politicalbetting.com » Blog Archive » At this stage in the marginals having enough volunteer foot

Philip_Thompson said:

Charles said:

Innocent_Abroad said:

Pulpstar said:

SouthamObserver said:

bigjohnowls said:

Fallon message is desperate.

When does the Tory campaign actually start?

Bet DH called last few days a narrow Tory win!!

He's talking to Tory UKIPers, no-one else. That is the Tory campaign strategy; just as the Labour strategy is to talk to the anti-Tories. There is no national debate taking place. The two big parties are having two very different conversations with two very distinct sections of the electorate. That's FPTP for you. Whoever wins is going to lack a proper mandate.

Turnout in England is going to be shocking.

And therefore "legitimacy". Say the poll is 67% and the winning Party polls a third of that - governing with the consent of 22% of the electorate isn't going to be much fun. No wonder the Lamb of God doesn't want to serve a full second term.

If I remember my theology correctly, the Lamb of God's second term will last ten thousand years. That's enough for anyone.

(Lamb of God is a specific title with great significance, not just an cute phrase to throw around for your own amusement)

So are other terms like Messiah etc but that doesn't stop people either. We don't live in a theology so people can break the second commandment if they wish.

Of course - and I wasn't proposing to behead @Innocent_Abroad (just yet, anyway )

I was just pointing out that Cameron has not claimed the title and that many people would find it surprising that it is applied to a politician.

TheScreamingEagles said:

Tykejohnno said:

Yet another depressing poll for the tories.

Keep calm.

Holiday polling is notoriously unreliable. There's a reason why YouGov didn't poll on Easter Sunday/Monday and why Ipsos Mori and ICM are waiting until this weekend to start polling.

If the phone polls show a similar story, then I will share your depression.

Today's YouGov:

Weighted sample:
18-24 (7 year range): 223 = 223/7 = 32 people per year
60+ (21 year range *): 565 = 565/21 = 27 people per year

OK, there may be a few more people in each year in the younger age groups but even allowing for that the poll is assuming approx. the same "turnout" for each age group - ie turnout of the whole population, not just registered voters.

Yet we know half of young people aren't even registered.

And look how few 18-24s YouGov actually found - only 126 and they had to then weight them up to 223.

* Average life expectancy = 81.5 per Wiki.

logical_song said:

Charles said:

Innocent_Abroad said:

Pulpstar said:

SouthamObserver said:

bigjohnowls said:

Fallon message is desperate.

When does the Tory campaign actually start?

Bet DH called last few days a narrow Tory win!!

He's talking to Tory UKIPers, no-one else. That is the Tory campaign strategy; just as the Labour strategy is to talk to the anti-Tories. There is no national debate taking place. The two big parties are having two very different conversations with two very distinct sections of the electorate. That's FPTP for you. Whoever wins is going to lack a proper mandate.

Turnout in England is going to be shocking.

And therefore "legitimacy". Say the poll is 67% and the winning Party polls a third of that - governing with the consent of 22% of the electorate isn't going to be much fun. No wonder the Lamb of God doesn't want to serve a full second term.

If I remember my theology correctly, the Lamb of God's second term will last ten thousand years. That's enough for anyone.

(Lamb of God is a specific title with great significance, not just an cute phrase to throw around for your own amusement)

Google seems to think that 'Lamb of God' is a "groove metal band'.

there's a difference between "Lamb of God" and "the Lamb of God"

Plato said:

What was the change in the methodology? I thought it was odd to alter it at this stage in the game myself.

taffys said:

''I think if they had stuck with their previous methodologists, the Tories would have been ahead in both YouGovs''

Yes, so the Yougov polls are the result of an arbitrary decision by f7ck knows who on f8ck knows what basis.

How many people, just how many, have a vested interest in this election being SEEN to go down to the wire, whether it is or it isn't?

Every news outlet and every pollster. So don;t trust them.

The only ones who don;t are the voters. Trust what they do when they go to the polls.

More here (about half-way down)
https://yougov.co.uk/news/2015/04/08/labour-lead-2/

When is a Red Line, not a Red Line...?

Sturgeon won't commit to voting against a Queen's Speech containing Trident

http://www.theguardian.com/politics/live/2015/apr/09/election-2015-live-blog-conservatives-challenge-ed-miliband-trident-nuclear-labour

What was the change in the methodology? I thought it was odd to alter it at this stage in the game myself.

A cynic would say the tories are doing too well too soon. Nobody's going to buy that special election pullout featuring Ed Miliband's inside leg measurement in that case.

