I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
You say it’s crap. I ask why you say it’s crap. I am on the Tories getting roughly twice as many seats as predicted here but I’m not going to be as hubristic as to suggest that a complex mathematical model produced by some quite clever people working for a historically accurate pollster that has inputted the voting intention of 34k people nationwide is “crap” without some pretty solid reasoning.
All I will say is watch the movement. The trend in all the published polls so far in the last couple of days is the Tories ticking upwards slightly.
Not enough to save them from a truly terrible result, but I think enough movement to save them from a result like Survation are forecasting. 100+. Probably about 120-140.
We have had poll and MRP overload in this election and I am still not quite sure what to think of it all. I am still not convinced I really trust MRPs, particularly non-Yougov ones.
I wonder why there is such disbelief on here this evening. The polls have been showing a huge Labour victory and the Lib Dem’s and the Tories being neck and neck for second place all the way through. Why are people surprised nothing has changed? The polls may be wrong of course, but pretending this is somehow a surprise is strange. It is just more of the same.
An utter crock. Labour c150 too high, Tories c180 too low, LDs c20 too high, Reform c7 too high
It is very confusing when you keep jumping between your real self and this persona.
I very much want the Conservatives out.
I do not believe the polling. I do not believe the Conservatives will be less than 200. I reckon Labour will be circa 350, which in itself is a decent enough working majority. I think the LDs will be around 50 and I am expecting 8/8 defeats for Farage and nul points for Reform. Does any of that sound outrageous?
All I will say is watch the movement. The trend in all the published polls so far in the last couple of days is the Tories ticking upwards slightly.
Not enough to save them from a truly terrible result, but I think enough movement to save them from a result like Survation are forecasting. 100+. Probably about 120-140.
We have had poll and MRP overload in this election and I am still not quite sure what to think of it all. I am still not convinced I really trust MRPs, particularly non-Yougov ones.
I think we have seen enough (non MRP) polls now to confirm a small swingback to the Tories is taking place in this final stage of the campaign.
How far that goes and whether it will be enough to save them from oblivion remains to to be seen.
I think I've worked it out. Every single poll for the last two years has been utterly and completely wrong. The pollsters have, collectively, fallen for the tricks played by the electorate. Millions of shy Tories have sat on their hands, refusing to tell pollsters their intentions. Meanwhile, a few thousand cocky Labour types have wormed their way on to the pollsters' panels to give a completely distorted result. Sunak's spreadsheets confirm this.
So, on Thursday, the truth will out. That pent-up Tory majority will proudly descend on polling stations up and down the country to mark their cross against the only truly patriotic party, and give Sunak another five years.
In previous elections, pollsters got criticism for herding.
This year, they certainly are not herding.
We cannot then criticise pollsters as BS just because they are outside what results we consider comfortable.
They are really full of crap this time. Or there are a lot of liars out there?
On what basis do you say that?
Nothing that will convince you and the others on here who do not agree with me. It is total bollocks. Labour did not do very well in the local elections. Local elections are not like general etc. We are being lead up the garden path. The Tories will do very badly and Labour will win. Fine. 385 seats I believe they will get. 410 absolute max. Roll on Friday. If I am wrong then please remind me about my prediction. I do not buy it. 200 rural Tories seats. They may loose 85 of them. They still have some other seats in the rest of the country that they will hold and may even gain the odd one. Looking forward to the truth Friday. 150 to 160 for the Tories.
I actually have no idea whether Survation is on the money or not, but I would challenge your take that Labour did poorly in recent local elections. They in fact flipped to Labour plurality every local authority in contention except Walsall and (borderline) Harlow. The other authorities were either Labour plurality already or in the Lib Dem camp. They didn't win every LA seat by any means. But if they had been fighting for Westminster at the time they would have won the vast majority of the larger Westminster seats. Very efficient seat management. They couldn't have had a better preparation for the GE.
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
You say it’s crap. I ask why you say it’s crap. I am on the Tories getting roughly twice as many seats as predicted here but I’m not going to be as hubristic as to suggest that a complex mathematical model produced by some quite clever people working for a historically accurate pollster that has inputted the voting intention of 34k people nationwide is “crap” without some pretty solid reasoning.
Let's wait till Friday.
Gut instinct. It is a con and we are being lead up the garden path. On Friday send me a message and if I am wrong I will admit it. If I am right then fine. I come out and say what some of you believe as well. Not everyone on here will say it. Some do. I am not the only one.
I remember when Osborne or Cameron said to much scoffing from those that follow polls that they thought they’d get a majority in 2015. While unlikely, it’s not completely completely inconceivable that there may be a huge shock coming Thursday night and it is more like hung parliament territory.
Of course shocks are possible, no one would deny a possibility, but even assuming all of the pollsters are just clueless this time, do we really think the circumstances are there to get us to a hung parliament territory? Labour vote failing to show up, Reform nowhere, Tories holding up far better than they are acting like it is, all of it?
I think if we’d been looking at shock hung parliament territory we’d at least have seen one poll that suggested it. It feels a bit of a reach to get to hung parliaments with 15%+ recorded leads.
The upper limit of Tory seats I can envisage is probably c.200. But that requires a lot to go right for them, and a big polling error.
An utter crock. Labour c150 too high, Tories c180 too low, LDs c20 too high, Reform c7 too high
It is very confusing when you keep jumping between your real self and this persona.
I very much want the Conservatives out.
I do not believe the polling. I do not believe the Conservatives will be less than 200. I reckon Labour will be circa 350, which in itself is a decent enough working majority. I think the LDs will be around 50 and I am expecting 8/8 defeats for Farage and nul points for Reform. Does any of that sound outrageous?
I also remember the disappointment of 1992.
You are so scarred by 1992 that is all you remember. The issue is that that is ALL you have. You are ignoring all the evidence available. The polls are the only evidence available. You may be right but you have to give us more than gut feeling.
I wonder why there is such disbelief on here this evening. The polls have been showing a huge Labour victory and the Lib Dem’s and the Tories being neck and neck for second place all the way through. Why are people surprised nothing has changed? The polls may be wrong of course, but pretending this is somehow a surprise is strange. It is just more of the same.
Normalcy bias
There was a caftan maker in Wales who used to talk about that.
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
Put your money where your mouth is then. Amazing odds available.
All evidence says that the Tories are going to get trounced beyond even the worst fears a month ago, and will be lucky to get 100 seats.
@highwayparadise306 , @Mexicanpete and others - I would genuinely like to discuss what trends you are seeing, data you think about etc that makes you bullish on the Tories.
This is a great forum because of the variety of opinion! And is what puts it a class above Twitter in many ways.
Whereas just looking at every poll outside a certain range and calling it crap/BS is the exact opposite of good discussion.
We could probably get a Twitter Robot that does exactly the same thing. Thus making this place the same as what Twitter has become…
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
You say it’s crap. I ask why you say it’s crap. I am on the Tories getting roughly twice as many seats as predicted here but I’m not going to be as hubristic as to suggest that a complex mathematical model produced by some quite clever people working for a historically accurate pollster that has inputted the voting intention of 34k people nationwide is “crap” without some pretty solid reasoning.
Gut instinct. No solid facts. Shy Tories and people lieing about who they will vote for. Plus reforms share will end up 10.0%, the rest of their vote will go to the Tories. My feeling.
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
Whatever happens there are going to be a number of stewards' enquiries as so much data and inference from it varies so wildly. FWIW I think about the same as Survation - for me Tories 70, Labour 460.
Looking at the data lists, projected votes and probabilities seat by seat, is a Gotterdammerung experience, Act III.
An utter crock. Labour c150 too high, Tories c180 too low, LDs c20 too high, Reform c7 too high
It is very confusing when you keep jumping between your real self and this persona.
I very much want the Conservatives out.
