How ICM made a 17% drop in Labour support into just 1%
April 18th, 2005
-
Do you believe the pollster’s mathematics?
FACT: For their Mirror survey published this morning 341 people told ICM that they had voted Labour last time. Yet only 291 said that they were planning to vote for the party on May 5th – a decline of about a sixth
YET the headline figure that was published was that Labour would get 41% - just one percent down on what the party got four years ago.
SO the opinion poll that has been driving the betting markets is based on ICM methodology that turns 17% into 1%.
With the Tories it worked the other way. Nearly a fifth more people told the pollster that they were voting for Michael Howard’s party than said they voted Conservative last time. Yet the headline figure from ICM had them with an increase of just over 1%.
With the Liberal Democrats the difference was even more marked. The number who told the pollster that they were supporting Charles Kennedy’s party was over a quarter than the number who said they voted this way last time. Yet ICM only put the headline figure up 1%.
-
All this is based on ICM’s view that people do not tell the truth when asked how they voted four years ago. To deal with this they have a complex equation which weights towards Labour and away from the Lib Dems and the Tories.
Are ICM right or are ICM wrong? We’ll know on May 6th.
In the meantime there might be good value in the betting markets. In 2001, in common with all other pollsters, ICM overstated Labour.
Mike Smithson
MessageSpace Advertising
if i do my own maths properly, that means that people are saying to ICM : L 35 C 38.5 LD 22.5 which looks to me a more credible outcome than 41/33/20 …
Astonishing! Thanks for posting this Mike - it explains a lot.
Mike,Are you saying that ICM is a phoney and that we shouldn’t take notice of them,or indeed all telephone and ‘face to face’ polls,because I believe that they all weight in a similar fashion????
So now all opinion polls according to Mike are rubbish, papers having spent loads of money on these polls are getting a duff load of information and we are actually going to get a tory govt on May 6th. Mike seems to hava a ‘thing’ for non-telephone polls, I think he has an agenda.
I have always thought that trying to account for “ashamed supporters” was a total sham. Let us not forget that it used to inflate the Tory figures, when everyone was profoundly ashamed to admit support for John Major.
Let’s adopt the more honest US approach: tell us the true findings, and then let us decide on their verity and value.
ICM should hang their heads in shame. I am truly taken aback by this - not something that a hardened political activist can say very often!
Maybe this explains Labours success in the polls rather more convincingly than that the pollsters are all fools or knaves!
http://politics.guardian.co.uk/columnist/story/0,,1462292,00.html
butchjoe,I agree with you that Mike may have an agenda.It could be that he simply wants to be controversial.It is a fact that people do not recall accurately how they voted previously.e.g.I think it is ICM that published a figure of 55% declaring that they voted Labour in the 2001 election.
If this is their finding then they must correct to allow for this obvious descrepancy
5. As i understand it, this isn’t quite the ‘ashamed supporter’ syndrome we hear about - rather it’s the ‘Manchester United syndrome’, where more people claim to have supported the winning team (Labour) in the last cupfinal (GE). Is that right Mike?
What I’m deeply curious to know is how the pollsters arrive at the idea that voters do this, and just how they weight for it. Any ideas?
If the pollsters have got this wrong…
Forgive me for posting this twice, but I just put it on the other thread and it seems very relevant.
After the debacle of 1992, ICM’s GE performance has been pretty good: in the last three Guardian polls of the campaign in
1997
ICM OVERSTATED THE TORIES by an average of 2.27
UNDERSTATED LABOUR by an average of 0.8%
OVERSTATED LIBDEMS by an average of 0.4%
2001
UNDERSTATED CONSERVATIVES by an average of 0.7%
OVERSTATED LABOUR by an average of 3.3%
UNDERSTATED LIBDEMS by an average of 1.5%
However, in 2001 their final poll was correct to 1% or less for each party, and in 1997 their biggest error was overstatement for the Conservatives. I don’t see why they should have suddenly started getting it so wrong.
Stuart. What this thread misses out is that if you ask any section of the public how they voted at the last election the sample will always be overwhelmingly Labour. Much more so than was actually the case. So there are two possibilities. Firstly that voters remembered wrongly which way they voted last time or secondly that most Tory voters hide between elections and only come out on polling day. ICM suspect the former but if you think the latter then go gambling!
