politicalbetting.com » Blog Archive » The big MRP message for Tory remainers is that Corbyn can’t be

System

politicalbetting.com » Blog Archive » The big MRP message for Tory remainers is that Corbyn can’t become PM

Most of the focus on the two MRP sets of projections that we got last night is that in certain key seats tactical voting has the potential to impede the pro-Brexit Tories. Certainly in the bellwether seat of Bedford where I live both YouGov and the Best for Britain suggest that the Tories are just one percent ahead and it won’t take much to have an impact.

Read the full story here

Big_G_NorthWales

First

Big_G_NorthWales

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Jonathan

With Boris looking likely to win a majority, it is time to give his plans serious scrutiny and get some firm commitments on the record.

His promise to 'Get Brexit Done' needs to be interrogated. We need to know precisely how he will will avoid a cold, hard Brexit in 12 months. A promise is required. We cannot afford to give him a blank cheque.

Endillion

FPT

IanB2 said:

Endillion said:

IanB2 said:

Pulpstar said:

IanB2 said:

Since the model works by starting from a very small sample of local voters - as I recall fewer than a hundred - and then projects demographic-based voting behaviour based on regional and national polling onto the seat to make up the result - I would have thought a weakness of this approach is determining accurately the level of support for Indy’s like Field or the TIGs?

I don’t expect them to win, just wonder how accurate the MRP can be for such candidates?

Last time round the MRP severely overestimated Claire Wright's appeal over the Tories in East Devon.

You’re right - and I remember wondering then how the model could do that. Nothing in any of the descriptions I have read suggests how you can use demographics to work out whether people are going to vote for an Independent standing in one constituency.

+1
I wrote the model off for this reason - it just made it look like it hadn't been through proper review by people who actually understood electoral mechanics.
On the other hand, she did have just about the largest error bar of any prediction in any seat in the country... but, even so ...

Do any statisticians happen to know roughly what the margin of error would be in a poll of just 100 people? Ignoring for the moment that they're not randomly sampled.

First stab: assume random sampling of 50k people (with or without replacement; doesn't matter much). Assume it's a two horse race and true result is 50:50 tossup. Then use normal approximation to the binomial distribution: gives you N(50, 25^2). Which implies a 95% CI between 0% and 100%. Which feels about right, actually.
I'm sure someone with a better stats background will be along soon to explain where I've gone wrong.

HYUFD

https://twitter.com/GoodwinMJ/status/1199953572118638593?s=20

Nigelb

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out
There is only one way to do that and it is to vote conservative

Of course you're not. You are a Tory diehard (though not in the debased HYUFD usage of the term).

HYUFD

I do expect the LDs to do a bit better than the MRP, especially in central London and Remain areas, by squeezing the Labour vote. They might gain a few more Tory Remainers too provided the Tories hold a clear lead over Corbyn Labour

Sandpit

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

Jonathan

Sandpit said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

How will Boris avoid a hard Brexit in 12 months? What guarantees do we have?

eek

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

And watch us crash out of Europe on 1st January 2021...

As Jonathan says Johnson has made promises he can't keep which means either he is a liar, a fool or both.

Now I know that neither option is good but as Labour can't win a majority at least he would be controlled by other factors - Boris with a majority won't be.

LostPassword

The MRP result seems to be perfectly calibrated for the satisfaction of Labour moderates. The Labour defeat is heavy enough that Corbyn will be seen to have clearly failed, but it's not so heavy as to do mortal damage to the Labour Party, nor are the Lib Dems making the advances that would put them closer to challenging Labour as the principal party of Opposition.
All they have to do now is find a candidate who can appeal to the membership, and then work out how to defeat the Tories.

geoffw

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out
There is only one way to do that and it is to vote conservative

The point of recent pushes for tactical voting is to keep the Cons out.

Endillion

Endillion said:

FPT

IanB2 said:

Endillion said:

IanB2 said:

Pulpstar said:

IanB2 said:

Since the model works by starting from a very small sample of local voters - as I recall fewer than a hundred - and then projects demographic-based voting behaviour based on regional and national polling onto the seat to make up the result - I would have thought a weakness of this approach is determining accurately the level of support for Indy’s like Field or the TIGs?

I don’t expect them to win, just wonder how accurate the MRP can be for such candidates?

Last time round the MRP severely overestimated Claire Wright's appeal over the Tories in East Devon.

You’re right - and I remember wondering then how the model could do that. Nothing in any of the descriptions I have read suggests how you can use demographics to work out whether people are going to vote for an Independent standing in one constituency.

+1
I wrote the model off for this reason - it just made it look like it hadn't been through proper review by people who actually understood electoral mechanics.
On the other hand, she did have just about the largest error bar of any prediction in any seat in the country... but, even so ...

Do any statisticians happen to know roughly what the margin of error would be in a poll of just 100 people? Ignoring for the moment that they're not randomly sampled.

First stab: assume random sampling of 50k people (with or without replacement; doesn't matter much). Assume it's a two horse race and true result is 50:50 tossup. Then use normal approximation to the binomial distribution: gives you N(50, 25^2). Which implies a 95% CI between 0% and 100%. Which feels about right, actually.
I'm sure someone with a better stats background will be along soon to explain where I've gone wrong.

On reflection, I think this completely wrong, but I can't edit to change. Sorry

alex_

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

HYUFD

https://twitter.com/GoodwinMJ/status/1199972391830589441?s=20

HYUFD

https://twitter.com/tnewtondunn/status/1199972526081875968?s=20

alex_

Jonathan said:

With Boris looking likely to win a majority, it is time to give his plans serious scrutiny and get some firm commitments on the record.

His promise to 'Get Brexit Done' needs to be interrogated. We need to know precisely how he will will avoid a cold, hard Brexit in 12 months. A promise is required. We cannot afford to give him a blank cheque.

Wonder what happens if the European Parliament reject the Withdrawal agreement...?

Richard_Tyndall

The idea that Corbyn can't become PM is simply not true. It really wouldn't take that much of a swing to lose the Tory majority and at that point it is entirely possible that Labour, as the next largest party could get a coalition with a couple of simple promises - Independence vote for Scotland and Second EU referendum for the Lib Dems.

Obviously OGH wants to push the narrative that it is safe to vote anti-Tory without the risk of Corbyn but that is just spin. Corbyn remains a genuine risk (if you regard his Premiership as a risk) and the only way to prevent him is by voting Tory.

Big_G_NorthWales

Jonathan said:

With Boris looking likely to win a majority, it is time to give his plans serious scrutiny and get some firm commitments on the record.

His promise to 'Get Brexit Done' needs to be interrogated. We need to know precisely how he will will avoid a cold, hard Brexit in 12 months. A promise is required. We cannot afford to give him a blank cheque.

