agents with red and yellow.
    I happened to be in Jericho last weekend. There were more yellow diamonds than I expected, but on the other hand there were more Labour posters per property than I can remember seeing anywhere in the last decade and a half. Admittedly the urban Oxford bit of OXWAB is not the whole of it, but I'm pretty bullish on a Tory hold here given the amount of tactical LD support that appears to be giving up and going back to Labour, or the principled LD support that's now Green.
    I don't doubt it. But having never been in a tight contest before I find it interesting. I've not been off the main drag in Jericho. Have you been down Botley Road and up Cumnor Hill?
    No, just Jericho (Red with bits of yellow), Wolvercote (Green with bits of yellow) and some surprisingly Blue areas around the Woodstock Road. Well, surprising for Oxford, those kind of houses would be solid Tory anywhere else in the country. Mansion Tax issues?
    Speaking to some colleagues in Abingdon earlier this weekand one said he was getting leaflets every day. I think Con hold but the LDs will keep it close enough to be competitive next time around.
  • A plausible result is a hung Parliament where Labour and the Scots Nationalists have more votes than the Tories and the Liberal Democrats, but the Tories have more votes than Labour and Liberal Democrats. A Tory government would be bound to fall since Labour and the SNP would oppose it. A Labour government would fall unless the Scots Nationalists were prepared to vote for it, since the Tories would oppose it. Miliband having ruled out a confidence and supply arrangement, can anyone see the Scots Nationalists going through the lobbies to support a Labour government, as opposed to abstaining on a confidence motion in it? Would the nationalist core vote really be happy if SNP MPs expressed confidence in a government that was notionally committed to reduce public expenditure and was committed to renew Trident?

    The country would appear ungovernable in those circumstances without the consent of the DUP...
  • Philip_ThompsonPhilip_Thompson Posts: 65,826
    Chris123 said:

    Chris123 said:

    kle4 said:

    Danny565 said:

    Populus will be the test tomorrow. Even as stable as they usually are, surely there'll be a Tory lead or atleast a tie if there genuinely has been movement this week.

    YG static & dead as a floorboard. Can we all agree to bin YG? Regardless of opinions no polling 7 days a week should have this little movement. Somethings badly wrong there.
    Well, we feel like there should have been more movement, but if they end up calling the election better than anyone else, it will have been our expectations that were wrong, not them. I don't think anyone should be certain YouGov must be wrong.
    MikeK said:

    Sad story of the night.

    MissLeahMarie ‏@LittleMissUKIP 4h4 hours ago
    The BBC are disgusting. Decided to tell everyone WHEN WE GOT THERE that only some of us are allowed in for "balance" reasons. BBC bastards.

    I don't see how that is particularly sad. A little inconvenient, but parties and their supporters whinge incessantly about balance, surely worth an individual's inconvenience.
    I find YouGov to be the most credible. The underlying fact is that the race has been incredibly static. I don't think the dial has moved very much. And pollsters that are all over the place from week to week are just showing us that their overall methodology isn't as tight as that of YouGov.
    Random sampling should have random errors. Outliers aren't a flaw of random polls there part and parcel of how it works. In a healthy sample of polls a certain proportion should be outside the margins of error. At a 5% confidence interval one in twenty will be.

    YouGov aren't acting randomly. That means something is wrong. YouGov clearly aren't randomly selecting a sample of the UK under standard distributions.
    Good point. But firstly their sample size is larger than other polls which reduces the margin of error. Secondly, they have also reported swings from day to day just not as large as some of the other polls.
    The swings are all within margin of error. When was the last YouGov outlier? When was the last poll that was outside the margin of error? Something is wrong with the method.

    They may be right, they may be wrong. But they're acting like a broken clock now and not following the laws of random sampling that is supposed to underpin opinion polling. If your sample is wrong your results coule be and since theirs is clearly not randomly representative of the UK it is wrong.
  • OblitusSumMeOblitusSumMe Posts: 9,143
    RodCrosby said:

    RodCrosby said:

    NoEasyDay said:

    Scott_P said:

    @PeteWishart: Important front page in the National tomorrow. This is only going to get bigger. All over for Lab, methinks. https://t.co/t3upVzfxGV

    Two things
    1. Milliband must be very worried about how the SNP issue is playing on the doorstep to keep going on about this.
    2. How does he explain getting a queens speech through, with what must be a deal with the SNP.
    Miliband was very clear about this tonight. Suppose that Cameron resigns, or loses a vote in the House, and so Miliband has a shot at being PM. What Miliband has said he will do is to put forward a Labour Queen Speech, without negotiating with the SNP to secure their support.
    Citation?
    This is what Miliband said tonight:

    "We do want to put forward a Labour Queen Speech, I want a majority Labour government, and it would be for parties in the House of Commons to vote for it."
    OK, thanks, I didn't hear that, and thought I paid close attention.

