Party's private Populus poll published

Labour 37% (-2)
Conservative 36% (+3)
Lib Dems 16% (+1)

UK Polling Report's Anthony Wells has forced the release of the leaked internal Populus poll, under the British Polling Council's disclosure rules:

"It is obviously a sharp drop in the Labour lead and, for the Conservatives, this is back to the sort of levels of support they were recording in Populus polls prior to Gordon Brown becoming leader. The only reason they are not ahead is that Gordon Brown has won back some of the support Labour had lost to the Liberal Democrats."

Full stats here. Taken from Saturday to Tuesday, it spanned the same bank holiday weekend as the disappointing YouGov poll. If next week's Populus poll for the Times shows the same sort of figures the risk of calling an early election would surely be too high for Brown.

Deputy Editor

Late update: The Wilted Rose has looked at the regional stats in detail and they paint a very mixed picture, including an 18% Conservative lead in the South East, and a 7% lead in the Midlands.

The blog has also compared these variations with those of the pre-election Populus poll in 2005 and concluded:

"If this is what happens with a Brown bounce: Tories well ahead in the South East / 6% ahead in the Midlands /Slight improvement in North and Wales/SW. Huge swing to SNP putting them only 4% behind Labour in Scotland.Then what would bad times for Brown be like?"

• South East: the Conservatives up 9% since the 2005 poll, LD down 9% and Labour down 2%. This will bring huge gains from Lab & LD.
• Midlands: Tories up 5%, Lab down 2% and LD 1%. Again, significant gains, although Electoral Calculus doesn’t do justice.
• North: Con +3%, LD down 3% and Lab down 1%. Some limited gains, but that could include Bolton W (Ruth Kelly).
• Wales/SW: Conservatives up 1% and LD and Lab down 1% with Plaid Cymru up 4%. Again, a few gains here and there for Con & PC.
• Scotland: Lib Dems and Conservatives decimated in Scotland with SNP up 15% and Lab up 11%. So there is a Brown Bounce – but it is only in Scotland (where Brown is popular). But SNP has increased significantly, at the expense of Con (16% to 8%) & LD (22% to 6%).

1st September update:

Comments

[David Sergeant]

Er, so lets get this staight. The YouGov poll says Labour vote is down by 1% and the Tory vote up by 1%. So this justifies a headline in the Telegraph (the Telegraph) "Cameron staring defeat in the face"? I wonder if Gordon Brown writes the headlines for the Telegraph although some Labour spin is a bit less hysterical.

[Man in a Shed]

Or is this a trap for Brown. Encouraging him to go when the going is less favourable for Labour than they realise.

The release of this poll perhaps suggests this isn't the case.

Labour must have their own private polling and suggestions are that they are clearing the decks right now ....

They may think that the advantage would be to move before the Lib Dems have disposed of Ming - and have a weak Ming as an insurance policy for a hung parliament.

The disadvantage will be the performance of the SNP.

But I think we can give them a bloody nose once a full campaign develops.

[DC]

Who is Anthony Wells?

[Pisaboy]

The YouGov poll WAS VERY BAD NEWS, David. We were only ahead on 3 out of 19 issues!!!
You wouldn't know that by reading this 'blog' though would you.

[Mountjoy (The Wilted Rose)]

I have just blogged on this very issue, by analysing the results and comparing to a previous Populus Poll for the Times.

The results would be spectacularly bad for Labour if Gordon Brown were stupid enough to go for a snap General Election. The Tories, although only one point ahead nationally of Labour, are :

• 18 points ahead in the South East (at 49% (+11% up from 38% in the Populus Times poll) with Labour at 31% (-6%) and LD 14% (+2%)
• 7 points ahead in the Midlands at 43% (+6% from 36%) and Labour 36% (no change) and LD 14% (-1%)
• 18 points behind in the North at 27% (down 2%) with Labour on 45% (+2%) the Lib Dems on 22% (+4%)
• 1 point ahead in Wales/SW at 32% (+1%) with Labour on 31% (-7%) and LD on 19% (+5%)
• In Scotland Labour are on 40% (+7%), SNP on 36% (+8%), Con 8% (-10%) and LD 6% (-7%).

As a result of these figures:

• Sweeping gains for the Conservatives in the South East and London (predominantly Labour, but a few LD seats too);

• Tories to pick up a large number of key marginal seats in the Midlands; the urban Midlands notoriously didn't shift for the Tories in the 2005 general election;

• Lib Dems gaining ground from Labour in the North but little Conservative progress;

• Major gains for the Tories in South West and to some extent in Wales;

• SNP gains from Labour and LDs in Scotland.