Mike In fairness the markets are not being particularly influenced by this poll. The You Gov poll today equates to a 92 majority-20 more than the middle of the current IG index. The ICM Mirror poll would equate to something like 150. My previous post on this seems to have got wiped but although I don’t believe for one minute that ICM are anywhere near right it’s in their commercial interest to try their best to be accurate. They’re not idiots who are programmed to get it wrong. You described them not so long ago as the’gold standard’ pollsters. Why not get all the pollsters to answer detailed questions on this site on their methodology and why they do what they do. In your opinion should any adjustment have been made to those figures and if so what would you suggest ICM do?
To 4 BJ. Yes I do have an agenda. I think it important that people are aware how opinion polls operate and the basis on which their figures are produced. The more understanding of this the better. So much of our political debate in the UK is framed by the polls yet there is very little examination of how the figures are produced.
I do think that ICM are right to take into account false recall when they deal with the past voting question. The point I’m making is that the headline figures are based on the way they deal with this. They might be completely right and we’ll know on May 6th.
Mike.I have just run Baxter with your corrected ICM numbers.The Labour seat position is now less that 300.Please advise::Should I take a sell position on Labour seats at 357???
The walls are closing in on Mike. He’s going to have no room for manoeuvre soon!
Do people honestly forget what they voted for last time? I find that very hard to believe.
Chrisco 9. ICM had two final polls in 2001 both starting on the same day. The first, for ther Standard, had the larger sample and reported a 17% Labour lead. The second in the Guardian had an 11% one.
Perhaps small “c” and big “C” conservatives are just that much more reticent - unwilling to cooperate with pollsters. Therefore, a sample may appear demagraphically balanced when in fact it isn’t.
Ray 13. Do as I have Ray - put your money into the turnout spread.
Well it is a little confusing if “the gold standard” suddenly stops being so!
Thanks Mike.Just a gentle wind up.
“ICM Research interviewed a random sample of 1,332 adults aged 18+ by telephone on 2-3rd June, 2001. Interviews were conducted across the country, with a boost in London so that the total number of London interviews was 502. The results have been weighted to the profile of all adults, with London boost interviews downweighted into the correct national profile.”
Could this London boosting have had anything to do with the result?
21 was for 16.
ICM full data (use the link to the PDF doc on each ICM poll on their web site to see this) shows that C1’s and C2’s are difficult to find using phone polling, so they “weight” to correct this low base number upwards.
The question with this is:- is the low base of C1’s caused by some kind of artifact of phone polling? Does it meen C1’s (and to a lesser extent C2’s are hiding from phone pollsters?)
CLI and/or answering machines can be used to screen out pollsters, so are C1’s more likely to do this because they lack AB’s nannies or DE free time???
The ICM marginal poll in particular triggered my “that can’t be right” sense… any comments??
The played-with ICM figures seem to be the polling equivalent of Sunny Delight. So many additives its hard to fine the core ingredient.
I have also just checked, and the average I have given in post 9 includes the Evening Standard poll.
I have no sense of this election at all. Most of the Tory canvassers say things are going particularly well. Some left wing journalists are saying Labour are having difficulties. Nick who I think we all respect seems to be suggesting something close to 2001. Michael Howard seems to be a problem. But so does Tony Blair. Until we have something better we can only look at the polls……
26, Still I have Labour at 80 -100 seat Maj, I can feel it in my water, which is a good thing looking at theis thread
Roger - 26 I completely agree with you - I cannot make it out either.
Populus - Lab 40 Con 31
Mori - Lab 40 Con 32
ICM - Lab 41 Con 33
A clear trend is developing Lab up LD up Con down
Will the Tories hold their nerve or will the panic set in, judging by the laughable Willetts/Osborne Press Conference this afternoon the latter already has….
If there is a trend towards Labour it has been caused by the Tories’ poor manifesto and their “are you thinking what we’re thinking?” campaign. Nothing is better designed to increase the Labour turn-out and provide waverers with a good reason to vote.
The Tory party should learn to communicate with the public in a non-insulting way.
26. Roger, if the swing is not uniform they could all be right.
It could be possible that Nick we’ll have the same result of 2001 (he seems a pretty active MP, so he could have a better than average swing). it could be possible that tory canvassers you know are right too (before 2001 if someone would have asked Romford tories canvassers about the situation, he could have thought that Labour was heading for a defeat).
In your humble opinion 29.