I do not believe the polling. I do not believe the Conservatives will be less than 200. I reckon Labour will be circa 350, which in itself is a decent enough working majority. I think the LDs will be around 50 and I am expecting 8/8 defeats for Farage and nul points for Reform. Does any of that sound outrageous?
An utter crock. Labour c150 too high, Tories c180 too low, LDs c20 too high, Reform c7 too high
It is very confusing when you keep jumping between your real self and this persona.
I very much want the Conservatives out.
I do not believe the polling. I do not believe the Conservatives will be less than 200. I reckon Labour will be circa 350, which in itself is a decent enough working majority. I think the LDs will be around 50 and I am expecting 8/8 defeats for Farage and nul points for Reform. Does any of that sound outrageous?
I also remember the disappointment of 1992.
If the Tories will not be under 200 Sunak has been campaigning in some very strange places. He was campaigning in Witney today. I suggest he thinks if he might lose Witney he does not agree with you. The Lib Dem’s have apparently been briefing the press that they ‘think they might get 50 seats’. Given the way political parties try to manage their messages I think that means they expect to win more. So maybe you could tell us what you see they do not?
I remember when Osborne or Cameron said to much scoffing from those that follow polls that they thought they’d get a majority in 2015. While unlikely, it’s not completely completely inconceivable that there may be a huge shock coming Thursday night and it is more like hung parliament territory.
Of course shocks are possible, no one would deny a possibility, but even assuming all of the pollsters are just clueless this time, do we really think the circumstances are there to get us to a hung parliament territory? Labour vote failing to show up, Reform nowhere, Tories holding up far better than they are acting like it is, all of it?
This is possible. Although Labour could be in the 385 to 410 seats territory.
An utter crock. Labour c150 too high, Tories c180 too low, LDs c20 too high, Reform c7 too high
It is very confusing when you keep jumping between your real self and this persona.
I very much want the Conservatives out.
I do not believe the polling. I do not believe the Conservatives will be less than 200. I reckon Labour will be circa 350, which in itself is a decent enough working majority. I think the LDs will be around 50 and I am expecting 8/8 defeats for Farage and nul points for Reform. Does any of that sound outrageous?
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
JLP, Savanta, and two More in Common polls have had a 15pp gap in the past week. There have been fifteen other polls with larger gaps.
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
The other thing that is interesting about the data is the current leader for each constitutency
Labour are currently leading in 498 Lib Dems are leading in 65 Tories are leading in 53 SNP are leading in 8 Green are leading in 2 Plaid are leading in 2 Reform are leading in 1
Could those calling out polls or MRP’s as “tripe” or a “crock” please explain why you think that. You may be right but some working is needed.
I think the individual seat projections a bit iffy. I don't think IOW West will get 20% Green but East 10% for example, more likely the other way around.
I think though that the seat numbers will be about right though, and it's good to see the 95% range shown on the website.
Here's to fun uncle Davey for LOTO. He deserves it on effort alone.
I wonder why there is such disbelief on here this evening. The polls have been showing a huge Labour victory and the Lib Dem’s and the Tories being neck and neck for second place all the way through. Why are people surprised nothing has changed? The polls may be wrong of course, but pretending this is somehow a surprise is strange. It is just more of the same.
Normalcy bias
There was a caftan maker in Wales who used to talk about that.
Indeed so. Didn’t he predict Covid in late January 2020 and everyone told him to shut up so they could talk about wood burning stoves, because “that could never happen”?
We are all prone to normalcy bias. I feel it myself looking at these polls - surely the Tories can’t go under 100 - surely the Lib Dem’s can’t be that close to being the opposition…
Yet that is what the polls say
But this is polling of politics not a pandemic (which has an inexorable and exponential logic). So maybe the polls ARE wrong - but they cannot be entirely dismissed
An utter crock. Labour c150 too high, Tories c180 too low, LDs c20 too high, Reform c7 too high
It is very confusing when you keep jumping between your real self and this persona.
I very much want the Conservatives out.
I do not believe the polling. I do not believe the Conservatives will be less than 200. I reckon Labour will be circa 350, which in itself is a decent enough working majority. I think the LDs will be around 50 and I am expecting 8/8 defeats for Farage and nul points for Reform. Does any of that sound outrageous?
I also remember the disappointment of 1992.
If the Tories will not be under 200 Sunak has been campaigning in some very strange places. He was campaigning in Witney today. I suggest he thinks if he might lose Witney he does not agree with you. The Lib Dem’s have apparently been briefing the press that they ‘think they might get 50 seats’. Given the way political parties try to manage their messages I think that means they expect to win more. So maybe you could tell us what you see they do not?
Yes if the Tories were anywhere close to 200 seats, Sunak would not be campaigning where he has been campaigning all election. Neither would Starmer.
All I will say is watch the movement. The trend in all the published polls so far in the last couple of days is the Tories ticking upwards slightly.
Not enough to save them from a truly terrible result, but I think enough movement to save them from a result like Survation are forecasting. 100+. Probably about 120-140.
We have had poll and MRP overload in this election and I am still not quite sure what to think of it all. I am still not convinced I really trust MRPs, particularly non-Yougov ones.
I think we have seen enough (non MRP) polls now to confirm a small swingback to the Tories is taking place in this final stage of the campaign.
How far that goes and whether it will be enough to save them from oblivion remains to to be seen.
Also the swingback is only taking the tories back to a 22% to 23% vote share which is what the survation mrp implies. And the raw data for those like jLP which puts the tories at 24% headline is actually 22%.
- Some 'traditional pollsters' will be wrong in terms of percentages. Tories towards mid 20s or in the teens fighting Reform. Others may be right, but they can't all be.
- Some models will be wrong in terms of implied seats for a given vote share. Others may be right, or at least close.
Other than for finding betting value, I'm not sure there's much point in arguing why X pollster is wrong. It's impossible to know yet, and also not long until we find out.
Personally I suspect a modest Tory recovery in recent days in the traditional polls saves their blushes vs. the most dire forecasts and we'll be in a position where everyone except Reform and SNP will be content (Labour with at least a good working majority, Tories avoiding oblivion, Lib Dem recovery continued). But who knows.
Lots of people saying the Survation MRP is wrong simply because it looks like a big change from the previous election.
Just because it’s a big swing doesn’t mean the poll is unreliable.
Also, Survation I think have one of the best records going in recent elections.
It is entirely possible for the MRPs to be correct: all it requires is that the Conservative vote is (roughly) halved everywhere, while the Labour vote is up slightly and the LibDem vote flat.
In this situation, a huge number of seats will change hands, especially if - in the South East - Labour supporters tactically vote LibDem, while in much of the rest of the country they go the other way. In that scenario, it's highly likely we see such a result.
What causes me not to take it as my central scenario is this:
(1) Conservative voters are older and more likely to trek to the polls. (2) I think Reform's vote is quite soft, especially outside the Red Wall, where a lot of it is people who don't normally vote (3) Lots of seats are due to be won by LD/Lab on very small margins, and a little bit of differential turnout can make big differences to seat count
My current forecast is Lab 40 Con 26 Ref 12 LD 12
On EC that gives 66 seats for the LDs, but I think that's generous. I think the Con vote holds up better in the South East. I think they are around 40, with the Tories on around 120-130.
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
You say it’s crap. I ask why you say it’s crap. I am on the Tories getting roughly twice as many seats as predicted here but I’m not going to be as hubristic as to suggest that a complex mathematical model produced by some quite clever people working for a historically accurate pollster that has inputted the voting intention of 34k people nationwide is “crap” without some pretty solid reasoning.
Let's wait till Friday.
Gut instinct. It is a con and we are being lead up the garden path. On Friday send me a message and if I am wrong I will admit it. If I am right then fine. I come out and say what some of you believe as well. Not everyone on here will say it. Some do. I am not the only one.
I’m not saying you’re wrong. I am saying you need to do more on a site that is nominally about discussing probabilistic outcomes to back up assertions that something is “crap”.