    But still, a bit muddled. Why would you want others to vote for it, if you already have a majority? You may want to put forward a QS, but in certain circumstances, discretion could be the better part of valour?

    I do detect a slight equivocation from "We will vote down a Tory Queen's Speech in a hung Parliament."
    The bit about Lab majority should probably be in parentheses since it was tangential to the question he was asked which was:

    "And you wouldn't put forward a Queen Speech on the hope they would support it, you just wouldn't go to the Queen and accept her invitation?"

    This was after one of the bits where he was clear that he would not make deals with the SNP. The question is at 43:39 if you are looking for it in the programme on BBC iPlayer.
  • ItwasriggedItwasrigged Posts: 154

    Neil: "Miliband's written off Scotland now."

    Yup! He is leaving it to the Press and Media up here to fight the fires with no water. They don't have the Sun either and growing speculation that the Record might ditch them as well.
  • OblitusSumMeOblitusSumMe Posts: 9,143
    edited April 2015

    A plausible result is a hung Parliament where Labour and the Scots Nationalists have more votes than the Tories and the Liberal Democrats, but the Tories have more votes than Labour and Liberal Democrats. A Tory government would be bound to fall since Labour and the SNP would oppose it. A Labour government would fall unless the Scots Nationalists were prepared to vote for it, since the Tories would oppose it. Miliband having ruled out a confidence and supply arrangement, can anyone see the Scots Nationalists going through the lobbies to support a Labour government, as opposed to abstaining on a confidence motion in it? Would the nationalist core vote really be happy if SNP MPs expressed confidence in a government that was notionally committed to reduce public expenditure and was committed to renew Trident?

    The country would appear ungovernable in those circumstances without the consent of the DUP...

    If the SNP do not support the Labour Queen Speech then recently defeated Labour MPs in Scotland can go back to their former constituents and argue that the SNP wouldn't support Labour to keep the Tories out, and by triggering another election they give the Tories another chance to win the election.

    Anyhow, even people such as I need some beauty sleep. Goodnight all!
  • Chris123Chris123 Posts: 174
    "I find yougov least credible for the same reason. There should be outliers fairly frequently e to the number of polls undertaken, but instead nothing moves. If you add to that the people on this forum who have been interviewed multiple times, then it could be the panel is too stale."

    I get your point. I think I read that they have a panel of up to 500.000 respondents. Big but still not necessarily a representative sample of the population as a whole. Phone pollsters obviously have a wider reach although not everybody will talk to them. It's still a self-selecting sample but on a much larger scale.

    Maybe a key difference is that the people that sign up to an internet panel are more interested in politics, generally more tech savvy and get their news/ information from a variety of online sources whereas the people that answer a phone poll may rely more on traditional (mainstream) media such as TV and newspapers. They may be more susceptible to endorsements and the narrative portrayed in whatever media they read.

    Another key difference is that people may be less willing to disclose things on the phone ("shy Kippers?").
  • Philip_ThompsonPhilip_Thompson Posts: 65,826

    A plausible result is a hung Parliament where Labour and the Scots Nationalists have more votes than the Tories and the Liberal Democrats, but the Tories have more votes than Labour and Liberal Democrats. A Tory government would be bound to fall since Labour and the SNP would oppose it. A Labour government would fall unless the Scots Nationalists were prepared to vote for it, since the Tories would oppose it. Miliband having ruled out a confidence and supply arrangement, can anyone see the Scots Nationalists going through the lobbies to support a Labour government, as opposed to abstaining on a confidence motion in it? Would the nationalist core vote really be happy if SNP MPs expressed confidence in a government that was notionally committed to reduce public expenditure and was committed to renew Trident?

    The country would appear ungovernable in those circumstances without the consent of the DUP...

    If the SNP do not support the Labour Queen Speech then recently defeated Labour MPs in Scotland can go back to their former constituents and argue that the SNP wouldn't support Labour to keep the Tories out, and by triggering another election they give the Tories another chance to win the election.
    Or the SNP can play havoc with Labour's legislative program and say that this isn't working and we need a second referendum. So long as they stay short of voting No Confidence they can cause plenty of mischief.