It has to be neck and neck right until the end to keep people reading and watching - and then a sensation at the end so that people tune in to the aftermath.

Pulpstar said:

I'd knock a few off the Labour score from that TNS personally. No chance they are 3% ahead

Talk about a buzzkill!

chestnut said:

TNS presentation is unreadable.

Sample from the net, heavily overloaded with 18-35.

Their front page is eminently readable, with all the data I need for ELBOW (same for Opinium). Populus also good. Ipsos MORI and Survation are OK, save for font size!

Can't say the same for ICM and Ashcroft, where you need to get a calculator out to get sample sizes for each party. Also, YouGov tends not to give Green sample size, or total voting sample size excluding Don't Knows/Refused.

Speedy said:

TheScreamingEagles said:

Tissue_Price said:

I laid Tory most seats @ 1.52 as soon as YouGov gave Labour a 2 point lead on Tuesday night. It would be nice if the market might acknowledge this week's polling!

Looking at it, people are adjusting for YouGov's methodological change.

The interesting thing from both YouGovs this week, that the Tories lead on the unweighted samples.

I think if they had stuck with their previous methodology, the Tories would have been ahead in both YouGovs

The biggest change is that they now use certainty to vote weighting, as yougov explained that should have favoured the tories since older people are more certain to vote, however they found out that was not the case in practice.

It was the no longer weighting to 2010 Voter ID that struck out of for me.

Especially as now the Tories lead in the unweighted samples.

Even some of the Tories leads in the last few months, they've not led on the unweighted numbers

TheScreamingEagles said:

Speedy said:

TheScreamingEagles said:

Tissue_Price said:

I laid Tory most seats @ 1.52 as soon as YouGov gave Labour a 2 point lead on Tuesday night. It would be nice if the market might acknowledge this week's polling!

Looking at it, people are adjusting for YouGov's methodological change.

The interesting thing from both YouGovs this week, that the Tories lead on the unweighted samples.

I think if they had stuck with their previous methodology, the Tories would have been ahead in both YouGovs

The biggest change is that they now use certainty to vote weighting, as yougov explained that should have favoured the tories since older people are more certain to vote, however they found out that was not the case in practice.

It was the no longer weighting to 2010 Voter ID that struck out of for me.

Especially as now the Tories lead in the unweighted samples.

Even some of the Tories leads in the last few months, they've not led on the unweighted numbers

See my post below.

Just go back to basics and think big picture.

Is it remotely conceivable that the total number of votes cast by 18 to 24s (a 7 year range) will be 40% of the total number of votes cast by people over 60 (a 21 year range)?

I'm sorry but I cannot see how this can possibly be the case.

Plato said:

Hmm.

YouGov are also returning to a method we successfully used in the 2005 general election. Our usual polling method relies upon weighting by party identification. However, for the campaign itself we will be drawing our daily polling samples from people who we previously contacted in January and February this year and weighting our data using how those people told us they were voting at that time (a period when the polls were broadly stable and Labour were, on average, slightly less than a point ahead).

This means that we can be confident that any material change in the polls from that position reflects a genuine shift in public opinion since January & February. There will be still be some random sample variation from poll to poll - it can never be eliminated completely - but it will mean any substantial change in the polls will be down to individual people changing their minds (or making their minds ups) since we interviewed them in February.

Sunil_Prasannan said:

Plato said:

What was the change in the methodology? I thought it was odd to alter it at this stage in the game myself.

taffys said:

''I think if they had stuck with their previous methodologists, the Tories would have been ahead in both YouGovs''

Yes, so the Yougov polls are the result of an arbitrary decision by f7ck knows who on f8ck knows what basis.

How many people, just how many, have a vested interest in this election being SEEN to go down to the wire, whether it is or it isn't?

Every news outlet and every pollster. So don;t trust them.

The only ones who don;t are the voters. Trust what they do when they go to the polls.

More here (about half-way down)
https://yougov.co.uk/news/2015/04/08/labour-lead-2/

I really don't like the Yougov methodology change.

''I'm sorry but I cannot see how this can possibly be the case.''

It stinks. But the pollsters know their customers.

TheScreamingEagles said:

Speedy said:

TheScreamingEagles said:

Tissue_Price said:

I laid Tory most seats @ 1.52 as soon as YouGov gave Labour a 2 point lead on Tuesday night. It would be nice if the market might acknowledge this week's polling!

Looking at it, people are adjusting for YouGov's methodological change.

The interesting thing from both YouGovs this week, that the Tories lead on the unweighted samples.