Is it possible to place bets on how many Conservative ideas will be pinched by the next Labour Gov.(if) I have often heard the Conservative Party referred to as Labour’s think tank!
It highlights the point that pollsters should place the raw data next to the adjusted data. This election is strange though. The Tories can’t win, but I’m not sure I am confident of what labour’s majority (if they have a majority) will be until polling day.
A couple of days ago I coined the phrase “Zogbyitis” for polls that tweak their raw data so much the original poll disappears under weightings and interpretation.
There seems reason to believe ICM have caught the condition in a big way.
“Populus - Lab 40 Con 31″
is the new Populus one?
35 - Populus and Mori out tomorrow
BTW apparently CK in the newsnight interview has said he wants to stay on as leader after this election to fight the next one.
Forget the polls. Nobody at Labour HQ takes them seriously. It’s a close election, but I’m still predicting a comfortable Labour majority in the region of 60 seats, assuming turnout of 60% or more, from anecdotal evidence and private polling.
As I’ve said before, Tories will do better in London and essex/kent. Million dollar question is what becomes of the Liberal Democrats. I think their campaign has been uninspired and poor and Kennedy seems to becoming more flakey with each day. Terrible against Paxman earlier.
It doesn’t really matter what a handful of people living in a few select wealthy suburbs in the south of England think is going to happen in the election; in the real world Conservative canvassers that I speak to are having good returns and are confident of doing much better than the polls suggest. All those C1s and C2s - you know the people who actually work for a living rather than just talk about it - can no longer afford the Labour Party.
31, Ask any canvaser and they will always tell you they are heading for victory, had Labour around tonight, Local MP none the less,
Mike, I think you have commented before that, whilst poll figures may not be entirely accurate as regards the level of support for the parties, what they can do is indicate a trend. At the moment, the trend seems to be Labour + LD support increasing and Tory support static (or falling).
My own feeling is that the most significant event of the campaign so far may be TB’s announcement (when launching Labour’s manifesto) that this would be his last election. Disgruntled voters now know for certain that within 4 years (and probably earlier), TB will be history. They don’t need to vote Tory to get rid of him.
David Clark in today’s “Guardian” (who has argued previously for anti-Labour tactical voting to reduce Blair’s majority) is now saying that, with a Brownite succession ensured, disillusioned ex-Labour voters should come back to the fold.
If I was a Labour candidate in a dicy marginal, I’d be stressing this to voters on the doorstep and plastering Gordon Brown’s face over my election leaflets. “Vote Blair, get Brown”, to quote an (abandoned) Tory election poster!
I do think we have to be a bit wary of labour moving up in the polls ostensibly because of a greater determination to vote among their supporters. It’s not difficult to vote. It must be remembered that a large number of people are not not voting to make some sort of point. It’s just come polling day they simply can’t be bothered. And no matter how much ‘enthusiasm’ they profess to have in advance, come polling day they once again will find some spurious reason to fail to make it to the ballot box again.
Polls right or wrong, they may be in percentage terms but they are indicating a trend Labour + Con - for all we know they might be right I doubt it my self, but us wonks do live in bubble of frantic politics, most of the population don’t give a stuff about politics
To 40 & 41. JB you sre right - the trend in some polls at least is towards Labour and that I believe. But what is the base starting point? Everybody seems to forget here what a rotten time Labour had on June 10 last year - the last time we had national elections. Pushed to third place in the locals when the Tories got 38% and finishing up with a mere 22% in the Euros.
On the turnout question I think there is evidence to show that a Labour 10/10 is not worth as much as a Lib Dem or Tory 10/10 - yet they are all counted the same by the pollsters.
33 & 34. To be very fair to ICM they do publish all the raw and adjusted data. It was because of this that I was able to find the figures for the article. The problem with the polls was well put by Nick Cohen in the Observer the Sunday before last. The media simply does not give them the same level of scrutiny that they give to other key political developments. So it’s left to nerds like me to go data mining
44, But Iraq was a big issue then, which it is not in this Election
Not forgetting the 500 lb gorilla in the corner …
Roughly 25% of people do not vote for their party of first choice (roughly split 50/50 between straight TV and ‘my vote doesn’t make a difference so I won’t bother this time’)
Polls tend to measure the party of principle…come voting day… who knows!