    The SNP don't need to vote down Labour's bills, they can propose amendments too to mess things up. The SNP want a mess so they can call and win a second referendum. To have them controlling the balance of power is like asking an arsonist to look after the fire extinguishers.
  • Philip_ThompsonPhilip_Thompson Posts: 65,826
    Chris123 said:

    "I find yougov least credible for the same reason. There should be outliers fairly frequently e to the number of polls undertaken, but instead nothing moves. If you add to that the people on this forum who have been interviewed multiple times, then it could be the panel is too stale."

    I get your point. I think I read that they have a panel of up to 500.000 respondents. Big but still not necessarily a representative sample of the population as a whole. Phone pollsters obviously have a wider reach although not everybody will talk to them. It's still a self-selecting sample but on a much larger scale.

    Maybe a key difference is that the people that sign up to an internet panel are more interested in politics, generally more tech savvy and get their news/ information from a variety of online sources whereas the people that answer a phone poll may rely more on traditional (mainstream) media such as TV and newspapers. They may be more susceptible to endorsements and the narrative portrayed in whatever media they read.

    Another key difference is that people may be less willing to disclose things on the phone ("shy Kippers?").

    No they don't have an active panel that big. Since they changed methodology they've cut their panel down to just those who were surveyed in January or February. Which means going back to the same individuals again and again.
  • Chris123Chris123 Posts: 174
    The swings are all within margin of error. When was the last YouGov outlier? When was the last poll that was outside the margin of error? Something is wrong with the method.

    They may be right, they may be wrong. But they're acting like a broken clock now and not following the laws of random sampling that is supposed to underpin opinion polling. If your sample is wrong your results coule be and since theirs is clearly not randomly representative of the UK it is wrong.

    Again a valid point. Is it possible that YouGov does not publish the outliers? They just do a 2nd poll when the results seem to far off and if that poll does not confirm the first poll they then transfer the outlier to an archive? It would be easy for them to do so.
  • Philip_ThompsonPhilip_Thompson Posts: 65,826
    It would be dishonest for a pollster to disregard a poll thats part of a series just because they don't like it's results. Would be far too open to abuse. I'd expect that to be against BPC rules and either way almost certainly not the practice.
  • Sunil_PrasannanSunil_Prasannan Posts: 52,150
    There were only 6 phone polls during January (out of 43 total polls), seven during February (again out of 43), nine during March (out of 51), and assuming there aren't any more late stragglers, 11 in April (out of 65!!)
  • Chris123Chris123 Posts: 174

    It would be dishonest for a pollster to disregard a poll thats part of a series just because they don't like it's results. Would be far too open to abuse. I'd expect that to be against BPC rules and either way almost certainly not the practice.

    Not sure about that. Mike Smithson reported that IPSOS Mori was thinking about not publishing their poll today. If that was against BPC rules why would it have been a subject of discussion? I need to look this up.
  • Chris123Chris123 Posts: 174

    It would be dishonest for a pollster to disregard a poll thats part of a series just because they don't like it's results. Would be far too open to abuse. I'd expect that to be against BPC rules and either way almost certainly not the practice.

    2.6. Organisations conducting privately commissioned surveys have the right to maintain the confidentiality of survey findings...

    http://www.britishpollingcouncil.org/objects.html

    Looks like it is possible in principle that they keep surveys out of the public domain.
  • Sunil_PrasannanSunil_Prasannan Posts: 52,150
    edited May 2015

    It would be dishonest for a pollster to disregard a poll thats part of a series just because they don't like it's results. Would be far too open to abuse. I'd expect that to be against BPC rules and either way almost certainly not the practice.

    There was an Opinium poll back in November that wasn't "intended for publication" and was pulled from their website.

    http://www1.politicalbetting.com/index.php/archives/2014/11/18/opinium-poll-that-slipped-out-has-con-ahead-with-the-lds-down-on-5/
  • Sunil_PrasannanSunil_Prasannan Posts: 52,150

    There were only 6 phone polls during January (out of 43 total polls), seven during February (again out of 43), nine during March (out of 51), and assuming there aren't any more late stragglers, 11 in April (out of 65!!)

