The Polls thread...

[Filuwaúrdjan]

ICM/Guardian:

Lab 40%, Con 33%, LD 20%

[Filuwaúrdjan]

Times/Populus Tracker: Lab 40 (n/c), Con 32 (+1), LD 21 (n/c), Oth 7 (-1)

YouGov/Torygraph: Lab 36 (-1), Con 32 (-1), LD 24 (n/c), Oth 8 (+2)

[Filuwaúrdjan]

The Sunday Polls are out:

CR/IoS: Lab 39 (-1), Con 31 (-4), LD 23 (+5)
YG: Lab 36 (n/c), Con 33 (+1), LD 23 (-1)
ICM/Sunday Torygraph: Lab 39 (-1), Con 31 (-2), LD 22 (+2)

My take: CR's poll looks suspiciously like they've changed their methodology (as indeed they have) and it now fits with the overall pattern. YG shows sod all change, while ICM agrees with Populus that the Tories "Liar! Liar! Pants on fire!" strategy has backfired (note that they've abandoned it).

In addition to that, MORI have done a survey of marginals: Marginal Seat Picture Offers Little Solace to Conservatives

[Filuwaúrdjan]

Not a poll as such, but...

A study by Proffessors Rallings and Thrasher (who apparently predicted the 1997 and 2001 elections accurately. They're the people that calculate notional results after boundary changes, btw) based on local authority by-elections then adjusted for turnout and protest voting has produced the following figures:

Labour: 37%
Tories: 34%
LibDems: 21%

Labour Majority: 96

[Filuwaúrdjan]

Times/Populus tracker update:

Today: Lab 42 (+2), Con 29 (-2), LD 21 (n/c), Oth 8 (n/c)

Tomorrow: Lab 41 (-1), Con 29 (n/c), LD 21 (n/c), Oth 9 (+1)

[Filuwaúrdjan]

The Guardian is publishing three constituency polls by ICM tomorrow.

Apparently they show Finchley & Golders Green neck and neck, David Davis looking suprisingly safe and (I'm stunned by this one) Leslie have a slim lead in Shipley.

No figures yet.

[Filuwaúrdjan]

Finchley & Golders Green: Lab 40/Con 37. When told it was a marginal with the Tories in second place: Lab 39/Con 39
Big rise in LD vote from 2001 apparently. Only way Tories can win back Thatcher's old seat is by a vote split, and it looks like it might happen.

Can't find the Shipley figures. Apparently ICM shows a small swing to Labour from 2001 with a rise in the LibDem vote taking from both parties, but apparently more from the Tories.

Haltemprice & Howden: Con 45/LD 38

In addition to that an ICM survey of marginals (it might be just those three transferred nationally but I don't think so) has Labour on 41% in Lab(held)/Con marginals (down from 47% in 2001) with the Tories staying on 36%
In the LibDem's 48 key marginals, they've actually fallen (according to ICM) from 36% in 2001 to 32% now.
In Con(held)/Lab marginals the Tories are still on 44%, but Labour has fallen to 33% (down 3 on 2001).

[Filuwaúrdjan]

New MORI/FT poll:

[Polkergeist]

New MORI/FT poll:

Using the BBC seat calculator we get....

LAB  403
CON  149
LIB DEM  65
OTH  29

On these figures it would be a disaster for the Tories going backward for them. Mind you if the polls favour anyone its Labour.

[Filuwaúrdjan]

Using the BBC seat calculator we get....

LAB  403
CON  149
LIB DEM  65
OTH  29

On these figures it would be a disaster for the Tories going backward for them.

If there's a uniform swing. It's possible that areas the Tories do badly in anyway swing hard away from them, while they stay roughly where they were last time round in the marginals and pick up seats via vote splitting.

You must be logged in to read this quote.

In the past that's been true, but there's been some big methodolgy changes so maybe not this time. I think that MORI now tends to exaggerate trends; it's first poll of the campaign actually had the Tories in front. They've fallen a staggering 8pts in MORI's polls since the beginning of the campaign.

[Polkergeist]

Using the BBC seat calculator we get....

LAB  403
CON  149
LIB DEM  65
OTH  29

On these figures it would be a disaster for the Tories going backward for them.

If there's a uniform swing. It's possible that areas the Tories do badly in anyway swing hard away from them, while they stay roughly where they were last time round in the marginals and pick up seats via vote splitting.

For the ignorant or those really far away:) are there any significant regional swings that may show up on Thursday ?

[Filuwaúrdjan]

For the ignorant or those really far away:) are there any significant regional swings that may show up on Thursday ?

You must be logged in to read this quote.

If ICM's constituency polls are at all accurate, there looks likely to be a stronger pro-Tory swing in the South than up North (and if the Shipley poll is accurate there might be a swing *to* Labour up North).

[Polkergeist]

For the ignorant or those really far away:) are there any significant regional swings that may show up on Thursday ?

You must be logged in to read this quote.

If ICM's constituency polls are at all accurate, there looks likely to be a stronger pro-Tory swing in the South than up North (and if the Shipley poll is accurate there might be a swing *to* Labour up North).