Re: 45 The main parties may not be talking about it but the war issue is still in the minds of many voters
http://politics.guardian.co.uk/election/story/0,15803,1462198,00.html
Or even
http://politics.guardian.co.uk/election/story/0,15803,1462198,00.html
47, But the kind of voter whoes mind it is on, will they be right or left, I suspect that they are to the left, not saying they are going to all vote laboutr but can you see them voting Tory?
if only the country could see the lib dems as they truely are…paxman makes mincemeat of kennedy but bbc decide to show it opposite coronation st……..dumbing down means blair wins again and kennedy escapes for most of the election with a free ride
RE 49 They have other choices including the don’t vote party. That can still damage Labour in some seats I assume
50, That’s the way it is, do you think even without competition the public would want to watch in any great numbers?
Do we have any confirmation of 29?
50, They should include some sort of phone in quiz with a cash prize that would get the viewing figures up. Handing out a free scratch card at polling stations would work wonders for turnout come to that.
53. the only thing I found about the new Populus poll is this:
http://www.timesonline.co.uk/article/0,,19809-1575245,00.html
55. Thanks I like to see it confirmed. If populus, MORI, ICM and Communicate have Lab over 40 it must be considered a noteworthy trend even if the phone/f2f pollsters are overestimating labour - presumably they’re not overstating more so there must have been an increase even if its from a much lower base than pollsters suggest…
It depends which version of mori is 40 32
Was the MORI poll certain to vote?
56. The Labour increase could also have come from a greater professed determination to vote, not a swing in voting intention. As discussed earlier in the tread, this could prove flakey.
Sky normally update their polls fairly quickly and there is nothing yet regarding a new MORI or Populus poll !
Really interesting comments about the validity of the different polls. My experience is that the most accurate poll is the one carried out privately for the Labour Party. Believe me when I say I don’t know what it is predicting for 5 May, but I do know that Tony and Gordon look pretty relaxed and not about to press the panic button. I suspect they have a pretty good idea of how much they think they will win by and that they will be able to live with it for another four years. By the way I ought to come clean - I am Sky Bet’s Political Consultant. As I am the ‘enemy’ I will understand it if you don’t want me to ever again sully the pages of politicalbetting.com.
Regards To All
59. You have to remember though that at a lot of people are already filling out their postal voting forms - so any trend now will still have an effect on the final result, even if the trend dissipates towards the end of the campaign.
61, If you are who you say you are sully on!
61, Who does the private polling, and has anyone seen any of them after the event?
BGR [54] - wasn’t there a suggestion from somewhere that you should be given a voucher worth perhaps £5 or £10, redeemable against Council tax, when you voted?
I don’t know whether it’s too early to mention it but Michael Howard’s campaign has been inexplicable. Having picked out his most photogenic lieutenants Osborne, Cameron etc) and hidden his others (Redwood) he has behaved as though he’s a one man party. Why?
66, He is the only one people know, but I don’t think they remember him otherwise he would be toast now
65 seems vaguely familiar. I prefer the prize idea myself, cheaper
Here we are in Reading East, a ‘target’ seat for a Conservative majority, and no canvassing so far from any of the parties (just some crumby leaflets). Is the fight purely restricted to denying Labour a (big) majority?
69. strange, considering the troubles on Labour Party in Reading East I expected tories to try harder and possibly making a better than average swing.
Seeing the number of PV cast, are any of the polling firms going to produce any figures for those who have already voted?
Somebody should ask them to?
Did I hear that Gorgeous George is intending take legal action to try to have PV declared invalid? (Presumably his team haven’t many of those ’sorted’).
Andrew @ 69, Don’t forget that Reading West had an electorate of 74,600 in 2001 (and is presumably larger still), so the chances of a canvassing visitation - while you’re actually in the house/flat - is statistically remote! I was offered a glass of sherry this evening by a delightful constituent: I may canvass there again!
Tony Blair defending sedgfield - a possible world Portillo moment - Paddy Power offered 1-3 on just a decrease in Blairs majority - difficult to get 30 % return in the stockmarket over 2 weeks so looks like a gift . The real action should be in TB to lose - who will give a price in size ?
http://www.bbc.co.uk/radio4/today/listenagain/
0740 Are the bookies predictions a better guide to election results than opinion polls? The chairman of internet pollster YouGov, Peter Kellner and the government’s adviser on betting, Professor Leighton Vaughan Williams discuss.