    Correction: January was seven phone polls from 43
  • RodCrosbyRodCrosby Posts: 7,737

    A plausible result is a hung Parliament where Labour and the Scots Nationalists have more votes than the Tories and the Liberal Democrats, but the Tories have more votes than Labour and Liberal Democrats. A Tory government would be bound to fall since Labour and the SNP would oppose it. A Labour government would fall unless the Scots Nationalists were prepared to vote for it, since the Tories would oppose it. Miliband having ruled out a confidence and supply arrangement, can anyone see the Scots Nationalists going through the lobbies to support a Labour government, as opposed to abstaining on a confidence motion in it? Would the nationalist core vote really be happy if SNP MPs expressed confidence in a government that was notionally committed to reduce public expenditure and was committed to renew Trident?

    The country would appear ungovernable in those circumstances without the consent of the DUP...

    By "votes", you mean seats, I presume?

    Votes are (virtually) irrelevant under the present constitutional conventions.

  • Chris123Chris123 Posts: 174


    No they don't have an active panel that big. Since they changed methodology they've cut their panel down to just those who were surveyed in January or February. Which means going back to the same individuals again and again.

    You're right. Their methodology states that each poll is based on approx. 2.000 respondents. Since they've conducted upwards of 40 polls during that period they're working with a sample of approx. 80.000 persons.

  • Chris123Chris123 Posts: 174
    But yes it is quite feasible that these 80.000 people are not representative of the UK population as a whole. I agree with you on that count.

    The problem is that I'm not convinced that the phone pollsters are getting a representative sample either. I suspect most people would be less willing to disclose personal preferences to an unknown person over the phone than anonymously via the net to an organization they trust.
  • edmundintokyoedmundintokyo Posts: 17,708
    For any poll with weighting or sampling the normal rules of random sampling (chi squared etc) are actually an over-simplification. If the thing you weight to exactly corresponds to the thing you're trying to find out, sampling error should be zero.

    For example, say you're taking a poll of Argentians and Brits on who owns the Falklands. Say for the sake of argument that the Argentinians unanimously say Argentina, and the Brits unanimously say Britain. Say you call Brits and Argentinians in exact proportion to their populations. Since nobody ever changes their mind, and you have exact correspndance to your weighting variable (nationality) you will get exactly the same result every time.

    What YouGov are doing is weighting to declared voting intention in Jan/Feb (which was in turn weighted/sampled to be representative of the population). This varies randomly a little bit each time because even without the overall vote shares changing, some Jan/Feb Labs will go Con and vice versa, so the weighting variable (Jan/Feb VI) isn't an exact proxy for the thing you're measuring (May VI). But it's pretty close, so it doesn't change much from day to day unless those people's voting intention changes.
  • RodCrosby said:

    By "votes", you mean seats, I presume?

    Votes are (virtually) irrelevant under the present constitutional conventions.

    Votes in the House of Commons are indeed all that matters.
  • Chris123Chris123 Posts: 174

    For any poll with weighting or sampling the normal rules of random sampling (chi squared etc) are actually an over-simplification. If the thing you weight to exactly corresponds to the thing you're trying to find out, sampling error should be zero.

    For example, say you're taking a poll of Argentians and Brits on who owns the Falklands. Say for the sake of argument that the Argentinians unanimously say Argentina, and the Brits unanimously say Britain. Say you call Brits and Argentinians in exact proportion to their populations. Since nobody ever changes their mind, and you have exact correspndance to your weighting variable (nationality) you will get exactly the same result every time.

    What YouGov are doing is weighting to declared voting intention in Jan/Feb (which was in turn weighted/sampled to be representative of the population). This varies randomly a little bit each time because even without the overall vote shares changing, some Jan/Feb Labs will go Con and vice versa, so the weighting variable (Jan/Feb VI) isn't an exact proxy for the thing you're measuring (May VI). But it's pretty close, so it doesn't change much from day to day unless those people's voting intention changes.

    Yes, very illuminating. Thanks!

  • PulpstarPulpstar Posts: 78,417
    UKIP Ordered places with largest chance of conversion - UK Elect Model

    Clacton, Thurrock, Thanet Holds/Gains.

    Rochester and Strood CC
    Castle Point BC
    Boston and Skegness CC
    Dudley North BC *3rd
    Plymouth, Moor View BC *3rd
    South Basildon and East Thurrock CC *3rd
    Great Yarmouth CC *3rd
    Rother Valley CC
  • AndyJSAndyJS Posts: 29,395
    So two Lab candidates now suspended, in Wellingborough and Banff&Buchan.
  • asjohnstoneasjohnstone Posts: 1,276
    It seems strange that in just a weeks time all these questions will be answered.