This election is not done yet and I accept your concern on a hard brexit but I doubt that will cut through

For me the key is the collapse of the lib dems due to the 'stupid' revoke policy poorly promoted by Jo Swinson. Labour are in trouble in leave areas and planting billions of trees is not going to move the dial as it is recognised leave voting areas are not impressed by climate change policies

The polls have not had time to reflect the dreadful last two days for Corbyn and even this morning when I switched on the tv Barry Gardiner was still trying to defend Corbyn's non apology

There are only 13 days left to turn the dial, tempus fugit, and it is running out rapidly

Sandpit

LostPassword said:

The MRP result seems to be perfectly calibrated for the satisfaction of Labour moderates. The Labour defeat is heavy enough that Corbyn will be seen to have clearly failed, but it's not so heavy as to do mortal damage to the Labour Party, nor are the Lib Dems making the advances that would put them closer to challenging Labour as the principal party of Opposition.
All they have to do now is find a candidate who can appeal to the membership, and then work out how to defeat the Tories.

They need to find a candidate who has wide appeal within the country, as well as within the party. Who’s the next Tony Blair - the only Labour leader to win a majority since Wilson squeaked one 45 years ago?

MikeSmithson

Richard_Tyndall said:

The idea that Corbyn can't become PM is simply not true. It really wouldn't take that much of a swing to lose the Tory majority and at that point it is entirely possible that Labour, as the next largest party could get a coalition with a couple of simple promises - Independence vote for Scotland and Second EU referendum for the Lib Dems.

Obviously OGH wants to push the narrative that it is safe to vote anti-Tory without the risk of Corbyn but that is just spin. Corbyn remains a genuine risk (if you regard his Premiership as a risk) and the only way to prevent him is by voting Tory.

So Prof John Curtice is wrong.

Sandpit

Ah, and I see Vanilla has had another “upgrade” in the last couple of days. Can someone please tell them to stop it?

Endillion

Endillion said:

FPT

IanB2 said:

Endillion said:

IanB2 said:

Pulpstar said:

IanB2 said:

Since the model works by starting from a very small sample of local voters - as I recall fewer than a hundred - and then projects demographic-based voting behaviour based on regional and national polling onto the seat to make up the result - I would have thought a weakness of this approach is determining accurately the level of support for Indy’s like Field or the TIGs?

I don’t expect them to win, just wonder how accurate the MRP can be for such candidates?

Last time round the MRP severely overestimated Claire Wright's appeal over the Tories in East Devon.

You’re right - and I remember wondering then how the model could do that. Nothing in any of the descriptions I have read suggests how you can use demographics to work out whether people are going to vote for an Independent standing in one constituency.

+1
I wrote the model off for this reason - it just made it look like it hadn't been through proper review by people who actually understood electoral mechanics.
On the other hand, she did have just about the largest error bar of any prediction in any seat in the country... but, even so ...

Do any statisticians happen to know roughly what the margin of error would be in a poll of just 100 people? Ignoring for the moment that they're not randomly sampled.

First stab: assume random sampling of 50k people (with or without replacement; doesn't matter much). Assume it's a two horse race and true result is 50:50 tossup. Then use normal approximation to the binomial distribution: gives you N(50, 25^2). Which implies a 95% CI between 0% and 100%. Which feels about right, actually.
I'm sure someone with a better stats background will be along soon to explain where I've gone wrong.

Where I went wrong is to confuse standard deviation with variance. It's actually N(50, 25), so SD = 5 and CI is about 10 points either side of central estimate. Load for a bit to allow for the simplifications I've made - probably looking at an error bar about 30 percentage points wide, at a guess.

MikeSmithson

alex_ said:

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

Agree 100%

geoffw

Voter complacency in the wake of favourable polls for the Cons could easily result in another hung parliament. Cummings shows acute awareness of this in his blog letter.

Big_G_NorthWales

Nigelb said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out
There is only one way to do that and it is to vote conservative

Of course you're not. You are a Tory diehard (though not in the debased HYUFD usage of the term).

I do not like diehard as a term and have called out HYUFD on many occasions over its use

I could be persuaded to vote otherwise in some circumstances but Corbyn has to be soundly defeated and gone by Xmas would be my best Xmas present

Richard_Tyndall

MikeSmithson said:

Richard_Tyndall said:

The idea that Corbyn can't become PM is simply not true. It really wouldn't take that much of a swing to lose the Tory majority and at that point it is entirely possible that Labour, as the next largest party could get a coalition with a couple of simple promises - Independence vote for Scotland and Second EU referendum for the Lib Dems.

Obviously OGH wants to push the narrative that it is safe to vote anti-Tory without the risk of Corbyn but that is just spin. Corbyn remains a genuine risk (if you regard his Premiership as a risk) and the only way to prevent him is by voting Tory.

So Prof John Curtice is wrong.

Quite clearly yes. And that is a terrible appeal to authority in the face of basic logic. The Comres poll yesterday has confirmed that we are very close to a Hung Parliament. Yougov themselves have shown that a Tory lead of less than between 4 and 6% will mean no majority. And once you have that, whatever John Curtice might think is immaterial because it all comes down to the choices of two people - Nicola Sturgeon and Jo Swinson - and how willing they are to take whatever bribes Corbyn can offer. And the professor certainly can't predict how that will go.

Of course you want to stop a Tory majority so will push this line but it is thoroughly misleading.

camel

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Sandpit

geoffw said:

Voter complacency in the wake of favourable polls for the Cons could easily result in another hung parliament. Cummings shows acute awareness of this in his blog letter.

Love him or loathe him, Dominic Cummings is incredibly astute in his observations. The reason he’s so polarising is that his political opponents know he’s incredibly good at his job.
For previous examples, see Andy Coulson, Damien McBride, Alastair Campbell, Willie Whitelaw...

Big_G_NorthWales

Jonathan said:

Sandpit said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

How will Boris avoid a hard Brexit in 12 months? What guarantees do we have?

None I am afraid but with a good majority I expect Boris to pivot to get a good deal

camel

HYUFD said:

https://twitter.com/GoodwinMJ/status/1199972391830589441?s=20

I agree that the tory seats will be more widely spread across this 'one nation'. However I will stake my trousers that Wakefield will not turn blue.

alex_

IanB2 said:

Chris said:

IanB2 said:

Endillion said:

IanB2 said:

Pulpstar said:

IanB2 said:

NickPalmer said:

Dadge said:

shiney2 said:

MRP & Don Valley..
We are given to believe es didn't stand..
UGov are having a giraffe.

Flint will lose.

The pr.

This is a familiar pattern, which we also saw w
it's early days for him.

https://www.realclearpolitics.com/epolls/latest_polls/

Since the model works by starting from a very small sample of local voters - as I recall fewer than a hundred - and then projects demographic-based voting behaviour based on regional and national polling onto the seat to make up the result - I would have thought a weakness of this approach is determining accurately the level of support for Indy’s like Field or the TIGs?