There was a bloke on here called @eadric who, in the few weeks before lockdown, was banging on about “normalcy bias”. He generally spoke a lot of shite but if he still posted on here he would be pointing out that a lot of people on here are dismissing this poll on the basis that “it could never happen here”. I get the criticisms, I’m not a fan of the very long period in which the survey was conducted, but it is still to be taken seriously..
I agree about normalcy bias !!
Everyone still has it in their heads that one of the two big parties gets a big majority that dwindles then swings the other way because that’s what happened 1983 to 2010
There is nothing that means that has to be the case !!
Cons could do badly … and never recover Labour could win HUGE … and collapse next time
We underestimate this because it doesn’t fit our mental model
Look at how people STILL interpret what has happened since 2010 to now using that model but it’s not what happened is it? Coalition then slim Con majority then Con minority then bigger Con majority then likely collapse
I just do not buy that Survation MRP. They are going to look very foolish come Friday.
How can you rely on a survey when the polling was done over close to a three week period?
MRPs are ridiculous.
It's an interesting point.
In 2017 the MRP did call it right when few others did.
In 2019 the first one was accurate but the second was not.
In 2019, I'm assuming you're taking about the YouGov MRPs?
The first one have the Tories 359 seats, the second 339 seats, the result was 365 Tory seats. So the error for the second one was only 26 Tory seats.
In the grand scheme of things that does not seem like a large error. If applied to this Survation MRP it would not be enough to take the Tories above 100 seats.
Fourteen of the forty five houses on it get CAM delivered to them. I think it stands for Cambridge Alumni Magazine
There can't be many streets outside of Cambridge with that ratio of CAs
I also, genuinely as far as I can tell, live next door to a friend of Banksy's
I know the guy a bit, I've met him several times but only as a friend of a friend. I weirdly don't know him at all as a neighbour or someone I deliver mail to; he has a box for all his mail and parcels, and I never see him when I'm at home
I only ever see him at my mate's house. He's always talking about building projects he's working on for musicians I've heard of. Everybody I know that knows him knows he's a friend of Banksy, but knows not to ask about it
Some who are closer to him, including my good friend, claim to have been to his house and to have seen his several original Banksy artworks
I don't know what to believe, but I don't think my mate is lying
I wonder why there is such disbelief on here this evening. The polls have been showing a huge Labour victory and the Lib Dem’s and the Tories being neck and neck for second place all the way through. Why are people surprised nothing has changed? The polls may be wrong of course, but pretending this is somehow a surprise is strange. It is just more of the same.
Normalcy bias
There was a caftan maker in Wales who used to talk about that.
Indeed so. Didn’t he predict Covid in late January 2020 and everyone told him to shut up so they could talk about wood burning stoves, because “that could never happen”?
We are all prone to normalcy bias. I feel it myself looking at these polls - surely the Tories can’t go under 100 - surely the Lib Dem’s can’t be that close to being the opposition…
Yet that is what the polls say
But this is polling of politics not a pandemic (which has an inexorable and exponential logic). So maybe the polls ARE wrong - but they cannot be entirely dismissed
Yes this is it. Some people are inherently conservative (small c) with their predictions.
It *can* happen here.
Put simply, to flip it, if this *was* the election where we got 475+ for Labour, or the Lib Dem’s close to 2nd, or the Tories on <100 seats - what more proof would we need that that was, at least, *possible*. Answer: not all that much.
In previous elections, pollsters got criticism for herding.
This year, they certainly are not herding.
We cannot then criticise pollsters as BS just because they are outside what results we consider comfortable.
They are really full of crap this time. Or there are a lot of liars out there?
On what basis do you say that?
Nothing that will convince you and the others on here who do not agree with me. It is total bollocks. Labour did not do very well in the local elections. Local elections are not like general etc. We are being lead up the garden path. The Tories will do very badly and Labour will win. Fine. 385 seats I believe they will get. 410 absolute max. Roll on Friday. If I am wrong then please remind me about my prediction. I do not buy it. 200 rural Tories seats. They may loose 85 of them. They still have some other seats in the rest of the country that they will hold and may even gain the odd one. Looking forward to the truth Friday. 150 to 160 for the Tories.
I actually have no idea whether Survation is on the money or not, but I would challenge your take that Labour did poorly in recent local elections. They in fact flipped to Labour plurality every local authority in contention except Walsall and (borderline) Harlow. The other authorities were either Labour plurality already or in the Lib Dem camp. They didn't win every LA seat by any means. But if they had been fighting for Westminster at the time they would have won the vast majority of the larger Westminster seats. Very efficient seat management. They couldn't have had a better preparation for the GE.
May elections are not a useful guide this time. This is a GE, and feelings are different. The consistent news from the polling is that 75-80% of voters don't want a Tory government - possibly more as some loyal Tories will vote Tory, as some Labour stalwarts vote for a Jezza they wanted to lose.
It's the opinions of the approx 80% of the voters who absolutely want them out which will determine the wipeout. Survation are about right. This is not like 1997. Lots of ordinary Tories then wanted a Tory government more than a Labour one, and this was not wholly irrational. I know, I was one of them. The vote split was 43/31 in 1997. Polls have consistently put the Tories on half the Labour vote this time.
All I will say is watch the movement. The trend in all the published polls so far in the last couple of days is the Tories ticking upwards slightly.
Not enough to save them from a truly terrible result, but I think enough movement to save them from a result like Survation are forecasting. 100+. Probably about 120-140.
We have had poll and MRP overload in this election and I am still not quite sure what to think of it all. I am still not convinced I really trust MRPs, particularly non-Yougov ones.
I think we have seen enough (non MRP) polls now to confirm a small swingback to the Tories is taking place in this final stage of the campaign.
How far that goes and whether it will be enough to save them from oblivion remains to to be seen.
Also the swingback is only taking the tories back to a 22% to 23% vote share which is what the survation mrp implies. And the raw data for those like jLP which puts the tories at 24% headline is actually 22%.
Yes - any swingback is helping the Tories move from truly abysmal to really dire, but no better than that.
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
You say it’s crap. I ask why you say it’s crap. I am on the Tories getting roughly twice as many seats as predicted here but I’m not going to be as hubristic as to suggest that a complex mathematical model produced by some quite clever people working for a historically accurate pollster that has inputted the voting intention of 34k people nationwide is “crap” without some pretty solid reasoning.
Let's wait till Friday.
Gut instinct. It is a con and we are being lead up the garden path. On Friday send me a message and if I am wrong I will admit it. If I am right then fine. I come out and say what some of you believe as well. Not everyone on here will say it. Some do. I am not the only one.
I’m not saying you’re wrong. I am saying you need to do more on a site that is nominally about discussing probabilistic outcomes to back up assertions that something is “crap”.
Fair enough. You seem like a decent person. Friday will be interesting!
Lots of people saying the Survation MRP is wrong simply because it looks like a big change from the previous election.
Just because it’s a big swing doesn’t mean the poll is unreliable.
Also, Survation I think have one of the best records going in recent elections.
It is entirely possible for the MRPs to be correct: all it requires is that the Conservative vote is (roughly) halved everywhere, while the Labour vote is up slightly and the LibDem vote flat.
In this situation, a huge number of seats will change hands, especially if - in the South East - Labour supporters tactically vote LibDem, while in much of the rest of the country they go the other way. In that scenario, it's highly likely we see such a result.
What causes me not to take it as my central scenario is this:
(1) Conservative voters are older and more likely to trek to the polls. (2) I think Reform's vote is quite soft, especially outside the Red Wall, where a lot of it is people who don't normally vote (3) Lots of seats are due to be won by LD/Lab on very small margins, and a little bit of differential turnout can make big differences to seat count
My current forecast is Lab 40 Con 26 Ref 12 LD 12
On EC that gives 66 seats for the LDs, but I think that's generous. I think the Con vote holds up better in the South East. I think they are around 40, with the Tories on around 120-130.