PS Ooops, of course I meant Reading East - should have had that glass!
66. Perhaps the Lib Dem “decapitation” plan is keeping some of his front bench team in their home constituencies ?
71- If GG will succeed to halt PV, what will happen?
Could the election be delayed?
71. I get the impression that the best Galloway hopes for is a separate count- that is the impression I felt he gave on Radio 4s PM programme.
I am hearing anecdotal evidence suggesting the Tories are quite pleased with their efforts nationally re: PV, particularly in target marginals, they might yet spring some major surprises?
69 - Depends where you are Andrew. No party does or can, canvass everywhere! We target our efforts. I know Reading East Tories are out every night canvassing.
Mike - has a mistake been made on Betting Odds - General Election - George Galloway Votes.It says
0-75000 6/4
7500-10000 6/4
10000 + 15/8
Surely the first should be 0 - 7500 or have I read it wrong?
62 - Postal Votes. Now I am know that some think that the result is already declared, but it is impossible that postal vote ballot papers are already being completed. Nominations dont close until 4 pm tomorrow - so it would be impossible to print the ballot papers till then. I believe you will find that most Returning Officers will not be sending out the first batch until Friday. So u can expect them to hit the mat on Saturday.
Maybe there are some County Council ones already delivered, but I doubt it.
71 - it would be illegal for returning officers to announce postal vote turnout at this election. Such information can only be given at all postal vote elections.
This getting harder to fathom. Surely those MORI figures cannot be certain to vote?
It is pretty clear the Tories have gone backwards; and it’s explicable given Howard plus blunders (manifesto, nastiness, tax and spend). But not sure why Lab is still going forward - don’t recall any posts to that effect on here. Perhaps just a lag in the firming up of the Lab vote, as some are slower than others to catch on to ‘Howard could be PM’?
PVs are not issued until Friday 22 April in our patch.
‘Only’ 291 out of 341? Strikes me as a lot higher than some earlier reports. Almost cast iron certainty many of the other 50 did not vote.
71 - The only way any candidate could get ANY postal votes made invalid would be thorugh an election petition, as was the case in Birmingham.
Back to the thread - I cannot believe that the number of votes cast for Labour this time will be more than in 2001. About a sixth lower sounds about right! The problem is what will be the turnout. The lack of enthusiasm for Howard and (I am afraid CK’s rather lack lustre efforts) may make 59% seem high.
BTW whether CK is good or not may make less difference than might be expected. Believe it or not I find voters enthused by Tuition Fees, Tax on over £100k earners and Local Income Tax!
Ref 74:::I posted on an earlier thread today that Prof Leighton Vaughn Williams is the subject of an interesting article in this weeks Economist.He is convinced that the spread betting market is a better guide to the election result than the polls,quoting U.S.A and Australia as recent examples.i.e.The market was always pionting to the winner..He believes that Labour will achieve a majority of 60.
Mike is right on this and I merely add to my comments yesterday. To me it’s clear these polls are duff, worse than that a whole load of tactically voters will vote on the basis of the duff polls and send Labour a far more severe message than they intended. Any methodology that seeks to draw straight lines on from the mood in 1997 are crazy. What happened in 1997 was that a large chunk of the electorate learned HOW to use their vote to bring about a result they wanted ie instead of voting for their normal preference. Watering down the clearly expressed VIs of people now will end up in a huge mess. Or so thinks I.
Sky now reporting Mori at 40 to 32 and Populus at 40 to 31. Sky had been using Mori at 39 to 35 which was a certain to vote figure so I assume, but cannot confirm, that 40 to 32 is also a certain to vote figure.
I’m feeling a bit over-exposed on my Lib Dem position. Ck hasn’t been that impressive recently, and the postal votes are making me nervous. Plus there seems to be something in the canvass returns. I’m looking for some Tory bankers on the seats markets to cover my losses. Any reccomendations?
Question on Tactical voting. The LibDem seats spread seems to assume that there will be tactical voting in both Con-LibDem and Lab-LibDem marginals. Presumably for the LibDems to make substantial gains there would need to be even more of a squeeze on the third party than in 2001 or 1997. Is there any evidence that this is actually happening ? How far can a third party vote be squeezed ? Are there really that many people out that who didn’t vote tactically last time but might now ? Also how squishy is the Tory vote in Lib-Lab marginals? Surely anyone who voted Tory last time is hardly going to switch now ?