I don’t expect them to win, just wonder how accurate the MRP can be for such candidates?

Last time round the MRP severely overestimated Claire Wright's appeal over the Tories in East Devon.

You’re right - and I remember wondering then how the model could do that. Nothing in any of the descriptions I have read suggests how you can use demographics to work out whether people are going to vote for an Independent standing in one constituency.

+1
I wrote the model off for this reason - it just made it look like it hadn't been through proper review by people who actually understood electoral mechanics.
On the other hand, she did have just about the largest error bar of any prediction in any seat in the country... but, even so ...

Do any statisticians happen to know roughly what the margin of error would be in a poll of just 100 people? Ignoring for the moment that they're not randomly sampled.

It should be inversely proportional to the square root of the sample size.

So for a 100 the margin of error would be 3 times as much as for 900, or 20 times as much as for 40,000.

Thank you kindly. The MOE for a 1000-1200 poll is usually put at around +/- 2.5%. So we're looking at an MOE of somewhere in the region of 7-10% for YouGov's mini-samples, which I guess isn't that bad of a start point for assessing how well the various Indys might do?

This is a bit misleading though isn’t it, because yougov’s projections are not based solely on the people sampled in each individual constituency. They are a combination of individual constituency polling, combined with wider demographic data from elsewhere combined with matching up constituencies with similar characteristics. It’s more of an exit poll approach.

Richard_Tyndall

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Yes but the trick is you then plant a new one. If, once you have your 2 billion trees you are continually planting new trees in rotation you keep the carbon locked up.

nunu2

https://mobile.twitter.com/326Pols/status/1199824621685682177

Is this correct? Are there any others?

YBarddCwsc

alex_ said:

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

So, it is back to Eric Joyce's fists again.

The election of Corbyn was absolutely key to the Brexit referendum result, agreed (and it is not acknowledged enough the role that Corbyn played, far more important than "Russian interference").

But, Corbyn himself happened because of the huge disillusionment in New Labour following the financial crisis & the Iraq War. Corbyn was inevitable.

GIN1138

LOL @ Labour planning to target their leave voting areas after spending the past two years telling them they are nothing but racist scum and then throwing them under the bus.

Good luck with that...

Morris_Dancer

Mr. Sandpit, there was another absolutely bloody brilliant reason McBride wasn't liked.

Andy_JS

Two interesting spreadsheets: the change in share for both Con and Lab in the YouGov study compared to GE2017.

https://docs.google.com/spreadsheets/d/1HYvNvaq3xaLjItv6PH0zFxoh3cwec3sL5WB5tByKbsY/edit#gid=0
https://docs.google.com/spreadsheets/d/1Sbmz9OAwHUz8fuFN2S9s011tDQw_RXmr4W1wjUb6bnw/edit#gid=0

camel

Richard_Tyndall said:

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Yes but the trick is you then plant a new one. If, once you have your 2 billion trees you are continually planting new trees in rotation you keep the carbon locked up.

I imagined it was better than that - like burying the trunks or something. I'm actually disappointed.

alex_

YBarddCwsc said:

alex_ said:

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

So, it is back to Eric Joyce's fists again.

The election of Corbyn was absolutely key to the Brexit referendum result, agreed (and it is not acknowledged enough the role that Corbyn played, far more important than "Russian interference").

But, Corbyn himself happened because of the huge disillusionment in New Labour following the financial crisis & the Iraq War. Corbyn was inevitable.

A moderate left wing platform growing in attractiveness as austerity took hold was inevitable. Corbyn was not. All it needed was to keep him off the ballot paper.

eek

YBarddCwsc said:

alex_ said:

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

So, it is back to Eric Joyce's fists again.

The election of Corbyn was absolutely key to the Brexit referendum result, agreed (and it is not acknowledged enough the role that Corbyn played, far more important than "Russian interference").

But, Corbyn himself happened because of the huge disillusionment in New Labour following the financial crisis & the Iraq War. Corbyn was inevitable.

Only because Labour was both stupid enough to put him on the paper while at the same time allowing anyone with £3 to vote.

Yes you can trace everything back to Corbyn being elected but for that you need to blame Ed Miliband and Margaret Beckett.

Stocky

Just got 6/5 Cambridge for Labour.
MRP poll has as clear Labour.
Thoughts?

SandyRentool

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Depends what happens to the tree at end of life. If it is used as building material, the CO2 remains locked in. If it is burnt then the CO2 returns to the atmosphere. If it is burnt with CCS, then there is a transfer of CO2 from the atmosphere to underground storage, so the overall tree growing and burning process becomes net negative. See "BECCS" - bioenergy with carbon capture and storage.

Second point, if the overall level of tree coverage is increased, then this will hold more CO2 in a dynamic steady state (!) than with lower tree coverage, so less CO2 in the atmosphere.

Big_G_NorthWales

MikeSmithson said:

alex_ said:

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

Agree 100%

And me

Jonathan

Big_G_NorthWales said:

Jonathan said:

Sandpit said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

How will Boris avoid a hard Brexit in 12 months? What guarantees do we have?

None I am afraid but with a good majority I expect Boris to pivot to get a good deal

Not good enough.

Barnesian

The MRP modelling technique is based on sampling the voting preferences of a large number of demographic groups and then applying the result to the demographics of each constituency. It is a model.
My model does not rely on demographics at all - not even remain/leave of each seat. It assumes that the demographics of each seat are "baked in" to the 2017 actual result including the relative strength of the activists base, council representation, history etc that I don't think the MRP model reflects. It then applies assumptions about tactical voting. It is a model.
The appeal of the MRP model is that it was close last time. But that is a sample of one. I think it was also used in the US so a sample of two.
Imagine a heavily promoted modelling technique (with whizzy multilevel Regression and Post-stratification) that correctly picked the 20/1 winner of last year's Grand National. Would its prediction for this year influence the betting? You bet it would - whether it was a sound model or not! "The computer says"
For this reason I'm betting against the MRP which has moved certain markets e.g LD over 25.5 seats at 3.35 looks good value. It shot up after the MRP was published.
EDIT. If my model proves to be spot on this time, you shouldn't assume it is a good model. It might just be lucky. But I suspect it would be followed with close attention next time!

DecrepiterJohnL

Sandpit said:

geoffw said:

Voter complacency in the wake of favourable polls for the Cons could easily result in another hung parliament. Cummings shows acute awareness of this in his blog letter.

Love him or loathe him, Dominic Cummings is incredibly astute in his observations. The reason he’s so polarising is that his political opponents know he’s incredibly good at his job.
For previous examples, see Andy Coulson, Damien McBride, Alastair Campbell, Willie Whitelaw...