Lets go back to 1997. Final polling average for the Tories was around 31% which was pretty bang on to their actual share. That is 8 to 9 points above what the final polling average is likely to show now. And that still only got the Tories 165 seats.
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
Put your money where your mouth is then. Amazing odds available.
All evidence says that the Tories are going to get trounced beyond even the worst fears a month ago, and will be lucky to get 100 seats.
@highwayparadise306 , @Mexicanpete and others - I would genuinely like to discuss what trends you are seeing, data you think about etc that makes you bullish on the Tories.
This is a great forum because of the variety of opinion! And is what puts it a class above Twitter in many ways.
Whereas just looking at every poll outside a certain range and calling it crap/BS is the exact opposite of good discussion.
We could probably get a Twitter Robot that does exactly the same thing. Thus making this place the same as what Twitter has become…
The trend is down for Labour and the latest trends are positive for Tories cf. Reform. Robert Hayward has warned of the undecideds and the shy Tories. Remember them from 1992? I think Curtice extrapolated a majority not dissimilar to mine from regular polls earlier. Was it Labour c 360?
I am also a conspiracy theorist and am awaiting the 3m hordes of ex -pat £1 poms and Apartheid era South African voters leaning blue. Postal voting and voter ID is making me sweat too.
All I will say is watch the movement. The trend in all the published polls so far in the last couple of days is the Tories ticking upwards slightly.
Not enough to save them from a truly terrible result, but I think enough movement to save them from a result like Survation are forecasting. 100+. Probably about 120-140.
We have had poll and MRP overload in this election and I am still not quite sure what to think of it all. I am still not convinced I really trust MRPs, particularly non-Yougov ones.
I think we have seen enough (non MRP) polls now to confirm a small swingback to the Tories is taking place in this final stage of the campaign.
How far that goes and whether it will be enough to save them from oblivion remains to to be seen.
Also the swingback is only taking the tories back to a 22% to 23% vote share which is what the survation mrp implies. And the raw data for those like jLP which puts the tories at 24% headline is actually 22%.
Yeah, but it could continue up to and including polling day and might get Con to mid to upper 20s% in the end?
Or this might be as far as it goes and Con do indeed end up polling around 22%. We probably won't know until the exit poll is revealed.
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
JLP, Savanta, and two More in Common polls have had a 15pp gap in the past week. There have been fifteen other polls with larger gaps.
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
Spot on. I’m not agreeing with this MRP. I’m disagreeing with people who just say it’s crap without any other explanation.
Lots of people saying the Survation MRP is wrong simply because it looks like a big change from the previous election.
Just because it’s a big swing doesn’t mean the poll is unreliable.
Also, Survation I think have one of the best records going in recent elections.
It is entirely possible for the MRPs to be correct: all it requires is that the Conservative vote is (roughly) halved everywhere, while the Labour vote is up slightly and the LibDem vote flat.
In this situation, a huge number of seats will change hands, especially if - in the South East - Labour supporters tactically vote LibDem, while in much of the rest of the country they go the other way. In that scenario, it's highly likely we see such a result.
What causes me not to take it as my central scenario is this:
(1) Conservative voters are older and more likely to trek to the polls. (2) I think Reform's vote is quite soft, especially outside the Red Wall, where a lot of it is people who don't normally vote (3) Lots of seats are due to be won by LD/Lab on very small margins, and a little bit of differential turnout can make big differences to seat count
My current forecast is Lab 40 Con 26 Ref 12 LD 12
On EC that gives 66 seats for the LDs, but I think that's generous. I think the Con vote holds up better in the South East. I think they are around 40, with the Tories on around 120-130.
I find it very hard to disagree with any of that.
I agree with you. The Tories may do slightly better. Let's wait and see.
I think I've worked it out. Every single poll for the last two years has been utterly and completely wrong. The pollsters have, collectively, fallen for the tricks played by the electorate. Millions of shy Tories have sat on their hands, refusing to tell pollsters their intentions. Meanwhile, a few thousand cocky Labour types have wormed their way on to the pollsters' panels to give a completely distorted result. Sunak's spreadsheets confirm this.
So, on Thursday, the truth will out. That pent-up Tory majority will proudly descend on polling stations up and down the country to mark their cross against the only truly patriotic party, and give Sunak another five years.
There's only one person likely to get five years on Thursday, and that's Trump.
Incidentally, it would be rather funny if the lunacy of the Supreme Court meant that Merchan felt free to impose a much harsher penalty than he otherwise would have done. And it may happen.
I think I've worked it out. Every single poll for the last two years has been utterly and completely wrong. The pollsters have, collectively, fallen for the tricks played by the electorate. Millions of shy Tories have sat on their hands, refusing to tell pollsters their intentions. Meanwhile, a few thousand cocky Labour types have wormed their way on to the pollsters' panels to give a completely distorted result. Sunak's spreadsheets confirm this.
So, on Thursday, the truth will out. That pent-up Tory majority will proudly descend on polling stations up and down the country to mark their cross against the only truly patriotic party, and give Sunak another five years.
A striking stat from the latest YouGov research is that, given a free choice, Labour would only poll around 28-29% of the voters actively wanting them as their first preference. All the rest - another 10% plus - comes from LibDem and Green and other minor party voters reluctantly throwing their lot in with Labour in seats the smaller parties will never win.
Worth remembering, when Labour bestride the stage with their humongous majority. Not that anyone will.
Lots of people saying the Survation MRP is wrong simply because it looks like a big change from the previous election.
Just because it’s a big swing doesn’t mean the poll is unreliable.
Also, Survation I think have one of the best records going in recent elections.
It is entirely possible for the MRPs to be correct: all it requires is that the Conservative vote is (roughly) halved everywhere, while the Labour vote is up slightly and the LibDem vote flat.
In this situation, a huge number of seats will change hands, especially if - in the South East - Labour supporters tactically vote LibDem, while in much of the rest of the country they go the other way. In that scenario, it's highly likely we see such a result.
What causes me not to take it as my central scenario is this:
(1) Conservative voters are older and more likely to trek to the polls. (2) I think Reform's vote is quite soft, especially outside the Red Wall, where a lot of it is people who don't normally vote (3) Lots of seats are due to be won by LD/Lab on very small margins, and a little bit of differential turnout can make big differences to seat count
My current forecast is Lab 40 Con 26 Ref 12 LD 12
On EC that gives 66 seats for the LDs, but I think that's generous. I think the Con vote holds up better in the South East. I think they are around 40, with the Tories on around 120-130.
Conservatives 100 to 149 seats has just moved out significantly on skybet now 7/4. Its not the central case now so you can fill your boots.
I just do not buy that Survation MRP. They are going to look very foolish come Friday.
How can you rely on a survey when the polling was done over close to a three week period?
MRPs are ridiculous.
It's an interesting point.
In 2017 the MRP did call it right when few others did.
In 2019 the first one was accurate but the second was not.
In 2019, I'm assuming you're taking about the YouGov MRPs?
The first one have the Tories 359 seats, the second 339 seats, the result was 365 Tory seats. So the error for the second one was only 26 Tory seats.
In the grand scheme of things that does not seem like a large error. If applied to this Survation MRP it would not be enough to take the Tories above 100 seats.
I think MRP might find postal votes being cast already a bit difficult to cope with, as indeed all polsters do, so the earlier one might be more accurate.
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
You say it’s crap. I ask why you say it’s crap. I am on the Tories getting roughly twice as many seats as predicted here but I’m not going to be as hubristic as to suggest that a complex mathematical model produced by some quite clever people working for a historically accurate pollster that has inputted the voting intention of 34k people nationwide is “crap” without some pretty solid reasoning.
Let's wait till Friday.
You can’t really just declare it’s crap and then not say why? Or then telling us to wait until the actual result.