However, the North-South divide is becoming more pronounced.

More regional analysis and implications here.

[EML]

I always find the regional breakdowns really interesting. The Conservatives appear to be ahead in the south east, midlands and the south west/Wales, whereas Labour are stacking up votes in Scotland the north of England. The averages comes out roughly the same, but there are a lot of marginals in the south and midlands that we can pick up, even being 1% behind overall.

This sort of result would not give us anywhere near an overall majority, of course, but it could very well create a hung parliament. And if these sort of polls continue then Brown will have no choice but to hold off until next summer or autumn, which will play straight into our hands.

If we can get through the next four months unscathed, repeatedly hitting the government and getting our message across, and hopefully staying within reach of Labour in the polls, then I think Brown is screwed. He will have blown his chance. His only chance of winning a majority of his own is to go for an election asap.

[Tony Makara]

This shows that the next election will be the most exciting for decades. I'm curious about how the Liberals will respond if the result is tight? Kennedy was against any deal with Labour, Ming has indicated that he is in favour. Would another Lib/Lab pact be likely? What would the Liberals want out of Labour? Hopefully Mr Cameron can get a working majority but the polls, which are just snapshots, show things could be very close.

[gadfly]

Just a little 'rain' on the populus poll. The simple question, when have populus been in the slightest accurate about any election? ICM, with a margin for error of some 3% and you gov with a margin of error of 1% have consistently called the elections correctly at national level, particularly in general elections. They both have labour much further ahead. I think it is best to wait for a few more of theirs before becoming too gloomy or too happy. The polls will really matter at the end of September/beginning of October.

[Sean Fear]

I think the evidence is mounting that Brown stacks up votes for Labour precisely where they don't need them.

[gadfly]

Just a little 'rain' on the populus poll. The simple question, when have populus been in the slightest accurate about any election? ICM, with a margin for error of some 3% and you gov with a margin of error of 1% have consistently called the elections correctly at national level, particularly in general elections. They both have labour much further ahead. I think it is best to wait for a few more of theirs before becoming too gloomy or too happy. The polls will really matter at the end of September/beginning of October.

[Sean Fear]

The simple question, when have populus been in the slightest accurate about any election?

On eve of poll, 2005, they performed as well as ICM and Yougov.

[Ben Surtees]

Excellent news IF reflected in the populus poll out next week and subsequent polls.

If replicated at a general election, the likely growth in the LibDem vote over the course of the campaign, the likelihood of substantial regional variations and the strength of the Conservative field organisation would make the contest exceedingly close in terms of seats as well as vote share.

But the most important thing is that this be reflected in other polls, most importantly the populus poll out next week – If it is reflected then we have some much needed positive news on the polling front and the first suggestion that the Brown Bounce could be fading, as its is we must wait and see.

It’s also worth noting that (and this goes for YouGov as much as this poll) that polls conducted over bank holidays can be notoriously volatile (IIRC populus did in fact record a huge Labour lead during the ’05 campaign in a poll conducted over a bank holiday), but as I say this goes for YouGov as well.

[Gadfly]

Actualy, you need to look back over the whol election to see that they were wildly out of step with the other two. It was strangely that on eve of poll they came suddenly back into line with the others having been all over the place.....

[Mountjoy (The Wilted Rose)]

Gadfly,

On 3rd May 2005, the Times Populus poll showed Con 32%, Lab 37%, LD 21%.

The actual results were Con 33%, Lab 36% and LD 22%.

In other words, Populus overstated Labour support and understated Con & LD support by 1%.

Populus is easily the most accurate opinion poll in that election.

[Gadfly]

Actualy, you need to look back over the whol election to see that they were wildly out of step with the other two. It was strangely that on eve of poll they came suddenly back into line with the others having been all over the place.....

[Ted]

Mountjoy - actually Labour was 35%, Conservatives 32% and LibDems 22%
http://news.bbc.co.uk/1/hi/uk_politics/vote_2005/constituencies/default.stm
Otherwise agree that they were close to the result.

[Mountjoy (The Wilted Rose)]

Thanks, Ted - sorry about the misleading figures.

That's bizarre - for some reason on Electoral Calculus (http://www.electoralcalculus.co.uk) it gives Con 33.2%, Lab 36.2%, LD 22.6%!