I think if they had stuck with their previous methodology, the Tories would have been ahead in both YouGovs

The biggest change is that they now use certainty to vote weighting, as yougov explained that should have favoured the tories since older people are more certain to vote, however they found out that was not the case in practice.

It was the no longer weighting to 2010 Voter ID that struck out of for me.

Especially as now the Tories lead in the unweighted samples.

Even some of the Tories leads in the last few months, they've not led on the unweighted numbers

Lets wait a little before you go all "unskewed polls" like a Republican in a presidential election.
So far only Yougov and TNS have shown a substantial improvement on Labour's position, while 3 other pollsters have not shown much change yet.

Indeed, Mr. 1983. Reminds me of that poll the other day which had the astounding revelation that people would like pay rises.

Plato said:

Hmm.

YouGov are also returning to a method we successfully used in the 2005 general election. Our usual polling method relies upon weighting by party identification. However, for the campaign itself we will be drawing our daily polling samples from people who we previously contacted in January and February this year and weighting our data using how those people told us they were voting at that time (a period when the polls were broadly stable and Labour were, on average, slightly less than a point ahead).

This means that we can be confident that any material change in the polls from that position reflects a genuine shift in public opinion since January & February. There will be still be some random sample variation from poll to poll - it can never be eliminated completely - but it will mean any substantial change in the polls will be down to individual people changing their minds (or making their minds ups) since we interviewed them in February.

Sunil_Prasannan said:

Plato said:

What was the change in the methodology? I thought it was odd to alter it at this stage in the game myself.

taffys said:

''I think if they had stuck with their previous methodologists, the Tories would have been ahead in both YouGovs''

Yes, so the Yougov polls are the result of an arbitrary decision by f7ck knows who on f8ck knows what basis.

How many people, just how many, have a vested interest in this election being SEEN to go down to the wire, whether it is or it isn't?

Every news outlet and every pollster. So don;t trust them.

The only ones who don;t are the voters. Trust what they do when they go to the polls.

More here (about half-way down)
https://yougov.co.uk/news/2015/04/08/labour-lead-2/

Trying to "un-skew" the polls by counting unweighted numbers strikes me as pretty futile. The weighting's generally there for a reason.

What I would be interested in is whether or not there's an observer effect going on here - by re-polling people again and again are they more or less likely to "change their mind"?

taffys said:

I really don't like the Yougov methodology change.

Given that many of the people polled are dyed in the wool supporters of one or the other main party, its simply a 'keep everybody' in the game methodology.

These guys want to sell polls. They aren't objective.

If they want to sell polls for the next 28 years, not just 28 days, then they need to be as accurate as possible. Remember the pollsters use opinion polls as the shop-window for their main business: market research. If they lose their reputation on the election then they lose their clients afterwards.

chestnut said:

TNS - Euro 2014

UKIP 31 (overstatement 3.5%)
Lab 28 (overstatement 2.6%)
Con 21 (understatement 2.9%)
LD 7 (overstatement 0.1%)

Lab-Con gap 7 (overstatement 5.5%)

By the same definition then the last TNS poll should have shown the Tories 6.5% ahead of Labour, equally problematic.
There is the danger of Tories sliding into "unskewed polls" territory in the last weeks of the GE.

http://www.wnd.com/2012/09/unskewed-polls-show-nearly-8-point-romney-lead/

TGOHF said:

taffys said:

''It would be nice if the market might acknowledge this week's polling!''

You and a stack of other cleverer-than-thou left backers

I'm pretty confident that the polls and the markets are not going to reconcile before election day.

There will presumably be a mega steward's inquiry after the election by most of the pollsters into how they got it so wrong.

Re Tories and policies - surely the absence of policy waving is the policy in itself. They are campaigning mainly on their record. It's what they deliver in government (sound money and recovery) versus what Labour delivers (mayhem and chaos). Why fog the message by joining in Labour's stupid pork auction?

Tissue_Price said:

Plato said:

Hmm.

YouGov are also returning to a method we successfully used in the 2005 general election. Our usual polling method relies upon weighting by party identification. However, for the campaign itself we will be drawing our daily polling samples from people who we previously contacted in January and February this year and weighting our data using how those people told us they were voting at that time (a period when the polls were broadly stable and Labour were, on average, slightly less than a point ahead).

This means that we can be confident that any material change in the polls from that position reflects a genuine shift in public opinion since January & February. There will be still be some random sample variation from poll to poll - it can never be eliminated completely - but it will mean any substantial change in the polls will be down to individual people changing their minds (or making their minds ups) since we interviewed them in February.