I have qualms about the timing of this, does this pressage a DUP collapse on Thursday?
Sir Jeffrey Donaldson, former leader of the DUP, is to face additional historical sex offence charges in court on Wednesday.
Donaldson, 61, is expected to appear at Newry magistrates’ court for a preliminary enquiry (PE) hearing to establish if there is sufficient evidence to send him for trial.
During previous court hearings, Donaldson had faced 11 holding charges brought by police.
However, an evidence file has since been reviewed by the Public Prosecution Service (PPS) and the former DUP leader will now face 18 charges.
He will face one charge of rape, four of gross indecency and 13 charges of indecent assault. The charges span a time period between 1985 and 2008.
The previous holding charges will be withdrawn in court on Wednesday.
If Sir Keir truly is a lucky general then Nicola Sturgeon gets charged tomorrow.
Why qualms? The criminal justice process should proceed in a normal way completely blind to electoral events. Decisions should neither be brought forward nor delayed due to an election.
Say he is found not guilty in the future and the DUP lose many seats on Thursday then I think that would be most unfair.
A delay of one week could have avoided a mess.
The bulk of the damage was done in the immediate aftermath of his arrest.
The DUP had cut all ties with him by first thing the next morning. There'd be very little positive benefit for them if he is eventually acquitted - in fact, it might even hurt them by making them look cruel for not having given him any benefit of the doubt whatsoever.
Manchin is not the most loyal senator. I suspect the reason he was persuaded to keep quiet is that he was told by Schumer that Biden is going to quit on his own.
I wonder why there is such disbelief on here this evening. The polls have been showing a huge Labour victory and the Lib Dem’s and the Tories being neck and neck for second place all the way through. Why are people surprised nothing has changed? The polls may be wrong of course, but pretending this is somehow a surprise is strange. It is just more of the same.
Normalcy bias
There was a caftan maker in Wales who used to talk about that.
Indeed so. Didn’t he predict Covid in late January 2020 and everyone told him to shut up so they could talk about wood burning stoves, because “that could never happen”?
We are all prone to normalcy bias. I feel it myself looking at these polls - surely the Tories can’t go under 100 - surely the Lib Dem’s can’t be that close to being the opposition…
Yet that is what the polls say
But this is polling of politics not a pandemic (which has an inexorable and exponential logic). So maybe the polls ARE wrong - but they cannot be entirely dismissed
It’s not a poll - it’s a poll, fed through a model.
The poll might be right but the model wrong. As Foxy says, when you dig into it, there are obviously anomalous local results. The question is whether, in aggregate, these cancel out so that the model is broadly right, or whether there are crucial local factors that the model is missing?
I was just thinking that this Tory government is in its last two or three days, after fourteen years. I'm often Tory-friendly, but I must admit that the thought gave me a smile.
Incidentally:
1979-1997 Conservative - ~18 years 1997-2010 Labour - ~13 years 2010-2024 Coalition / Conservative - ~14 years.
I think a Labour government will be in for at least three terms; not the least because the Tories will be in no position to challenge them in the next term.
how ling Labour have in power depends on who they choose as leader over the next two, three or four parliaments....
Indeed. But Thatcher got nearly twelve years; Blair ten years. I think Starmer will be set for at least two elections, if his health remains (hope so...) and he remains interested in the job and he doesn't try a referendum on the EU...
By the way I do think those suggesting that the Tories stay above 200 seats have forgotten the fact that in 1997 they only won 165 and that was with 30& of the vote.
My view was roughly 130 based on the current circumstances, but the MRP might be picking up an incredibly brutal tactical vote.
I think I've worked it out. Every single poll for the last two years has been utterly and completely wrong. The pollsters have, collectively, fallen for the tricks played by the electorate. Millions of shy Tories have sat on their hands, refusing to tell pollsters their intentions. Meanwhile, a few thousand cocky Labour types have wormed their way on to the pollsters' panels to give a completely distorted result. Sunak's spreadsheets confirm this.
So, on Thursday, the truth will out. That pent-up Tory majority will proudly descend on polling stations up and down the country to mark their cross against the only truly patriotic party, and give Sunak another five years.
There's only one person likely to get five years on Thursday, and that's Trump.
Incidentally, it would be rather funny if the lunacy of the Supreme Court meant that Merchan felt free to impose a much harsher penalty than he otherwise would have done. And it may happen.
I'm expecting either probation, or a sentence to be set for after the election. Even though appeals would mean he wouldn't serve time beforehand it avoids argument of even trying to enforce it.
I've been doing a rewatch of Star Trek TNG for the first time in 25 years, and coming upon an episode where an elderly legend is past their prime and being coddled by their loved ones to achieve a last great task even at cost of their pride, health, and dignity, is hitting close to home.
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
JLP, Savanta, and two More in Common polls have had a 15pp gap in the past week. There have been fifteen other polls with larger gaps.
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
Spot on. I’m not agreeing with this MRP. I’m disagreeing with people who just say it’s crap without any other explanation.
And that is classic normalcy bias. It is generally found when people are confronting highly unusual and UNPLEASANT phenomena - plagues, invasions, electoral catastrophes, planes flying into New York towers
Some of the people dismissing these polls don’t want to believe them because they presage an electoral wipe out which is distressing to them. That doesn’t mean these PBers are wrong, but that emotion is surely in play
I suspect a couple of PBers don’t want to believe these polls because they can’t believe the result will be THAT good and desirable
I always think you can tell plenty from where the party leaders go. Sunak went to Witney today. People were generally surprised that he chose to go to a seemingly safe Tory seat. The Survation current state of the parties is Tory 32.9% Lib Dem’s 32.2%. Maybe this MRP is what Tory HQ is seeing.
By the way I do think those suggesting that the Tories stay above 200 seats have forgotten the fact that in 1997 they only won 165 and that was with 30& of the vote.
My view was roughly 130 based on the current circumstances, but the MRP might be picking up an incredibly brutal tactical vote.
150 would be a fantastic result for them. Sunak's words and their strategy in 20k seats acknowledging they are potentially at risk shows that very clearly.
The other thing that is interesting about the data is the current leader for each constitutency
Labour are currently leading in 498 Lib Dems are leading in 65 Tories are leading in 53 SNP are leading in 8 Green are leading in 2 Plaid are leading in 2 Reform are leading in 1
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
Put your money where your mouth is then. Amazing odds available.
All evidence says that the Tories are going to get trounced beyond even the worst fears a month ago, and will be lucky to get 100 seats.
@highwayparadise306 , @Mexicanpete and others - I would genuinely like to discuss what trends you are seeing, data you think about etc that makes you bullish on the Tories.
This is a great forum because of the variety of opinion! And is what puts it a class above Twitter in many ways.
Whereas just looking at every poll outside a certain range and calling it crap/BS is the exact opposite of good discussion.
We could probably get a Twitter Robot that does exactly the same thing. Thus making this place the same as what Twitter has become…
The trend is down for Labour and the latest trends are positive for Tories cf. Reform. Robert Hayward has warned of the undecideds and the shy Tories. Remember them from 1992? I think Curtice extrapolated a majority not dissimilar to mine from regular polls earlier. Was it Labour c 360?
I am also a conspiracy theorist and am awaiting the 3m hordes of ex -pat £1 poms and Apartheid era South African voters leaning blue. Postal voting and voter ID is making me sweat too.
Remember 1992!
This parody isn’t funny anymore. Even I gave up the Truss schtick eventually. The ex-pat angle is particularly unfunny.
And c.360 Tory seats is not what Curtice is predicting.
To get to a 1992 style result the polls would have to be twice as wrong as they were leading into that election. Accuracy has got better, not worse, since then.
Maybe I should have stuck to my beliefs two years ago that the tories were viscerally loathed by large swathes of Britain.
Clearly been spending too long lately in leafy Surrey. Although they’re not popular here either, if you can hear the voices over the din of the leaf blowers.