I checked the report by Professors Rawling & Thrasher on the Electoral Commission website and they agree with the BBC figures.

[Yet Another Anon]

Mountjoy - actually Labour was 35%, Conservatives 32% and LibDems 22%
Opinion poll figures and most national vote figures given by the media exclude Ulster. Excluding Ulster Labour got 36%, Conservative 33% and Liberal Democrat about 22.5%; however SDLP are affiliated to Labour and Alliance Party are affiliated to the Liberal Democrats so neither party fields candidates in Ulster. The Conservative Party field the odd candidate but except for getting the odd councillor they don't get many votes; including Ulster Labour got 35.2%, Conservative 32.3% and the Liberal Democrats about 22.1%.

[Yet Another Anon]

That's bizarre - for some reason on Electoral Calculus (http://www.electoralcalculus.co.uk) it gives Con 33.2%, Lab 36.2%, LD 22.6%!
Those will be figures excluding Ulster, the figures the BBC gave will have included Ulster.

[Ted]

Yet Another Anon - thanks, reason I checked Mountjoy's figures is I keep seeing 36% quoted but remembered 35%/32% and exclusion of NI explains it.
It was the 32% that made me a moderniser - if after 8 years of a Labour Government, with a hugely unpopular war, suggestions of sleaze, we couldn't even get a third of the electorate to believe in us we needed to accept that our party needed a radical change to make it a party of the 21st Century not one of 1979.

[Oberon Houston]

Sean Fear, help us out here, what is going on? How can the Telegraph publish those results and stand by them? One other observation is this - often political broadsheets talk up the 'opposition' in polls, the Telegraph seems to be doing it here, and the Guardian, I have noticed does the same, why would editors want this? I'm now completely confused as to what if anything polling can tell us.

[Andrew Woodman]

Probably trying to produce poll results to convince Brown to call an election Oberon. Elections sell papers.

[Mountjoy (The Wilted Rose)]

Yet Another Anon, thanks from an Ulsterman!

[Yet Another Anon]

if after 8 years of a Labour Government, with a hugely unpopular war, suggestions of sleaze, we couldn't even get a third of the electorate to believe in us we needed to accept that our party needed a radical change to make it a party of the 21st Century not one of 1979.
The War in Iraq was right and IDS and Michael Howard were right to support the war, every Conservative leader back to Margaret Thatcher supported the war.

Certainly the war strategy was flawed, British forces weren't and aren't well enough equipped and the dossier was plagiaristic and largely rubbish. The real reason to remove Saddam Hussein should have been because he and the Iraqi Ba'ath Party were destroying the economy, environment and people of Iraq and were a menace to their neighbours, and were developing new missiles prohibited under UN Resolutions and the 1991 Ceasefire Agreement and while they might have temporarily ceased developing WMD's this was only until international attention moved away from them at which point no doubt Iraq would have resumed development of nuclear weapons.

So far as the proportion of the electorate goes - I assume you mean a third of the Popular Vote as the proportion of the electorate voting Conservative dropped sharply in 2001 with actually fewer total votes than Labour got in 1983; the total Conservative vote in 2005 was actually a sizeable recovery towards that of 1997 in terms of proportion of the electorate.

I think that the next General Election, which I am still convinced will be on 11 June 2009 will see quite a sizeable recovery in both the Labour and Conservative votes, Labour's vote in 2005 having been their lowest in absolute numbers since 1983 and as a percentage of the popular vote almost as low as 1992.

[Dominic]

Middle England are taxed to the hilt over the years the Conservative party have lost thousands of their core voters thousands have joined UKIP they have lost the Essex man vote years ago the pinkie feely touchy clap trap stance is not working look at the polls out today look at the Epping result last night beaten into fourth place with the BNP coming top in Epping . Telegraph today Tory paper statement” polls show Cameron could face landslide defeat if snap election is called” hardly impressive I have been a Tory supporter all my life I would love the Tories to top the polls it would be good for Britain but it is not going to happen.
With a Government that has been in power for ten years with open goals nearly every month Cameron has not laid a glove on them.
Good Conservative values are needed to bring back in the fold all the people that gave Thatcher land slide victories just compare the Shadow Cabinet then and now.

[Mountjoy (The Wilted Rose)]

So there are now rumours of an announcement of a general election being made by "Kamikaze" Brown on Tuesday? .