That is not quite right. Dominic Cummings alienated his erstwhile allies before his opponents. Andy Coulson, whom you also mention, blocked Cummings' recruitment before Cameron acquiesced.

Scrapheap_as_was

As a Tory remainer I''m not persuaded that Corbyn is safely barred from No. 10 just yet....

Barnesian

Stocky said:

Just got 6/5 Cambridge for Labour.
MRP poll has as clear Labour.
Thoughts?

Not a value bet

EDIT But not a bad bet either. I think it is about right. I have Cambridge 44% LD, 41% Lab.

OnlyLivingBoy

camel said:

Richard_Tyndall said:

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Yes but the trick is you then plant a new one. If, once you have your 2 billion trees you are continually planting new trees in rotation you keep the carbon locked up.

I imagined it was better than that - like burying the trunks or something. I'm actually disappointed.

Does it release CO2 when it decomposes? If so, where did oil and coal come from? Isn't that just dead trees and animals that have been buried under successive layers of the earth's surface? Anyway, as others have said, the key is to keep replacing the trees over time.

YBarddCwsc

alex_ said:

YBarddCwsc said:

alex_ said:

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

So, it is back to Eric Joyce's fists again.

The election of Corbyn was absolutely key to the Brexit referendum result, agreed (and it is not acknowledged enough the role that Corbyn played, far more important than "Russian interference").

But, Corbyn himself happened because of the huge disillusionment in New Labour following the financial crisis & the Iraq War. Corbyn was inevitable.

A moderate left wing platform growing in attractiveness as austerity took hold was inevitable. Corbyn was not. All it needed was to keep him off the ballot paper.

There were moderate left candidates on the ballot paper. Why did they not beat Corbyn?

The answers really come down to the fact that inequality has risen so much in the UK that someone like Corbyn is a very attractive figure to many. And the EU has been one of the main drivers of inequality, with places like Cambridge awash with EU cash, and places like Blaenau Gwent with hardly any.

NorthofStoke

MikeSmithson said:

alex_ said:

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

Agree 100%

I also agree 100%. The Sarajavo pistol shot was the election of Ed rather than David by the narrowest of margins. There had to be a lot of mistakes after that to be sure but that vote has had a catastrophic effect.

Verulamius

A new forest has been planted just north of St Albans over the past 10 years.

https://heartwood.woodlandtrust.org.uk/

600,000 new trees.

Morris_Dancer

It is interesting to consider the counter-factual.
What if Corbyn had never made the shortlist?
We'd likely have Cooper or Burnham as Labour leader. Neither titans, but neither far left nutcases either. Both pro-EU. Likely would've helped win the pro-EU side the referendum.
Osborne would probably be PM. Jews would feel more relaxed about living here.
Would there be a push for greater EU integration? Would UKIP be riding high in the polls?
Hard to say. But the far left squatting on Labour's front bench would've been avoided, and that would be a very good thing.
Edited for para-spacing.

noneoftheabove

Richard_Tyndall said:

The idea that Corbyn can't become PM is simply not true. It really wouldn't take that much of a swing to lose the Tory majority and at that point it is entirely possible that Labour, as the next largest party could get a coalition with a couple of simple promises - Independence vote for Scotland and Second EU referendum for the Lib Dems.

Obviously OGH wants to push the narrative that it is safe to vote anti-Tory without the risk of Corbyn but that is just spin. Corbyn remains a genuine risk (if you regard his Premiership as a risk) and the only way to prevent him is by voting Tory.

You are right on the numbers, a hung parliament is perfectly possible still, and Corbyn is likely to be PM in a lot of the hung parliament scenarios.
However, I think Mike is right on the sentiment, the risk from the Corbyn PM is now far lower than it was pre-election as the scenarios of Labour doing well enough to give him real power have virtually disappeared. At best (for him) he will be able to govern for six months to deliver a referendum and then call another GE, which is presumably in line with Tory remainers wishes. In that six months we are not going to be nationalising anything or bunging billions to the waspis.

Big_G_NorthWales

Jonathan said:

Big_G_NorthWales said:

Jonathan said:

Sandpit said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

How will Boris avoid a hard Brexit in 12 months? What guarantees do we have?

None I am afraid but with a good majority I expect Boris to pivot to get a good deal

Not good enough.

With respect nothing but remaining is good enough for you

OnlyLivingBoy

Jonathan said:

Big_G_NorthWales said:

Jonathan said:

Sandpit said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

How will Boris avoid a hard Brexit in 12 months? What guarantees do we have?

None I am afraid but with a good majority I expect Boris to pivot to get a good deal

Not good enough.

He is at least as likely to use a big majority to deliver a 'proper Brexit' quickly and then use the next four years to get over the economic hit that will follow. Does he really want to still be bogged down in tricky FTA negotiations going into the next election? And how big will the ERG be in the next parliament? I don't imagine the new tranche of Tory MPs in leave voting areas will be a new swathe of Dominic Grieves.

Dura_Ace

eek said:

Now I know that neither option is good but as Labour can't win a majority at least he would be controlled by other factors - Boris with a majority won't be.

Corbo constrained by Sturgeon and Swinson is a far less dangerous proposition for the UK than Full Metal Boris.

But we know it's party before country for the tories.

Every single time.

OnlyLivingBoy

And you can vote Labour in Bedford, Mike. 😉

Barnesian

Scrapheap_as_was said:

As a Tory remainer I''m not persuaded that Corbyn is safely barred from No. 10 just yet....

There is a difference between Corbyn in number 10 enacting his manifesto and Corbyn in number 10 only being allowed to arrange a second referendum before being VONCed - which is a much more likely scenario.

noneoftheabove

alex_ said:

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

For me it started with QE becoming the norm, separating the country into asset owners supported by the govt and wage earners footing the bill. That has driven Corbynism, and in turn Brexit and Bluekipisation of the Tory party.

Big_G_NorthWales

Dura_Ace said:

eek said:

Now I know that neither option is good but as Labour can't win a majority at least he would be controlled by other factors - Boris with a majority won't be.

Corbo constrained by Sturgeon and Swinson is a far less dangerous proposition for the UK than Full Metal Boris.

But we know it's party before country for the tories.

Every single time.