In previous elections, pollsters got criticism for herding.
This year, they certainly are not herding.
We cannot then criticise pollsters as BS just because they are outside what results we consider comfortable.
They are really full of crap this time. Or there are a lot of liars out there?
On what basis do you say that?
Nothing that will convince you and the others on here who do not agree with me. It is total bollocks. Labour did not do very well in the local elections. Local elections are not like general etc. We are being lead up the garden path. The Tories will do very badly and Labour will win. Fine. 385 seats I believe they will get. 410 absolute max. Roll on Friday. If I am wrong then please remind me about my prediction. I do not buy it. 200 rural Tories seats. They may loose 85 of them. They still have some other seats in the rest of the country that they will hold and may even gain the odd one. Looking forward to the truth Friday. 150 to 160 for the Tories.
I actually have no idea whether Survation is on the money or not, but I would challenge your take that Labour did poorly in recent local elections. They in fact flipped to Labour plurality every local authority in contention except Walsall and (borderline) Harlow. The other authorities were either Labour plurality already or in the Lib Dem camp. They didn't win every LA seat by any means. But if they had been fighting for Westminster at the time they would have won the vast majority of the larger Westminster seats. Very efficient seat management. They couldn't have had a better preparation for the GE.
May elections are not a useful guide this time. This is a GE, and feelings are different. The consistent news from the polling is that 75-80% of voters don't want a Tory government - possibly more as some loyal Tories will vote Tory, as some Labour stalwarts vote for a Jezza they wanted to lose.
It's the opinions of the approx 80% of the voters who absolutely want them out which will determine the wipeout. Survation are about right. This is not like 1997. Lots of ordinary Tories then wanted a Tory government more than a Labour one, and this was not wholly irrational. I know, I was one of them. The vote split was 43/31 in 1997. Polls have consistently put the Tories on half the Labour vote this time.
This is totally different from 1997.
I am at work 5 am, Friday morning, if I wasn't, I think I would stay up all night, could be a night to tell your Grandkids about, were you up for Sunak
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
JLP, Savanta, and two More in Common polls have had a 15pp gap in the past week. There have been fifteen other polls with larger gaps.
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
Spot on. I’m not agreeing with this MRP. I’m disagreeing with people who just say it’s crap without any other explanation.
And that is classic normalcy bias. It is generally found when people are confronting highly unusual and UNPLEASANT phenomena - plagues, invasions, electoral catastrophes, planes flying into New York towers
Some of the people dismissing these polls don’t want to believe them because they presage an electoral wipe out which is distressing to them. That doesn’t mean these PBers are wrong, but that emotion is surely in play
I suspect a couple of PBers don’t want to believe these polls because they can’t believe the result will be THAT good and desirable
I've never voted Conservative in my life, but I don't want to see them eviscerated. They are part of the national furniture. What I would like to see is a revitalised Conservative Party led by a Rory Stewartesque leader. Surely you would too.
I've been doing a rewatch of Star Trek TNG for the first time in 25 years, and coming upon an episode where an elderly legend is past their prime and being coddled by their loved ones to achieve a last great task even at cost of their pride, health, and dignity, is hitting close to home.
I just do not buy that Survation MRP. They are going to look very foolish come Friday.
How can you rely on a survey when the polling was done over close to a three week period?
MRPs are ridiculous.
Their reputation is mostly based on the fact that they called Canterbury correctly in 2017. But when you think about it, that's not a huge amount of evidence that they're as good as people say.
I simply don’t buy its projection especially with a likely late swing from Reform to the Tories .
Yes. It is complete garage.
Another bullshit poll out the way.
Bullshit poll = it can't be true it can't be true please let it not be true
It is not true. It is crap.
Put your money where your mouth is then. Amazing odds available.
All evidence says that the Tories are going to get trounced beyond even the worst fears a month ago, and will be lucky to get 100 seats.
@highwayparadise306 , @Mexicanpete and others - I would genuinely like to discuss what trends you are seeing, data you think about etc that makes you bullish on the Tories.
This is a great forum because of the variety of opinion! And is what puts it a class above Twitter in many ways.
Whereas just looking at every poll outside a certain range and calling it crap/BS is the exact opposite of good discussion.
We could probably get a Twitter Robot that does exactly the same thing. Thus making this place the same as what Twitter has become…
The trend is down for Labour and the latest trends are positive for Tories cf. Reform. Robert Hayward has warned of the undecideds and the shy Tories. Remember them from 1992? I think Curtice extrapolated a majority not dissimilar to mine from regular polls earlier. Was it Labour c 360?
I am also a conspiracy theorist and am awaiting the 3m hordes of ex -pat £1 poms and Apartheid era South African voters leaning blue. Postal voting and voter ID is making me sweat too.
Remember 1992!
This parody isn’t funny anymore. Even I gave up the Truss schtick eventually. The ex-pat angle is particularly unfunny.
And c.360 Tory seats is not what Curtice is predicting.
To get to a 1992 style result the polls would have to be twice as wrong as they were leading into that election. Accuracy has got better, not worse, since then.
Hunter Biden has joined meetings with President Biden and his top aides this week at the White House, four people familiar with the matter tell my colleagues, who are told the reaction from some senior White House staff has been, “What the hell is happening?”
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
JLP, Savanta, and two More in Common polls have had a 15pp gap in the past week. There have been fifteen other polls with larger gaps.
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
Spot on. I’m not agreeing with this MRP. I’m disagreeing with people who just say it’s crap without any other explanation.
And that is classic normalcy bias. It is generally found when people are confronting highly unusual and UNPLEASANT phenomena - plagues, invasions, electoral catastrophes, planes flying into New York towers
Some of the people dismissing these polls don’t want to believe them because they presage an electoral wipe out which is distressing to them. That doesn’t mean these PBers are wrong, but that emotion is surely in play
I suspect a couple of PBers don’t want to believe these polls because they can’t believe the result will be THAT good and desirable
I've never voted Conservative in my life, but I don't want to see them eviscerated. They are part of the national furniture. What I would like to see is a revitalised Conservative Party led by a Rory Stewartesque leader. Surely you would too.
It's not going to happen. They need a punishment beating a second time before they learn their lesson.
Clicking round the Survation map, I’m struck by a few three-way marginals which must cause some uncertainty. Obviously there’s straight margin of error and changes between fieldwork and polling day.
But it must hamper any concerted tactical effort (“everyone winning here”), so IMO probably benefits Con at the margins??
Eg: Ynys Mon (L/C/PC all within 5 pts) West Worcs (G/L/C within 2) S.Shrops (C/L/LD within 3) St Neots (C/L/LD within 4) Bicester within 5 Didcot 4
Survation is showing the Greens taking West Worcestershire, but not Bristol Central, Waveney Valley or indeed North Herefordshire. That's... unexpected.
I've been doing a rewatch of Star Trek TNG for the first time in 25 years, and coming upon an episode where an elderly legend is past their prime and being coddled by their loved ones to achieve a last great task even at cost of their pride, health, and dignity, is hitting close to home.
Sarek?
Spot on.
S1 was of, er, variable quality, S2 was hit or miss, but S3 has a lot of great episodes I had forgotten.
I'm thinking about this purely in the change in Conservative vote percentage since 2019.
Even the most optimistic recent vote percentage of 25% (in the 15-27 June Survation MRP) sees the Tories losing 19 percentage points. There is no way that standard modelling can take that into account. Due to the probability of increased tactical voting it's likely that the opposition parties will become much more efficient in their votes and there'll be a much larger structural swing away from the Tories. ie. after the election, put in the 2019 percentages into a calculator with the new results and you'll get nothing like the notional 2019 results.
also to put it into context in 1997 John major only lost 11.2 percentage points from the 1992 election win. The only real comparison is the 1918 general election where the Liberal party lost 31.2 percentage points and 236 seats.