Sunil_Prasannan said:

Plato said:

What was the change in the methodology? I thought it was odd to alter it at this stage in the game myself.

taffys said:

''I think if they had stuck with their previous methodologists, the Tories would have been ahead in both YouGovs''

Yes, so the Yougov polls are the result of an arbitrary decision by f7ck knows who on f8ck knows what basis.

How many people, just how many, have a vested interest in this election being SEEN to go down to the wire, whether it is or it isn't?

Every news outlet and every pollster. So don;t trust them.

The only ones who don;t are the voters. Trust what they do when they go to the polls.

More here (about half-way down)
https://yougov.co.uk/news/2015/04/08/labour-lead-2/

Trying to "un-skew" the polls by counting unweighted numbers strikes me as pretty futile. The weighting's generally there for a reason.

What I would be interested in is whether or not there's an observer effect going on here - by re-polling people again and again are they more or less likely to "change their mind"?

I agree, weighted numbers are always the best, is why I said, when the first few YouGovs with Tory leads really weren't Tory leads as Lab led on the unweighted basis.

My point was, if we we're comparing on a like for like basis, this week's YouGov using last week's methodology would probably have the Tories ahead.

Unweighted polling is what, in part led to the great polling disaster of 1992

I notice some early Dick Morris style 'unskewed polls' now appearing on PB...

Talking of which has Romney won Ohio yet???

Luckyguy1983 said:

Aparently Nigel Farage has backed Trident replacement. If that's so, he can forget (such as it is) my vote. This country will only be on its way back to greatness when we eschew expensive status symbols and get back to sound economics and genuine capability, especially in defence. If UKIP isn't for that, what's the point?

Oh come on. We all know what UKIP's for.

They're a market niche that a handful of individuals have chosen to occupy because it's there.

My question is this.

Will there be another methodology change in the last week where they ask whether people are registered to vote and then bingo 18 to 24s go right back down in the weighted sample?

Sorry to be a fuss but I want an answer to my question:

Are psephological experts really expecting the total number of votes cast by 18 to 24s to be 40% of the total number of votes cast by over 60s?

If they are fine. But I would like to see the calculation. Because it seems to me that even the most basic Primary School numeracy skills would show that this cannot possibly be right.

bigjohnowls said:

Pulpstar said:

I'd knock a few off the Labour score from that TNS personally. No chance they are 3% ahead

Decimal place wrong with TNS.

I reckon LAB 0.3% ahead right now

Still no strong evidence against the roughly dead-heat theory. Good news for Labour whilst it lasts, but a tenuous 'lead' indeed.

Bond_James_Bond said:

Luckyguy1983 said:

Aparently Nigel Farage has backed Trident replacement. If that's so, he can forget (such as it is) my vote. This country will only be on its way back to greatness when we eschew expensive status symbols and get back to sound economics and genuine capability, especially in defence. If UKIP isn't for that, what's the point?

Oh come on. We all know what UKIP's for.

They're a market niche that a handful of individuals have chosen to occupy because it's there.

Everything that exists does so because it occupies a market niche.

TheScreamingEagles said:

MikeL said:

My question is this.

Will there be another methodology change in the last week where they ask whether people are registered to vote and then bingo 18 to 24s go right back down in the weighted sample?

Sorry to be a fuss but I want an answer to my question:

Are psephological experts really expecting the total number of votes cast by 18 to 24s to be 40% of the total number of votes cast by over 60s?

If they are fine. But I would like to see the calculation. Because it seems to me that even the most basic Primary School numeracy skills would show that this cannot possibly be right.

I think the next methodological change will be from next week, when they ask people if they have already voted, and those voters will not form part of the sample.

They SO should though

Luckyguy1983 said:

Oh, and can I also say, I've tried to 'get' David Coburn, and I'm sure he's a very nice gentleman with his heart in the right place, but he just comes over as a sort of basement dwelling bad version of Nigel, made up of all the left over bits, and always getting it wrong when he attempts Farageisms. If you had to create someone worse to represent UKIP in Scotland (not an easy sell at the best of times), you'd struggle.

Certainly agree with you on this,he was awful.

TheScreamingEagles said:

Speedy said:

TheScreamingEagles said:

Tissue_Price said:

I laid Tory most seats @ 1.52 as soon as YouGov gave Labour a 2 point lead on Tuesday night. It would be nice if the market might acknowledge this week's polling!

Looking at it, people are adjusting for YouGov's methodological change.

The interesting thing from both YouGovs this week, that the Tories lead on the unweighted samples.

I think if they had stuck with their previous methodology, the Tories would have been ahead in both YouGovs

The biggest change is that they now use certainty to vote weighting, as yougov explained that should have favoured the tories since older people are more certain to vote, however they found out that was not the case in practice.