Not sure this makes sense

RochdalePioneers

MRP doesn't match reports that I am hearing from the ground. That bit was right last time doesn't mean it's right this time as it's a very different election.
However, it's impact will be real. The parties will react to it and that will make changes.
Labour seem to want to target their leavers. So a more explicit pro-leave message, reinforcing the obvious reality that a Labour government would back a Labour negotiated leave deal*. This combined with Tories feeling more comfortable that Corbyn wont win means less Tory one nation remainers clinging to a party they no longer recognise. Thus fragmenting the growing squeeze.
*Labour's plans to steal vast billions from pension funds by nationalizing companies at a price they set rather than the market price is screamingly illegal under EU law. As I've been saying for ages Corbyn is a leaver so nobody can stop him renationalising National Carriers and BOAC. This latest confirmation that the EU would tie us up in court just reinforces it

Andy_JS

Morris_Dancer said:

It is interesting to consider the counter-factual.
What if Corbyn had never made the shortlist?
We'd likely have Cooper or Burnham as Labour leader. Neither titans, but neither far left nutcases either. Both pro-EU. Likely would've helped win the pro-EU side the referendum.
Osborne would probably be PM. Jews would feel more relaxed about living here.
Would there be a push for greater EU integration? Would UKIP be riding high in the polls?
Hard to say. But the far left squatting on Labour's front bench would've been avoided, and that would be a very good thing.
Edited for para-spacing.

One could plausibly blame the internet for all of that, because Corbyn's narrow but enthusiastic fan-base probably wouldn't have been to gain the traction it did without the influence of social media.

IshmaelZ

OnlyLivingBoy said:

camel said:

Richard_Tyndall said:

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Yes but the trick is you then plant a new one. If, once you have your 2 billion trees you are continually planting new trees in rotation you keep the carbon locked up.

I imagined it was better than that - like burying the trunks or something. I'm actually disappointed.

Does it release CO2 when it decomposes? If so, where did oil and coal come from? Isn't that just dead trees and animals that have been buried under successive layers of the earth's surface? Anyway, as others have said, the key is to keep replacing the trees over time.

Yes it does. Coal is wood which hasn't aerobically decomposed. I thought until 5 minutes ago this was because it was deprived of air by water, but rivetingly wikipedia says it's because there was a gap between trees learning how to make wood, and microbes evolving to be able to decompose it.

Morris_Dancer

If the Conservatives do get a majority, I hope Labour MPs grow some balls and either axe Corbyn or walk out en masse, instead of re-enacting Watson's wibbling. Now the critics of Corbyn are campaigning to make the unacceptable PM.

Endillion

Stocky said:

Just got 6/5 Cambridge for Labour.
MRP poll has as clear Labour.
Thoughts?

AFAIK city is trending LD again and students will have long since gone home following end of term. So I'm backing LDs. No idea how/if yougov deal with that kind of stuff but my guess is it's tricky.

Andy_JS

Verulamius said:

A new forest has been planted just north of St Albans over the past 10 years.

https://heartwood.woodlandtrust.org.uk/

600,000 new trees.

The search engine Ecosia plants new trees every time you use it.

https://www.ecosia.org/

IanB2

FPT

Barnesian said:

The MRP modelling technique is based on sampling the voting preferences of a large number of demographic groups and then applying the result to the demographics of each constituency. It is a model.
My model does not rely on demographics at all - not even remain/leave of each seat. It assumes that the demographics of each seat are "baked in" to the 2017 actual result including the relative strength of the activists base, council representation, history etc that I don't think the MRP model reflects. It then applies assumptions about tactical voting. It is a model.
The appeal of the MRP model is that it was close last time. But that is a sample of one. I think it was also used in the US so a sample of two.
Imagine a heavily promoted modelling technique (with whizzy multilevel Regression and Post-stratification) that correctly picked the 20/1 winner of last year's Grand National. Would its prediction for this year influence the betting? You bet it would - whether it was a sound model or not! "The computer says"
For this reason I'm betting against the MRP which has moved certain markets e.g LD over 25.5 seats at 3.35 looks good value. It shot up after the MRP was published.
EDIT. If my model proves to be spot on this time, you shouldn't assume it is a good model. It might just be lucky. But I suspect it would be followed with close attention next time!

MRP therefore wins over a 'static' model if the change in voting behaviour is driven by demographics - as has been the case in recent years. The MRP also starts with the data from its panellists (around 100?) in each seat, rather than starting with a blank piece of paper.

Where you have a point is that the MRP will itself affect the result. Last time it came late and few people believed it. This time it comes early and, as you say, with a lot of credibility.

nunu2

Libdems seem to be getting really close in some southern shires, perhaps BXP not standing there has saved a lot of tory seats.

DecrepiterJohnL

NorthofStoke said:

MikeSmithson said:

alex_ said:

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

Agree 100%

I also agree 100%. The Sarajavo pistol shot was the election of Ed rather than David by the narrowest of margins. There had to be a lot of mistakes after that to be sure but that vote has had a catastrophic effect.

Wrong party. David Cameron was the worst prime minister since Lord North, and is largely responsible for the current mess vis-a-vis Europe, the economy, and the dislocation of Britain, so we need to trace Cameron's rise to power. Perhaps the Sarajevo assassination was that William Hague resigned too quickly, and should have stayed on to fight another election.

humbugger

Good morning all.

I see from their website that the Daily Mail has now cottoned on to the fact that Labour's dividend tax rises will really hit lots of small businesses very hard. This is another example of Labour totally misleading voters with their claim that only the top 5% would pay more tax.

Corbyn is becoming estranged from the truth. Yesterday's ludicrous claims about the NHS and today's fantasy plan to plan zillions of trees suggest that panic is beginning to set in and Labour's response seems to be even more outlandish claims. Whatever next?

Corbyn and McDonnell are absolutely ruthless.

camel

OnlyLivingBoy said:

camel said:

Richard_Tyndall said:

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Yes but the trick is you then plant a new one. If, once you have your 2 billion trees you are continually planting new trees in rotation you keep the carbon locked up.

I imagined it was better than that - like burying the trunks or something. I'm actually disappointed.

Does it release CO2 when it decomposes? If so, where did oil and coal come from? Isn't that just dead trees and animals that have been buried under successive layers of the earth's surface? Anyway, as others have said, the key is to keep replacing the trees over time.

I think coal and oil was formed in anaerobic conditions and, in the case of coal, require something quite literally seismic to form. If dead trees are left on the surface they will return their carbon to the atmosphere. What I've learned today is that you can store x amount of carbon per hectare. Once you've run out of hectares, you've run out of potential. And one thing we know about the UK is that hectares are expensive - at the last auctions I attended farmland here was selling at £45k/hectare. That's without the opportunity cost of not growing food on it.
On the plus side, after 37 years wondering, we may have found a genuine use for the Falkland Islands where land cost 1000x less.

HYUFD

LostPassword said:

The MRP result seems to be perfectly calibrated for the satisfaction of Labour moderates. The Labour defeat is heavy enough that Corbyn will be seen to have clearly failed, but it's not so heavy as to do mortal damage to the Labour Party, nor are the Lib Dems making the advances that would put them closer to challenging Labour as the principal party of Opposition.
All they have to do now is find a candidate who can appeal to the membership, and then work out how to defeat the Tories.