In the meantime, here is a rough-and-ready analysis of the likely Conservative gains according to this Populus poll. Is Brown ready for the loss of significant numbers of his MPs? He is walking a tightrope in Scotland, too, with the SNP on 36% of the vote there according to Populus (or it could be 39% or 41%, you never know). Can Brown take that risk?

[malcolm]

Dominic you forgot to mention the UKIP vote in the Epping by-election. 28 votes.
I rather think your remark that thousands have left the Conservative party for UKIP is wishful thinking too.I'm pretty sure that it has been reported that UKIP membership has almost halved in the last two years. Can someone from UKIP confirm?
In Essex the Conservative party won every seat in non metropolitan part of the county excepting Colchester and made sweeping gains in the last locals all over the county so I don't think your argument really stands up to any scrutiny at all.If you want conservative government you only have one choice and that is the Conservative party.

[Mountjoy (The Wilted Rose)]

malcolm, absolutely right. Using the Populus poll, in Essex/Herts I have Con Gains in Basildon South and East Thurrock; Harlow; Milton Keynes South; Watford; and Colchester. And, given the regional variations, Stevenage may fall too - not sure about Thurrock, though.

[Edison Smith]

South East Con gains potentially:

Dover
Chatham and Aylesford
Brighton Kemptown
Dartford
Hastings and Rye
Hove
Crawley
Lewes
Sittingbourne and Sheppey
Gillingham and Rainham
Thanet South
Rochester and Strood
Oxford West and Abingdon
Bath
Thornbury and Yate
Cheltenham
Dorset South
Portsmouth South
Eastleigh
Winchester
Romsey and Southampton North

About time with Bath if you ask me.

http://thewiltedrose.wordpress.com/about-2/election-predictions-based-on-the-august-populus-poll/

[John Scott]

A word of warning about this "internal" poll, which everyone is getting excited about. It is out of kilter with most other polls - today's MORI in the Sun has Lab: 41, Con 36, Lib 16. Similar to ICM and YouGov gave Labour a 10-point lead. Take and average and then do the calculations of seats - it might be more sobering, and also more accurate.

[Yet Another Anon]

South East Con gains potentially
Most of those places you listed are not in the South East - Bath, Winchester, Oxford & Abingdon, Romsey and Southampton North, Dorset North, Cheltenham, Portsmouth South and Brighton Kempton being notable examples. Indeed most of them are South Central or South West. They might be possible Conservative gains, but not South Eastern ones.

[Annabel Herriott]

When I opened the DT, my first thought was that the Editor is trying to TEMPT Gordie into running early I would tell GB, bring it on, b ut McCavity wasnt there. He was sitting in his control centre, Mekon like, staring at his screens, and biting his fingernails!

[Traditional Tory]

I think I tend to value the weighty opinions of Prof. King above those of Mountjoy/Wilted Rose or whatever he chooses to call himself.

On his personal blog this clever little chappie refers to 'the Conservatives on 33% - under David Cameron, its best ever leader.'

Really?

Eat your hearts out Disraeli, Churchill, Thatcher...

[Moral minority]

Some of the seats that Edison Smith mentions are nominally Tory due to boundary changes, e.g. in Kent. Several used to have large Tory majorities.

Mr Smith would benefit from a geography lesson, as "Yet Another Anon" notes. We used to hold Gloucester and a few of the Bristol seats. He does not mention Eastleigh or other target seats on the Kent and Sussex coast. Has Edison given up on them?

There is little chance of getting Norman Baker out in Lewes. He is very popular has has a huge personal vote.

[Mountjoy (The Wilted Rose)]

Yet Another Anon - I have used the Government Office Region/Regional Development Agency definition of the "South East", which does actually include Hampshire (i.e. Winchester, Southampton), Sussex E & W (Brighton) and Oxfordshire. I will move Bath, Dorset etc to the South West as it was rather late last night and I mixed up my regions. I campaigned against the NE Elected Regional Assembly, so am a sceptic of regionalisation: these GOR/statistical definitions just don't ring true!

Moral Minority - it is I who would benefit from a geography lesson, not Edison Smith (although I have a PhD in Geography!!!), as I apologised for above. However, Eastleigh is on the list here - Bristol NW is on the list on the link he provides. Gloucester has too big a Labour majority to be a Con Gain on these opinion polling figures. On these opinion poll figures, personal vote or not, Norman Baker is out!

Traditional Tory - again I emphasise that I am not "Felicity" - but I was being deliberately provocative when I said Mr Cameron is the 'best ever leader' - and perhaps a little tongue in cheek!