I think I've worked it out. Every single poll for the last two years has been utterly and completely wrong. The pollsters have, collectively, fallen for the tricks played by the electorate. Millions of shy Tories have sat on their hands, refusing to tell pollsters their intentions. Meanwhile, a few thousand cocky Labour types have wormed their way on to the pollsters' panels to give a completely distorted result. Sunak's spreadsheets confirm this.
So, on Thursday, the truth will out. That pent-up Tory majority will proudly descend on polling stations up and down the country to mark their cross against the only truly patriotic party, and give Sunak another five years.
A striking stat from the latest YouGov research is that, given a free choice, Labour would only poll around 28-29% of the voters actively wanting them as their first preference. All the rest - another 10% plus - comes from LibDem and Green and other minor party voters reluctantly throwing their lot in with Labour in seats the smaller parties will never win.
Worth remembering, when Labour bestride the stage with their humongous majority. Not that anyone will.
Yes. And that is before Labour have done anything in government to piss people off.
If the Tories are reduced to under 100 seats then it could be hard for Labour to use the Tory boogeyman to keep those voters in line. It's not implausible that the 2029 GE would see no party win more than 30% of the vote. And then, with FPTP, anything could happen.
I've been doing a rewatch of Star Trek TNG for the first time in 25 years, and coming upon an episode where an elderly legend is past their prime and being coddled by their loved ones to achieve a last great task even at cost of their pride, health, and dignity, is hitting close to home.
Sarek?
Spot on.
S1 was of, er, variable quality, S2 was hit or miss, but S3 has a lot of great episodes I had forgotten.
Season 3 was when Michael Piller joined as head of the writing staff and helped give us many more episodes of Trek we would not have had.
Maybe I should have stuck to my beliefs two years ago that the tories were viscerally loathed by large swathes of Britain.
Clearly been spending too long lately in leafy Surrey. Although they’re not popular here either, if you can hear the voices over the din of the leaf blowers.
If I have this correct, a lot of PBers are more engaged with the issue of Tory possible wipeout than lots of Labour seats.
It seems to me this is right and captures the zeitgeist. The utterly distinctive feel of this election is Get Tories Smashed not Get Labour In - that was 1997.
Which is why, I think, Survation are about right. 75-80% of voters are ganging up in effect to ensure the Tories are decimated and both conscious and unconscious tactical voting will be enormous. In a sense I am an invisible example. I shall vote Labour. If I voted in one of the Tory/LD seats I would transfer to LD without thinking at all, and wouldn't even reckon I was voting tactically.
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
JLP, Savanta, and two More in Common polls have had a 15pp gap in the past week. There have been fifteen other polls with larger gaps.
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
Spot on. I’m not agreeing with this MRP. I’m disagreeing with people who just say it’s crap without any other explanation.
And that is classic normalcy bias. It is generally found when people are confronting highly unusual and UNPLEASANT phenomena - plagues, invasions, electoral catastrophes, planes flying into New York towers
Some of the people dismissing these polls don’t want to believe them because they presage an electoral wipe out which is distressing to them. That doesn’t mean these PBers are wrong, but that emotion is surely in play
I suspect a couple of PBers don’t want to believe these polls because they can’t believe the result will be THAT good and desirable
I've never voted Conservative in my life, but I don't want to see them eviscerated. They are part of the national furniture. What I would like to see is a revitalised Conservative Party led by a Rory Stewartesque leader. Surely you would too.
Are you serious? YES i want them eviscerated: I want the Tories dead and buried, and then I want the corpse of the Tories exhumed so the body can be pulped and the resultant gore burned in a nuclear furnace which turns it to ashes which can be eaten by pigs which are flown to the moon to excrete the dung
I haven't commented for a long time, but I think this Survation MRP must be wrong. There is surely ample evidence of a modest tory recovery and a significant labour decline towards say 38% The gap on other polls seems to be around 15% I cannot believe that the Tories will be on 64 and my hunch is that the SNP will do much better than this poll.
I see this as an outlier - if all the stars aligned for Labour this is a conceivable option.
I am not in the habit of Eating Hats, but I would be astonished if the Tories don't get over 100 seats
JLP, Savanta, and two More in Common polls have had a 15pp gap in the past week. There have been fifteen other polls with larger gaps.
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
Spot on. I’m not agreeing with this MRP. I’m disagreeing with people who just say it’s crap without any other explanation.
And that is classic normalcy bias. It is generally found when people are confronting highly unusual and UNPLEASANT phenomena - plagues, invasions, electoral catastrophes, planes flying into New York towers
Some of the people dismissing these polls don’t want to believe them because they presage an electoral wipe out which is distressing to them. That doesn’t mean these PBers are wrong, but that emotion is surely in play
I suspect a couple of PBers don’t want to believe these polls because they can’t believe the result will be THAT good and desirable
I suspect I may be one of the latter ones. Still forecasting a 70 majority based on feelings. Which defies all available empirical evidence.
Comments
Not enough to save them from a truly terrible result, but I think enough movement to save them from a result like Survation are forecasting. 100+. Probably about 120-140.
We have had poll and MRP overload in this election and I am still not quite sure what to think of it all. I am still not convinced I really trust MRPs, particularly non-Yougov ones.
I do not believe the polling. I do not believe the Conservatives will be less than 200. I reckon Labour will be circa 350, which in itself is a decent enough working majority. I think the LDs will be around 50 and I am expecting 8/8 defeats for Farage and nul points for Reform. Does any of that sound outrageous?
I also remember the disappointment of 1992.
Survation phone (updates tomorrow am)
Whitestone Insight
Lord Ashcroft
Ipsos Mori
People Polling
How far that goes and whether it will be enough to save them from oblivion remains to to be seen.
So, on Thursday, the truth will out. That pent-up Tory majority will proudly descend on polling stations up and down the country to mark their cross against the only truly patriotic party, and give Sunak another five years.
The upper limit of Tory seats I can envisage is probably c.200. But that requires a lot to go right for them, and a big polling error.
Latecomers to the scene can still get 9/4 with BetVictor. Early bids can sit snugly on what will at the very worst be a good value loser.
My guess is that it's even money, and unlikely to be more than a couple of thousand votes in it either way.
This is a great forum because of the variety of opinion! And is what puts it a class above Twitter in many ways.
Whereas just looking at every poll outside a certain range and calling it crap/BS is the exact opposite of good discussion.
We could probably get a Twitter Robot that does exactly the same thing. Thus making this place the same as what Twitter has become…
Looking at the data lists, projected votes and probabilities seat by seat, is a Gotterdammerung experience, Act III.
So maybe you could tell us what you see they do not?
This Survation MRP has an implied gap of 19pp, exactly the same as the R&W poll an hour earlier, and lower than the gap in the most recent Deltapoll, We Think, Techne, and Opinium polls.
The MRP algorithm may be wrong, or the application of it may be wrong. But the underlying sample doesn't seem like a massive outlier to me.
Labour are currently leading in 498
Lib Dems are leading in 65
Tories are leading in 53
SNP are leading in 8
Green are leading in 2
Plaid are leading in 2
Reform are leading in 1
I think though that the seat numbers will be about right though, and it's good to see the 95% range shown on the website.
Here's to fun uncle Davey for LOTO. He deserves it on effort alone.
We are all prone to normalcy bias. I feel it myself looking at these polls - surely the Tories can’t go under 100 - surely the Lib Dem’s can’t be that close to being the opposition…
Yet that is what the polls say
But this is polling of politics not a pandemic (which has an inexorable and exponential logic). So maybe the polls ARE wrong - but they cannot be entirely dismissed
- Some 'traditional pollsters' will be wrong in terms of percentages. Tories towards mid 20s or in the teens fighting Reform. Others may be right, but they can't all be.
- Some models will be wrong in terms of implied seats for a given vote share. Others may be right, or at least close.