True if Labour lose and replace Corbyn with a moderate. If after a second heavier Corbyn defeat Labour pick another hard left leader the LDs have their chance, especially if ex Labour Chuka wins Cities of London and Westminster and succeeds Swinson as LD leader

isam

MikeSmithson said:

alex_ said:

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

Agree 100%

Blame the virtue signalling centrists who put him on the ballot to look fair minded and have been gnashing their teeth ever since

IanB2

eek said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

And watch us crash out of Europe on 1st January 2021...

As Jonathan says Johnson has made promises he can't keep which means either he is a liar, a fool or both.

Now I know that neither option is good but as Labour can't win a majority at least he would be controlled by other factors - Boris with a majority won't be.

As yesterday's Newsnight reminded us, he also claimed (on the steps of no. 10, no less) to have an already worked up plan to solve the Social Care crisis. One which they now say doesn't exist.

Peter_the_Punter

Richard_Tyndall said:

The idea that Corbyn can't become PM is simply not true. It really wouldn't take that much of a swing to lose the Tory majority and at that point it is entirely possible that Labour, as the next largest party could get a coalition with a couple of simple promises - Independence vote for Scotland and Second EU referendum for the Lib Dems.

Obviously OGH wants to push the narrative that it is safe to vote anti-Tory without the risk of Corbyn but that is just spin. Corbyn remains a genuine risk (if you regard his Premiership as a risk) and the only way to prevent him is by voting Tory.

Oh I regard a Corbyn Premiership as a risk, Richard, but I think OGH has it more right than you this time.

The only way Corbyn can realistically become PM is with an Overall Majority. That's 33/1 against as I write and I'd be a layer, not a backer, even at that price.

He might just about manage it, you might think, with a little help from his friends if he comes up a few short of 325 seats. But he ain't got many friends. The price for them supporting a Labour Government would almost certainly be his removal and replacement with a much more acceptable kind of Leader. It would also imply a much more centreist Labour government in Office.

You can see the appeal for Leavers like me. I can see it too for Remain Tories. Boris would go, the traditional Tory Party would resume control, the Brexiteers who hijacked the Party would be vanquished and we might even have a second look at the whole question of revisiting the referendum in the light of what we have learned since.

Seems a very positive way to use one's vote to me.

Xtrain

Richard_Tyndall said:

MikeSmithson said:

Richard_Tyndall said:

The idea that Corbyn can't become PM is simply not true. It really wouldn't take that much of a swing to lose the Tory majority and at that point it is entirely possible that Labour, as the next largest party could get a coalition with a couple of simple promises - Independence vote for Scotland and Second EU referendum for the Lib Dems.

Obviously OGH wants to push the narrative that it is safe to vote anti-Tory without the risk of Corbyn but that is just spin. Corbyn remains a genuine risk (if you regard his Premiership as a risk) and the only way to prevent him is by voting Tory.

So Prof John Curtice is wrong.

Quite clearly yes. And that is a terrible appeal to authority in the face of basic logic. The Comres poll yesterday has confirmed that we are very close to a Hung Parliament. Yougov themselves have shown that a Tory lead of less than between 4 and 6% will mean no majority. And once you have that, whatever John Curtice might think is immaterial because it all comes down to the choices of two people - Nicola Sturgeon and Jo Swinson - and how willing they are to take whatever bribes Corbyn can offer. And the professor certainly can't predict how that will go.

Of course you want to stop a Tory majority so will push this line but it is thoroughly misleading.

I agree.
There is no room for complacency.
Corbyn only needs NOM.
There are still 2 weeks to go.

Andy_JS

Sandpit said:

Ah, and I see Vanilla has had another “upgrade” in the last couple of days. Can someone please tell them to stop it?

Obviously an upgrade by someone who doesn't actually use the service very much.

Pulpstar

Hold up, this reminds me of @Casino_Royale message right before the referendum that it was safe to vote leave because they couldn't win in order to close the gap and not give us a mandate to join the Euro or some such.

IanB2

Endillion said:

Endillion said:

FPT

IanB2 said:

Endillion said:

IanB2 said:

Pulpstar said:

IanB2 said:

Since the model works by starting from a very small sample of local voters - as I recall fewer than a hundred - and then projects demographic-based voting behaviour based on regional and national polling onto the seat to make up the result - I would have thought a weakness of this approach is determining accurately the level of support for Indy’s like Field or the TIGs?

I don’t expect them to win, just wonder how accurate the MRP can be for such candidates?

Last time round the MRP severely overestimated Claire Wright's appeal over the Tories in East Devon.

You’re right - and I remember wondering then how the model could do that. Nothing in any of the descriptions I have read suggests how you can use demographics to work out whether people are going to vote for an Independent standing in one constituency.

+1
I wrote the model off for this reason - it just made it look like it hadn't been through proper review by people who actually understood electoral mechanics.
On the other hand, she did have just about the largest error bar of any prediction in any seat in the country... but, even so ...

Do any statisticians happen to know roughly what the margin of error would be in a poll of just 100 people? Ignoring for the moment that they're not randomly sampled.

First stab: assume random sampling of 50k people (with or without replacement; doesn't matter much). Assume it's a two horse race and true result is 50:50 tossup. Then use normal approximation to the binomial distribution: gives you N(50, 25^2). Which implies a 95% CI between 0% and 100%. Which feels about right, actually.
I'm sure someone with a better stats background will be along soon to explain where I've gone wrong.

On reflection, I think this completely wrong, but I can't edit to change. Sorry

Its main weakness appears to be the absence of an answer.

geoffw

Scrapheap_as_was said:

As a Tory remainer I''m not persuaded that Corbyn is safely barred from No. 10 just yet....

Well it's up to remainers like you to choose your poison.

Andy_JS

nunu2 said:

Libdems seem to be getting really close in some southern shires, perhaps BXP not standing there has saved a lot of tory seats.

LDs below 10% in 316 seats is pretty poor.

nunu2

What if indeed....

https://mobile.twitter.com/hugorifkind/status/1199970855394119683

Casino_Royale

FPT - the Conservatives have created a large polarisation in their voting coalition because their economics and cultural values work for older people, and don't for younger people.

If the Conservatives want the votes of younger people they need to not saddle them with so much debt, make buying houses easier/renting cheaper and show that whilst they are patriotic they are also relaxed about multi-levelled identities.

I don't climate change really features here, actually, as the Tories do actually have a good practical set of policies in place that are both deliverable and realistic.

Sandpit

Jonathan said:

Sandpit said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

How will Boris avoid a hard Brexit in 12 months? What guarantees do we have?

The main difference between the withdrawal negotiations and the trade negotiations, is that the latter are much less political and adversarial in nature. Those on the EU side of these negotiations will want to protect trade flows above all else, rather than to make a political point in bad faith as they did with the Withdrawal Agreement.
We already have a pretty good idea of where we end up, thanks to other agreements the EU has already signed with Canada and Japan. Trade in goods will be pretty much the same as those two agreements, with hopefully an agreement on access to services if the EU can accept regulatory equivalence rather than alignment. Once the EU agreement is concluded we can look to joining other trade groups such as the TPP.