    A lot of people have made some very firm predictions over a long period of time. In just 7 short days we'll bet getting the first marginals starting to come in.

    EICIPM or EMWNBPM ?
  • asjohnstoneasjohnstone Posts: 1,276
    edited May 2015
    Phrases you never thought you'd hear in your lifetime;

    "Labour are 1.65 to win 5 or less seats in Scotland."

  • RodCrosbyRodCrosby Posts: 7,737

    RodCrosby said:

    By "votes", you mean seats, I presume?

    Votes are (virtually) irrelevant under the present constitutional conventions.

    Votes in the House of Commons are indeed all that matters.
    OK, clarified.

    Riddle me this...

    In the past, in this country, Westminster Model countries, and as far as I can find, parliamentary systems in general.

    If a government loses a crucial vote, it must either

    a) resign (or be dismissed), leading necessarily to the appointment of another government, or

    b) seek a dissolution, under prerogative powers, although such request may occasionally* be denied, leading usually to a)

    Proposition
    The FTPA, in a nutshell, removes b) as an option, leaving a) as the only remaining option.

    * it is accepted that an incumbent government failing its first parliamentary test would in almost no circumstances be granted a dissolution. There are also other circumstances where if an alternative government is possible a dissolution could be denied.
  • RodCrosby said:

    OK, clarified.

    Riddle me this...

    There are a number of views as to the constitutional, as opposed to legal effect of the 2011 Act. The preferable view is that it does not remove option (b), but makes it subject to certain statutory formalities in section 2 of the Act. Once those formalities have been complied with, the Crown has the power by proclamation to dissolve Parliament. An incumbent government could engineer a vote of no confidence in itself, having sought a guarantee from Her Majesty that it would not be unilaterally sacked for doing so. The only occasion where such a contrivance would be impossible would be a government facing its first vote after the meeting of a new Parliament (e.g Baldwin, 1924).

    In any event, if the logic of your argument were right, Callaghan would have had to have resigned in 1979, rather than waiting two weeks for an election. That cannot have been the intention of Parliament in passing the 2011 Act.

    Lord Norton of Louth's view (in the most recent edition of Parliamentary Affairs) is that the Act is far more revolutionary than you suggest. He argues it redefines what you describe as a "crucial vote", such that only a motion in the form "That this House has no confidence in Her Majesty's Government" will amount to such a vote. On his argument, the Government could lose the votes on the Address, on the estimates and on the Finance and Appropriation Bills and remain in office, provided no motion in the statutory form were passed.

    I think the reality in all this is that the we will have to wait and see. I am immensely sceptical of the view that constitutional conventions will necessitate any outcome. Unlike the 2011 Act, they are not statutes, but flexible. After all, many in Australia in 1975 thought the reserve powers of the Crown under the Commonwealth of Australia Constitution Act 1900 were in abeyance or obsolete, only for those powers to be used to determine Whitlam's commission unilaterally.
  • JohnLoonyJohnLoony Posts: 1,790
    I have been phone-polled by Populus. It seemed to be a specific constituency poll for Croydon Central.
  • SimonStClareSimonStClare Posts: 7,976
    It was all going so well, - until he fell off the stage…! #Aerosmith
  • CarlottaVanceCarlottaVance Posts: 60,216
    I think the Tories & the SNP have easily won the poster war in this campaign:

    http://ichef.bbci.co.uk/news/624/media/images/82687000/png/_82687908_locktoriessnp.png
  • PaulMPaulM Posts: 613

    Any thought on what happens if the Royal Baby is born on election day? Its a slight possibility isn't it? Could it affect the media and possibly even turnout?

    Isn't the media only allowed to say things about the weather and "turnout was brisk" etc in the hours the polls are open ?
  • RobDRobD Posts: 60,045
    PaulM said:

    Any thought on what happens if the Royal Baby is born on election day? Its a slight possibility isn't it? Could it affect the media and possibly even turnout?

    Isn't the media only allowed to say things about the weather and "turnout was brisk" etc in the hours the polls are open ?
    The broadcast media, at least. I think the newspaper websites are still allowed to cover things? In any case, a royal babe would be a blessing for editors desperate for a story that can occupy the day!
  • RodCrosbyRodCrosby Posts: 7,737



    I think the reality in all this is that the we will have to wait and see. I am immensely sceptical of the view that constitutional conventions will necessitate any outcome. Unlike the 2011 Act, they are not statutes, but flexible. After all, many in Australia in 1975 thought the reserve powers of the Crown under the Commonwealth of Australia Constitution Act 1900 were in abeyance or obsolete, only for those powers to be used to determine Whitlam's commission unilaterally.