It was the no longer weighting to 2010 Voter ID that struck out of for me.

Especially as now the Tories lead in the unweighted samples.

Even some of the Tories leads in the last few months, they've not led on the unweighted numbers

That, and instead they are only polling members of their panel who answered surveys in January and February, so as to weight to people's responses at that time. I think this means their polling panel for their polls is now roughly the size of the electorate for one constituency.

New Thread

Luckyguy1983 said:

Danny565 said:

Tim Montgomerie hits the nail on the head: If voters have a choice between an individualist right and a big-government left, the latter will prove more popular

http://www.thetimes.co.uk/tto/opinion/columnists/article4405688.ece

The Tories are simply well to the Right of mainstream opinion on economics. The only reason the Tories are still in the game is because people trust Cameron a lot more on the world stage / in negotiations with the baddies a lot more than they trust Miliband. Nothing else.

That has always been the case. A majority of people always prefer instant pleasure and deferred pain, and prefer emotion to calculation. That isn't something to celebrate.

There was an interesting item about the psychology of political campaigns / voting patterns on radio 4 yesterday, with evidence from many countries. Basic message was that assuming people listen to arguments and vote according to policy preferences is dead wrong. It's much more visceral and emotive than that. Labour's policy and competence and reason and execution and coherence all suck harder than the vacuum of deep space - but they're terribly terribly good at summoning up a well of unreasoned emotion from the lumpen blob. Tories = head. Labour = heart. Heart is stronger than head when it comes to voting in most cases. How very depressing for those of us with a brain.

Charles said:

Plato said:

My knowledge here is very patchy - is The Lamb the thingy that was sacrificed in Revelations to mark the Apocalypse?

Charles said:

logical_song said:

Charles said:

Innocent_Abroad said:

Pulpstar said:

SouthamObserver said:

bigjohnowls said:

Fallon message is desperate.

When does the Tory campaign actually start?

Bet DH called last few days a narrow Tory win!!

He's talking to Tory UKIPers, no-one else. That is the Tory campaign strategy; just as the Labour strategy is to talk to the anti-Tories. There is no national debate taking place. The two big parties are having two very different conversations with two very distinct sections of the electorate. That's FPTP for you. Whoever wins is going to lack a proper mandate.

Turnout in England is going to be shocking.

And therefore "legitimacy". Say the poll is 67% and the winning Party polls a third of that - governing with the consent of 22% of the electorate isn't going to be much fun. No wonder the Lamb of God doesn't want to serve a full second term.

If I remember my theology correctly, the Lamb of God's second term will last ten thousand years. That's enough for anyone.

(Lamb of God is a specific title with great significance, not just an cute phrase to throw around for your own amusement)

Google seems to think that 'Lamb of God' is a "groove metal band'.

there's a difference between "Lamb of God" and "the Lamb of God"

No: it's the term that John the Baptist used to refer to Jesus and, as the Agnus Dei ("behold the Lamb of God who takest away the sins of the world") is used in the Eucharistic Prayer

Also gets a mention in "Jerusalem"

Indeed - a cull of the top-brass and a reversal of the one-to-four ratio of civil service to military personnel would not go amiss either.

Plato said:

A better ratio between top brass and troops would be good too.

Luckyguy1983 said:

Aparently Nigel Farage has backed Trident replacement. If that's so, he can forget (such as it is) my vote. This country will only be on its way back to greatness when we eschew expensive status symbols and get back to sound economics and genuine capability, especially in defence. If UKIP isn't for that, what's the point?

Luckyguy1983 said:

Plato said:

I'm all for keeping our nuclear deterrent. I wish the world didn't have them, but we do and in an international game of MAD - I want both our place on the Security Council and our own nukes updated.

The world is becoming unstable again - it's not as simple as the old Cold War anymore. And given all our Londistan issues - frankly I'm pleased to hear we're getting 4 new subs et al.

A better ratio between top brass and troops would be good too.

Luckyguy1983 said:

Aparently Nigel Farage has backed Trident replacement. If that's so, he can forget (such as it is) my vote. This country will only be on its way back to greatness when we eschew expensive status symbols and get back to sound economics and genuine capability, especially in defence. If UKIP isn't for that, what's the point?

But how exactly can strategic nuclear weapons (of dubious genuine capability I might add) protect us from something like ISIS? Surely better spent (as you rightly indicate) on conventional forces.

256 mirv warheads should deal with the middle east very nicely. There wouldn't be much to visit afterwards though and the oilfields a little fiery.