David Kendrick - “Seeing the number of PV cast, are any of the polling firms going to produce any figures for those who have already voted?”
Not unless they enjoy being prosecuted they won’t - it counts as an exit poll and it’s illegal to publish it until after the polls close. After the Times published polling figures for areas where there was compulsory postal voting in the European Elections they were referred to the CPS. In the event, the CPS did not take further action and the Electoral Commission referred the issue to the British Polling Council. They have issued guidance stating that pollsters can include people with postal votes in their samples, but they are not allowed to publish separate figures for people with postal votes.
Here is the link to the MORI poll. Among all respondents, it’s 41/31/21 so the 8 point lead at 40 to 32 IS certain to vote figures. Not great for the Tories.
http://news.ft.com/cms/s/cd25bb68-b049-11d9-ab98-00000e2511c8.html
MORI’s figures are certain to vote - their website has the latest details on their general election webpage
MORI’s figures are certain to vote - their website has the latest details on their general election webpage
MORI and Populus polls are lagging I think. Duff.
91 - Paul M, it’s generally accepted that you don’t get squeeze below 10%, plus Conservatives wouldn’t vote for us in a month of Sundays. The evidence seems to be that they’re pulling votes away from us now.
Details about Populus
http://www.itnews.it/risorse/EuroNews,Zj0xMjAyOTE0
Howard in last position when asked voters who impressed the most so far in the campaign. Kennedy leading between swing voters (Howard in last place also between them).
Also a 5 point lead for Labour in the Independent which I assume is NOP, down 1% from last week’s 6% lead.
Can you be a bit more specific re: canvass returns Tabman. Is this in the last few days ?
I am not sure whether this is certain to vote or not.
MORI’s poll adds to recent signs that Labour has established a firm lead. It puts Labour on 41 points, the Conservatives on 31 and the Liberal Democrats on 21 among all voters naming a party.
Among all voters naming a party suggests it is not.
This confirmation of Populus and MORI makes Austin Powers in msg. 29 very well informed and about 2 hours ahead of the media embargo. I feared we were being ramped again!!
nop poll - lab 37 (-1), con 32(u/c), lib 21 (u/c)
Addition to 100
I think the poll is not certain to vote. Thanks to a link to an italian site. I read this.
The poll also showed that Conservative supporters are still more likely than Labour supporters to say they are certain to vote, by 72 to 63 per cent. This advantage makes Labour’s lead two points lower than it would otherwise be. The poll implies a turnout of 63 per cent, up from 59 per cent four years ago.
90 + 97 - Tabman - is that really you?Whats the name of the book shop near your place of work?
latest you gov poll for daily telegraph puts Labour a single poing ahead of the Tories.
102. the NOP Poll is not that bad for the conservatives.
Vino - Foleys. Yes, it is me.
these polls are all up and down and the only 1 i think is right is You Gov.
97. According to Richard Willis’ canvassing in S&C, there are so many switchers from Lab to Con, the Labour vote must be heading for well below 10%.
Also bear in mind that there were a large number of abstainers (mostly non-Tories) in 2001. One of them voting for the Lib Dems (in a Con/LD marginal) is as good as a Labour switcher.
About NOP Polls:
http://icessex.icnetwork.co.uk/news/tm_objectid=15416635&method=full&siteid=50102&headline=brown-could-boost-labour-victory-name_page.html
With Brown as leader Labour will poll 11% higher: 23% of Libdems will switch and 16% of tories too!
Re 91 on how far can the 3rd vote be squeezed.
My view as a tactical voter in 2001 is that it has lost its shine for a lot of people. I moved from Labour to Lim Dem to try an oust the Tory. But the Tory vote held firm and a good campaign by them lead to an increased maj. So do I vote tactically this time? My view is that the Lid dems can’t win the seat and that the Labour vote has been squeezed as far as it can go. Thus I shall be swing back to Labour.
I’m sure that for many Labour Tactical voters they will be thinking the same in those constituencies where the Lib Dems did not win in 2001. Where it paid dividends I would have thought that the tactical vote would stay firm. Look at Dorset to se how my old mate Billy Bragg is still pushing the anti-tory line.
On the basis of the above, I don’t see any Lib Dem gains from Tory this election, but they will hold on to what they have got.