    I get Lord Norton's point, which is surely just a procedural nicety.

    Callaghan could have resigned in 1979, as any PM (on behalf of his government) can do at any time, leading (almost invariably and naturally) to the appointment of the Opposition. He had, however, the option of dissolution, and took it.

    Had the FTPA existed in 1979, the Tories would no doubt have indicated they were not interested in forming a government without an election, just as they did in the real 1979.

    As far as I can envisage, Callaghan would still have been obligated to resign, but would have been re-appointed on a "caretaker" basis (since no-one else was available) by HMQ for the duration of the election campaign - not substantively different to what occurred in the real 1979.

    Balfour simply resigned in 1905, and did not ask for a dissolution (for tactical reasons). The King thus automatically sent for Campbell-Bannerman, even though he was in a small minority in the House (C-B then immediately obtained a dissolution, with spectacular results).

    The FTPA would have made little difference here, surely? Balfour could have brought a No Confidence against C-B, but after first putting him in office by his own resignation, could scarcely have expected to be given another chance himself. An election would have ensued, with C-B as PM or "caretaker" PM.

    It would be interesting to discuss each anomalous situation with reference to the FTPA.

    Starting with Salisbury, 1885.
    Background: a majority Liberal government had resigned mid-way, resulting in the appointment of a minority Tory PM, who sought and obtained a dissolution, resulting in the election of a hung parliament in which the Liberals were the largest party. The incumbent minority Tory PM, nevertheless met Parliament (as second party, the last occasion this has occurred) and was defeated on a Queens Speech, leading to the re-appointment of the previous Liberal PM.
  • @RodCrosby
    It is obviously the case that there must be a government, and if the incumbent Prime Minister resigns, the Leader of the Opposition is likely to be commissioned on the basis that he is the most plausible alternative. That, of course, depends on the incumbent tendering his resignation. It must be considered very unlikely that Her Majesty would unilaterally determine the commission of a premier. She did after all watch "events in Canberra [in 1975] with close interest and attention", and She is a very shrewd political operator. The Act's consequences are therefore likely to be defined by the first Prime Ministers who operate under it without the confidence of the House. If one resigns on a loss of confidence, no doubt it will be said that the "convention" is to do so. If one doesn't a different "convention" may develop...

    Whether a government is described as a "caretaker" or not is perhaps of less importance than whether it can take its case to the country. The Churchill Ministry after VE Day was after all a caretaker ministry, notwithstanding that it had a majority.

    Things would be much simpler if this monstrosity were repealed!
  • Casino_RoyaleCasino_Royale Posts: 60,712
    NEW THREAD
  • AlistairAlistair Posts: 23,670

    Chris123 said:

    kle4 said:

    Danny565 said:

    Populus will be the test tomorrow. Even as stable as they usually are, surely there'll be a Tory lead or atleast a tie if there genuinely has been movement this week.

    YG static & dead as a floorboard. Can we all agree to bin YG? Regardless of opinions no polling 7 days a week should have this little movement. Somethings badly wrong there.
    Well, we feel like there should have been more movement, but if they end up calling the election better than anyone else, it will have been our expectations that were wrong, not them. I don't think anyone should be certain YouGov must be wrong.
    MikeK said:

    Sad story of the night.

    MissLeahMarie ‏@LittleMissUKIP 4h4 hours ago
    The BBC are disgusting. Decided to tell everyone WHEN WE GOT THERE that only some of us are allowed in for "balance" reasons. BBC bastards.

    I don't see how that is particularly sad. A little inconvenient, but parties and their supporters whinge incessantly about balance, surely worth an individual's inconvenience.
    I find YouGov to be the most credible. The underlying fact is that the race has been incredibly static. I don't think the dial has moved very much. And pollsters that are all over the place from week to week are just showing us that their overall methodology isn't as tight as that of YouGov.
    Random sampling should have random errors. Outliers aren't a flaw of random polls there part and parcel of how it works. In a healthy sample of polls a certain proportion should be outside the margins of error. At a 5% confidence interval one in twenty will be.

    YouGov aren't acting randomly. That means something is wrong. YouGov clearly aren't randomly selecting a sample of the UK under standard distributions.
    So that means after 20 polls there is a 1 in 3 chance that there will be NO outliers, the thing about random chance is that it is not predictable.
This discussion has been closed.