Jonathan

Big_G_NorthWales said:

Jonathan said:

Big_G_NorthWales said:

Jonathan said:

Sandpit said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

Exactly. Those who would prop up a Corbyn government are going to have to agree his and McDonnell’s budget, not to mention all the things that a PM can do unilaterally without recourse to Parliament.
For all Boris’ faults, he’s not leading an institutionally racist party and doesn’t want to take Britain back to the 1970s. His party will also not hesitate to knife him in the front, if he becomes a liability.

How will Boris avoid a hard Brexit in 12 months? What guarantees do we have?

None I am afraid but with a good majority I expect Boris to pivot to get a good deal

Not good enough.

With respect nothing but remaining is good enough for you

That’s not true. Perfectly willing to accept EFTA. I just can’t take on a pure faith Boris avoiding a hard Brexit and neither should you.

IanB2

Endillion said:

Endillion said:

FPT

IanB2 said:

Endillion said:

IanB2 said:

Pulpstar said:

IanB2 said:

Since the model works by starting from a very small sample of local voters - as I recall fewer than a hundred - and then projects demographic-based voting behaviour based on regional and national polling onto the seat to make up the result - I would have thought a weakness of this approach is determining accurately the level of support for Indy’s like Field or the TIGs?

I don’t expect them to win, just wonder how accurate the MRP can be for such candidates?

Last time round the MRP severely overestimated Claire Wright's appeal over the Tories in East Devon.

You’re right - and I remember wondering then how the model could do that. Nothing in any of the descriptions I have read suggests how you can use demographics to work out whether people are going to vote for an Independent standing in one constituency.

+1
I wrote the model off for this reason - it just made it look like it hadn't been through proper review by people who actually understood electoral mechanics.
On the other hand, she did have just about the largest error bar of any prediction in any seat in the country... but, even so ...

Do any statisticians happen to know roughly what the margin of error would be in a poll of just 100 people? Ignoring for the moment that they're not randomly sampled.

First stab: assume random sampling of 50k people (with or without replacement; doesn't matter much). Assume it's a two horse race and true result is 50:50 tossup. Then use normal approximation to the binomial distribution: gives you N(50, 25^2). Which implies a 95% CI between 0% and 100%. Which feels about right, actually.
I'm sure someone with a better stats background will be along soon to explain where I've gone wrong.

Where I went wrong is to confuse standard deviation with variance. It's actually N(50, 25), so SD = 5 and CI is about 10 points either side of central estimate. Load for a bit to allow for the simplifications I've made - probably looking at an error bar about 30 percentage points wide, at a guess.

Lesson one in statistics is that the population size is irrelevant to the accuracy of a sample. So any sums working from the population size have got off on the wrong foot.

148grss

Richard_Tyndall said:

MikeSmithson said:

Richard_Tyndall said:

The idea that Corbyn can't become PM is simply not true. It really wouldn't take that much of a swing to lose the Tory majority and at that point it is entirely possible that Labour, as the next largest party could get a coalition with a couple of simple promises - Independence vote for Scotland and Second EU referendum for the Lib Dems.
Obviously OGH wants to push the narrative that it is safe to vote anti-Tory without the risk of Corbyn but that is just spin. Corbyn remains a genuine risk (if you regard his Premiership as a risk) and the only way to prevent him is by voting Tory.

So Prof John Curtice is wrong.

Quite clearly yes. And that is a terrible appeal to authority in the face of basic logic. The Comres poll yesterday has confirmed that we are very close to a Hung Parliament. Yougov themselves have shown that a Tory lead of less than between 4 and 6% will mean no majority. And once you have that, whatever John Curtice might think is immaterial because it all comes down to the choices of two people - Nicola Sturgeon and Jo Swinson - and how willing they are to take whatever bribes Corbyn can offer. And the professor certainly can't predict how that will go.
Of course you want to stop a Tory majority so will push this line but it is thoroughly misleading.

Not to diminish the powers of OGH and his posts here, but do his suggestions have such an impact on the electorate?

Casino_Royale

Jonathan said:

With Boris looking likely to win a majority, it is time to give his plans serious scrutiny and get some firm commitments on the record.

His promise to 'Get Brexit Done' needs to be interrogated. We need to know precisely how he will will avoid a cold, hard Brexit in 12 months. A promise is required. We cannot afford to give him a blank cheque.

We won't get that.

But, we know he's a snake and will throw anyone under a bus to save himself.

In my view, that means he will ratfuck the ERG to get a good, close relationship with the EU, and build on that with a nominal "thin" US one, and similar with Oz, Nz, Japan and Canada.

I don't think he's a No Dealer, but he will be boxed in if he only gets a majority of ten - he really needs 60-80.

eek

IanB2 said:

eek said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out

There is only one way to do that and it is to vote conservative

And watch us crash out of Europe on 1st January 2021...

As Jonathan says Johnson has made promises he can't keep which means either he is a liar, a fool or both.

Now I know that neither option is good but as Labour can't win a majority at least he would be controlled by other factors - Boris with a majority won't be.

As yesterday's Newsnight reminded us, he also claimed (on the steps of no. 10, no less) to have an already worked up plan to solve the Social Care crisis. One which they now say doesn't exist.

Really? - Johnson claimed something that wasn't true - I'm shocked...

Barnesian

IanB2 said:

FPT

Barnesian said:

The MRP modelling technique is based on sampling the voting preferences of a large number of demographic groups and then applying the result to the demographics of each constituency. It is a model.
My model does not rely on demographics at all - not even remain/leave of each seat. It assumes that the demographics of each seat are "baked in" to the 2017 actual result including the relative strength of the activists base, council representation, history etc that I don't think the MRP model reflects. It then applies assumptions about tactical voting. It is a model.
The appeal of the MRP model is that it was close last time. But that is a sample of one. I think it was also used in the US so a sample of two.
Imagine a heavily promoted modelling technique (with whizzy multilevel Regression and Post-stratification) that correctly picked the 20/1 winner of last year's Grand National. Would its prediction for this year influence the betting? You bet it would - whether it was a sound model or not! "The computer says"
For this reason I'm betting against the MRP which has moved certain markets e.g LD over 25.5 seats at 3.35 looks good value. It shot up after the MRP was published.
EDIT. If my model proves to be spot on this time, you shouldn't assume it is a good model. It might just be lucky. But I suspect it would be followed with close attention next time!

MRP therefore wins over a 'static' model if the change in voting behaviour is driven by demographics - as has been the case in recent years. The MRP also starts with the data from its panellists (around 100?) in each seat, rather than starting with a blank piece of paper.