So are we look at a polarising election with swings to the Tories in Tory held seats and swings to Labour in Labour held seats ?

[Filuwaúrdjan]

So are we look at a polarising election with swings to the Tories in Tory held seats and swings to Labour in Labour held seats ?

Sort of. It looks (from comparing various polls. This could be wrong) that Labour has lost ground from 2001 in marginals (usually defined as a seat with a majority of under 10%) but may have actually gained in seats that are already safe. If the ICM polls are accurate, this could be complicated by an element of regional polarisation; we could see Labour hanging onto very marginal textiles seats in Northern England, while losing more affluent seats in the South with larger majorities.
The media have been spinning this all as "no one knows what's happening in the marginals etc." or that "the polls don't make sense/contradict each other". But I don't think they (the media) are looking at things logically.

[minionofmidas]

So are we look at a polarising election with swings to the Tories in Tory held seats and swings to Labour in Labour held seats ?

Sort of. It looks (from comparing various polls. This could be wrong) that Labour has lost ground from 2001 in marginals (usually defined as a seat with a majority of under 10%) but may have actually gained in seats that are already safe.

Which may be due to higher turnout. May it? PPLLEEAASSEE, can we have some higher turnout?

[Filuwaúrdjan]

Which may be due to higher turnout. May it?

Possibly. Logically it's more a "probably" (certainly MORI is showing a higher turnout than at the start of the campaign) but I don't want to tempt fate (my ideal outcome, btw, is for a Labour majority of about 80 with Labour on a higher vote share than 2001 and with turnout being much higher. Probably won't happen but I'd like it to).
Apparently the Unions have been sending their members postal ballots which might explain a lot actually.

You must be logged in to read this quote.

*fingers crossed*

[Democratic Hawk]

New MORI/FT poll:

Using the BBC seat calculator we get....

LAB  403
CON  149
LIB DEM  65
OTH  29

On these figures it would be a disaster for the Tories going backward for them. Mind you if the polls favour anyone its Labour.

Among those 'naming a party' (n = 1007), Labour are on 41%, the Tories 26% and the Lib Dem's 24%, which would result in Labour having 423 seats; the Tories 115 and the Lib Dems 80 - an overall Labour majority of 200 (www.electoralcalculus.co.uk)

Dave

[Banana Republic]

I don't know how much drugs someone would have to be on in order to believe this MORI poll is even close to being accurate. I have never trusted thier polls, even when they put the Tories on top by 5 points last month(what does that tell you?)

Yesterday's YouGov poll is closest, surely. There couldn't have been this big of a swing in two days when absolutely nothing has been going on campaign-wise

[Filuwaúrdjan]

MORI tend to exaggerate any trends nowadays

[Democratic Hawk]

I don't know how much drugs someone would have to be on in order to believe this MORI poll is even close to being accurate. I have never trusted thier polls, even when they put the Tories on top by 5 points last month(what does that tell you?)

Yesterday's YouGov poll is closest, surely. There couldn't have been this big of a swing in two days when absolutely nothing has been going on campaign-wise

Perhaps folk have realised just how dire Michael Howard is ! No seriously, I think this poll is a bit wide-off-the-mark

YouGov will probably be the most accurate; though, I'd rather ICM were

Dave

[Mr. Morden]

Slightly off-topic (but still on the subject of the UK election--I just don't think it's worth opening up a new thread for this):

What time of day on Thursday are the first results likely to start trickling in?  Also, is there any chance we'll get some leaked exit polls before official results are released?

[Filuwaúrdjan]

Polls close at 10 PM. Exit polls are published as soon as they're ready.

First results come in from about 10:45, usually in safe Labour seats first off.
Most seats declare between 12 AM and 4 AM.
Northern Ireland declare the next day, as do County Council elections.

[Erc]

Polls close at 10 PM. Exit polls are published as soon as they're ready.

First results come in from about 10:45, usually in safe Labour seats first off.
Most seats declare between 12 AM and 4 AM.
Northern Ireland declare the next day, as do County Council elections.

Lessee...that's 5:00 to 11:00 PM EDT...so we Yanks don't have to stay up until the wee hours of the morning to find out.

Convenient, that.

Back on topic...

As for YouGov...is it just me, or does their methodology seem a bit Zogbyish?

[Filuwaúrdjan]

This might also explain something: Class divide

So things *have* changed during the campaign;

Labour have surged with working class voters while collapsing with professionals etc. And these are big swings, up 10 and down 10... big increase in Labour support from lower middle class voters as well (which might explain the Shipley poll).

[Filuwaúrdjan]

New Times/Populus tracker:

Labour: 41 (n/c)
Tory: 27 (-2)
LibDem: 23 (+2)
Oth: 9 (n/c)

Notes:

1. Tracking polls like this are only supposed to measure momentum... Populus will apparently do a final poll tomorrow. Having said that, dropping 6 points over the course of the campaign probably *isn't* the sort of news Howard wanted to hear...

2. For the reasons indicated in the article below, trying a uniform swing probably wouldn't be advisable.

3. No party has ever won power from a position of 27% in the polls.