Other than for finding betting value, I'm not sure there's much point in arguing why X pollster is wrong. It's impossible to know yet, and also not long until we find out.
Personally I suspect a modest Tory recovery in recent days in the traditional polls saves their blushes vs. the most dire forecasts and we'll be in a position where everyone except Reform and SNP will be content (Labour with at least a good working majority, Tories avoiding oblivion, Lib Dem recovery continued). But who knows.
In this situation, a huge number of seats will change hands, especially if - in the South East - Labour supporters tactically vote LibDem, while in much of the rest of the country they go the other way. In that scenario, it's highly likely we see such a result.
What causes me not to take it as my central scenario is this:
(1) Conservative voters are older and more likely to trek to the polls.
(2) I think Reform's vote is quite soft, especially outside the Red Wall, where a lot of it is people who don't normally vote
(3) Lots of seats are due to be won by LD/Lab on very small margins, and a little bit of differential turnout can make big differences to seat count
My current forecast is
Lab 40
Con 26
Ref 12
LD 12
On EC that gives 66 seats for the LDs, but I think that's generous. I think the Con vote holds up better in the South East. I think they are around 40, with the Tories on around 120-130.
Everyone still has it in their heads that one of the two big parties gets a big majority that dwindles then swings the other way because that’s what happened 1983 to 2010
There is nothing that means that has to be the case !!
Cons could do badly … and never recover
Labour could win HUGE … and collapse next time
We underestimate this because it doesn’t fit our mental model
Look at how people STILL interpret what has happened since 2010 to now using that model but it’s not what happened is it? Coalition then slim Con majority then Con minority then bigger Con majority then likely collapse
The first one have the Tories 359 seats, the second 339 seats, the result was 365 Tory seats. So the error for the second one was only 26 Tory seats.
In the grand scheme of things that does not seem like a large error. If applied to this Survation MRP it would not be enough to take the Tories above 100 seats.
I know the guy a bit, I've met him several times but only as a friend of a friend. I weirdly don't know him at all as a neighbour or someone I deliver mail to; he has a box for all his mail and parcels, and I never see him when I'm at home
I only ever see him at my mate's house. He's always talking about building projects he's working on for musicians I've heard of. Everybody I know that knows him knows he's a friend of Banksy, but knows not to ask about it
Some who are closer to him, including my good friend, claim to have been to his house and to have seen his several original Banksy artworks
I don't know what to believe, but I don't think my mate is lying
It's equally likely they're rong the other way.
It *can* happen here.
Put simply, to flip it, if this *was* the election where we got 475+ for Labour, or the Lib Dem’s close to 2nd, or the Tories on <100 seats - what more proof would we need that that was, at least, *possible*. Answer: not all that much.
It's the opinions of the approx 80% of the voters who absolutely want them out which will determine the wipeout. Survation are about right. This is not like 1997. Lots of ordinary Tories then wanted a Tory government more than a Labour one, and this was not wholly irrational. I know, I was one of them. The vote split was 43/31 in 1997. Polls have consistently put the Tories on half the Labour vote this time.
This is totally different from 1997.
I am also a conspiracy theorist and am awaiting the 3m hordes of ex -pat £1 poms and Apartheid era South African voters leaning blue. Postal voting and voter ID is making me sweat too.
Remember 1992!
Or this might be as far as it goes and Con do indeed end up polling around 22%. We probably won't know until the exit poll is revealed.
Incidentally, it would be rather funny if the lunacy of the Supreme Court meant that Merchan felt free to impose a much harsher penalty than he otherwise would have done. And it may happen.
Worth remembering, when Labour bestride the stage with their humongous majority. Not that anyone will.
It may be, but please explain why?
The DUP had cut all ties with him by first thing the next morning. There'd be very little positive benefit for them if he is eventually acquitted - in fact, it might even hurt them by making them look cruel for not having given him any benefit of the doubt whatsoever.
Manchin is not the most loyal senator. I suspect the reason he was persuaded to keep quiet is that he was told by Schumer that Biden is going to quit on his own.
If I'm right, bet on Harris
The poll might be right but the model wrong. As Foxy says, when you dig into it, there are obviously anomalous local results. The question is whether, in aggregate, these cancel out so that the model is broadly right, or whether there are crucial local factors that the model is missing?
My view was roughly 130 based on the current circumstances, but the MRP might be picking up an incredibly brutal tactical vote.
Some of the people dismissing these polls don’t want to believe them because they presage an electoral wipe out which is distressing to them. That doesn’t mean these PBers are wrong, but that emotion is surely in play
I suspect a couple of PBers don’t want to believe these polls because they can’t believe the result will be THAT good and desirable
And c.360 Tory seats is not what Curtice is predicting.
To get to a 1992 style result the polls would have to be twice as wrong as they were leading into that election. Accuracy has got better, not worse, since then.
https://en.m.wikipedia.org/wiki/Opinion_polling_for_the_1992_United_Kingdom_general_election
Clearly been spending too long lately in leafy Surrey. Although they’re not popular here either, if you can hear the voices over the din of the leaf blowers.
Counted stakeboards again (different route): 16 LD, 0 Lab, 0 Con, 0 Green.
But did see 5 Labour canvassers in a team, led by one (1) Nick Palmer.
Stopped for a brief an congenial chat, wished each other luck (both of us with crossed fingers, naturally), and continued on.
Where is Sunak spending his final days campaigning? Normally safe seats. We’ve been to…
Stratford on Avon (notional Con majority 19,020)
Whitney (15,674)
Banbury (13,799)
Nuneaton (13,144)
Stoke on Trent South (15,393)
Hinckley & Bosworth (22,851)
@JackSurvation
These are the seats with a new leader in our final MRP projection vs. the 28th June - a lot of toss-ups.
https://x.com/JackSurvation/status/1808213390634242254
@robfordmancs
Weald of Kent was definitely on my list of "seats Cons will never lose" - and yet...
16/1 with the magic sign...
Hunter Biden has joined meetings with President Biden and his top aides this week at the White House, four people familiar with the matter tell my colleagues, who are told the reaction from some senior White House staff has been, “What the hell is happening?”
But it must hamper any concerted tactical effort (“everyone winning here”), so IMO probably benefits Con at the margins??
Eg:
Ynys Mon (L/C/PC all within 5 pts)
West Worcs (G/L/C within 2)
S.Shrops (C/L/LD within 3)
St Neots (C/L/LD within 4)
Bicester within 5
Didcot 4
S1 was of, er, variable quality, S2 was hit or miss, but S3 has a lot of great episodes I had forgotten.
Even the most optimistic recent vote percentage of 25% (in the 15-27 June Survation MRP) sees the Tories losing 19 percentage points. There is no way that standard modelling can take that into account. Due to the probability of increased tactical voting it's likely that the opposition parties will become much more efficient in their votes and there'll be a much larger structural swing away from the Tories. ie. after the election, put in the 2019 percentages into a calculator with the new results and you'll get nothing like the notional 2019 results.
also to put it into context in 1997 John major only lost 11.2 percentage points from the 1992 election win. The only real comparison is the 1918 general election where the Liberal party lost 31.2 percentage points and 236 seats.
If the Tories are reduced to under 100 seats then it could be hard for Labour to use the Tory boogeyman to keep those voters in line. It's not implausible that the 2029 GE would see no party win more than 30% of the vote. And then, with FPTP, anything could happen.
It seems to me this is right and captures the zeitgeist. The utterly distinctive feel of this election is Get Tories Smashed not Get Labour In - that was 1997.
Which is why, I think, Survation are about right. 75-80% of voters are ganging up in effect to ensure the Tories are decimated and both conscious and unconscious tactical voting will be enormous. In a sense I am an invisible example. I shall vote Labour. If I voted in one of the Tory/LD seats I would transfer to LD without thinking at all, and wouldn't even reckon I was voting tactically.
Still forecasting a 70 majority based on feelings.
Which defies all available empirical evidence.