Where you have a point is that the MRP will itself affect the result. Last time it came late and few people believed it. This time it comes early and, as you say, with a lot of credibility.

As I understand it, the panellists (around 100?) in each seat are not chosen to be representative of that seat though in total across the nation they are chosen to be representative. So a huge MOE for the panellists - more than +/- 10%.
The big changes in demographics are reflected in the change in vote share. My model is not static. If it were, it would produce the same results as last time.

Richard_Tyndall

camel said:

OnlyLivingBoy said:

camel said:

Richard_Tyndall said:

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Yes but the trick is you then plant a new one. If, once you have your 2 billion trees you are continually planting new trees in rotation you keep the carbon locked up.

I imagined it was better than that - like burying the trunks or something. I'm actually disappointed.

Does it release CO2 when it decomposes? If so, where did oil and coal come from? Isn't that just dead trees and animals that have been buried under successive layers of the earth's surface? Anyway, as others have said, the key is to keep replacing the trees over time.

I think coal and oil was formed in anaerobic conditions and, in the case of coal, require something quite literally seismic to form. If dead trees are left on the surface they will return their carbon to the atmosphere. What I've learned today is that you can store x amount of carbon per hectare. Once you've run out of hectares, you've run out of potential. And one thing we know about the UK is that hectares are expensive - at the last auctions I attended farmland here was selling at £45k/hectare. That's without the opportunity cost of not growing food on it.
On the plus side, after 37 years wondering, we may have found a genuine use for the Falkland Islands where land cost 1000x less.

But as I say, once you have your trees grown to maturity, as long as you keep replacing the ones that die, then the carbon remains locked up. And trees do live a fair amount of time. It used to be said that an oak tree was 400 years in the growing, 400 years in the living and 400 years in the dying. But more recent research has shown that those numbers are probably nearer 600 years in each stage. That means that you can lock up the carbon for up to 1800 years just with a single tree lifespan.

Peter_the_Punter

Casino_Royale said:

FPT - the Conservatives have created a large polarisation in their voting coalition because their economics and cultural values work for older people, and don't for younger people.

If the Conservatives want the votes of younger people they need to not saddle them with so much debt, make buying houses easier/renting cheaper and show that whilst they are patriotic they are also relaxed about multi-levelled identities.

I don't climate change really features here, actually, as the Tories do actually have a good practical set of policies in place that are both deliverable and realistic.

So good in fact that it would be a waste of Johnson's time to put them forward on tonite's Channel 4 Debate on the matter, it seems.

Must be just about damned perfect.

camel

Casino_Royale said:

Jonathan said:

With Boris looking likely to win a majority, it is time to give his plans serious scrutiny and get some firm commitments on the record.

His promise to 'Get Brexit Done' needs to be interrogated. We need to know precisely how he will will avoid a cold, hard Brexit in 12 months. A promise is required. We cannot afford to give him a blank cheque.

We won't get that.

But, we know he's a snake and will throw anyone under a bus to save himself.

In my view, that means he will ratfuck the ERG to get a good, close relationship with the EU, and build on that with a nominal "thin" US one, and similar with Oz, Nz, Japan and Canada.

I don't think he's a No Dealer, but he will be boxed in if he only gets a majority of ten - he really needs 60-80.

An absolutely perfect analysis, IMO.

Andy_JS

isam said:

MikeSmithson said:

alex_ said:

The flaw is that many want Corbyn(ism) to be demonstrably shown to have failed as an election platform and be killed off completely. They desperately need a return to sanity in the main opposition and an alternative to vote for with confidence.

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

Agree 100%

Blame the virtue signalling centrists who put him on the ballot to look fair minded and have been gnashing their teeth ever since

Or was Ed's £3 membership to blame?

PeterC

Sandpit said:

Ah, and I see Vanilla has had another “upgrade” in the last couple of days. Can someone please tell them to stop it?

The expansion of long threads with grey shading is truly awul. IIRC it has happened before but the original format of collapsed threads was restored fairly promptly. Let's hope the same thing happens again.

IshmaelZ

camel said:

OnlyLivingBoy said:

camel said:

Richard_Tyndall said:

camel said:

Scientific info request:
How does planting a tree help with CO2 levels? I understand it will absorb CO2 during its life, but when it dies surely it will release the carbon as it decomposes or is burned.

Yes but the trick is you then plant a new one. If, once you have your 2 billion trees you are continually planting new trees in rotation you keep the carbon locked up.

I imagined it was better than that - like burying the trunks or something. I'm actually disappointed.

Does it release CO2 when it decomposes? If so, where did oil and coal come from? Isn't that just dead trees and animals that have been buried under successive layers of the earth's surface? Anyway, as others have said, the key is to keep replacing the trees over time.

I think coal and oil was formed in anaerobic conditions and, in the case of coal, require something quite literally seismic to form. If dead trees are left on the surface they will return their carbon to the atmosphere. What I've learned today is that you can store x amount of carbon per hectare. Once you've run out of hectares, you've run out of potential. And one thing we know about the UK is that hectares are expensive - at the last auctions I attended farmland here was selling at £45k/hectare. That's without the opportunity cost of not growing food on it.
On the plus side, after 37 years wondering, we may have found a genuine use for the Falkland Islands where land cost 1000x less.

No native trees in the Falklands, which possibly tells us something about the viability of introducing them there.

IanB2

Big_G_NorthWales said:

Nigelb said:

Big_G_NorthWales said:

I am not persuaded you can safely tactically vote and keep Corbyn out
There is only one way to do that and it is to vote conservative

Of course you're not. You are a Tory diehard (though not in the debased HYUFD usage of the term).

I do not like diehard as a term and have called out HYUFD on many occasions over its use

I could be persuaded to vote otherwise in some circumstances but Corbyn has to be soundly defeated and gone by Xmas would be my best Xmas present

Its no good getting rid of Corbyn if he is replaced by another nutter.

The implied logic of OGH's lead is that, once Corbyn has no chance of winning, his -ism is best defeated by lending a hand to the LibDems in replacing Labour as the opposition, if not nationally (this time, so it seems) then in as many seats as possible.

Alistair

YBarddCwsc said:

alex_ said:

Personally I think the disaster that has befallen U.K. politics in recent years can almost all be traced back to the election of Corbyn as leader of the Labour Party. Virtually everything else has flowed from there.

So, it is back to Eric Joyce's fists again.

The election of Corbyn was absolutely key to the Brexit referendum result, agreed (and it is not acknowledged enough the role that Corbyn played, far more important than "Russian interference").

But, Corbyn himself happened because of the huge disillusionment in New Labour following the financial crisis & the Iraq War. Corbyn was inevitable.

He only got m the ballot because dull fucker MPs who didn't support him put him there.