107 - Tabman - sorry to doubt you,but the tone of your posts seem down which I do not associate with you.
Alan
The same is happening in Welwyn Hatfield as i have been canvassing in that area.
Labour voters are switching to Conservatives.
110 With Brown as leader Labour will poll 11% higher: 23% of Libdems will switch and 16% of tories too!
Andrea, it is well known that if u ask a hypothetical question, the result will always be different than if it was not the case. As people know that Gordon Brown is not the Leader, he will always score more.
Then to You Gov poll - sorry but with all this yo yoing about it just made me laugh !
109 - I would have to agree now that Labour are heading for less than 10%. Having just canvassed a Council estate type area tonight I have advised my neighbour candidate to do some test canvassing in his less good areas. We had another great night with posters given out and a big increase in our pledge totals. I will have a poster in the newsagent on the estate and an 8′ x 4′ poster going up shortly right on the main road! I had 8 canvassers with me tonight who were amazed at how good the reaction was. We hardly found a Lib Dem all night and there were NO repeat NO, Lib Dem posters anywhere on the estate. There are now at least 20 of mine up!!!
112 - some impressive Clarke boards out in Rushcliffe. A lot of don’t knows during canvassing, which seems to be shorthand for not you (so expereinced canvassers tell me). Plus, having been through the 80s I know how effective the Tory machine can be
I got sweopt away in a tide of optimism and over-egged my LD position. I need to balance it with something and from what I’m hearing a few TOry bankers would be good.
113 - Darren, if that is the case (and I dont disbelieve u), explain the polls ????????????? Because I cant
“As people know that Gordon Brown is not the Leader, he will always score more”
I don’t think that if Clare Short will led Labour, she will score more.
There were also a lot of Tory abstainers in 2001. Getting them back on board could be crucial in a lot of marginal seats. The classic example is Guildford where the Tory vote went down by just over 4000 and the Lib Dem vote only rose a couple of hundred.
Ste
At the last election we the tories in welwyn hatfield actually did better than the national vote.
Our candidate has been working the constituancy for the last 3 years.
Ste
Also u can tell when out canvassing if you are being told the truth about the way people intend voting.
I was in a strong Labour area in welhat and even there people are either not voting or switching from labour to conservative.
Its funny that in SW London the council estates (or rather the bought up one’s which most are) are often good territory for the Tories. In Wandsworth they amazingly manage to have councillors (by fairly comfortable margins) all over the Alton estate, a lot of which probably isn’t bought up.
122 - why is that funny?
A Lib Dem canvasser in Richmond Park tonight told me they were hoping that UKIP were going to have an impact on the Tory vote. Not a statement which suggested huge confidence.
UKIP could hit the Tory vote in Twickenham where a popular ex Tory councillor is standing, but v.unlikely in Richmond. UKIP will poll less than a 1000 and by no means all of that will be Tory.
41,31,21 is very interesting- grim reading for CCO
BTW I have come to the conclusion that the “marginal poll” by ICM is fairly cludge-y- with only an average of 30 respondents per seat, I think the margin of error is too large for meaningful data. Looking at larger data sets: In one marginal seat I have been into, there has been an over 5% swing Con-Lib Dem. In another the swing is much larger. So in answer to [91], yes there is still some squishiness in the Tory vote- reason is MH and the fact that so many of his front bench have to defend their seats so tightly emphasises his personality.
Could do with a new Scottish poll - if the Tories are really doing as badly as the polls suggest my sell on the SNP isn’t looking good.
23: ICM full data (use the link to the PDF doc on each ICM poll on their web site to see this) shows that C1’s and C2’s are difficult to find using phone polling, so they “weight” to correct this low base number upwards.
Good news for Lynton then as it’s C1/C2s where the Conservatives are doing best.
121 - I note what u say, and know that to be the case, but still tell me why the polls are the way they are. Soon the public will say that they dont believe them
Dave at 73: I think it’s more likely that Tony Blair will be kidnapped by aliens than defeated in Sedgfield. In fact I think you’ll find that in a few years he’ll retire ‘not out’, without ever having lost a major vote in General Elections, his own seat, Parliament or the European referendum (if the French don’t torpedo it first).
Paul M at 91: I think there is scope for new TVs where a seat wasn’t very close last time but has become so. However, the polls so far have tended to suggest a mild tendency for Labour to strengthen rather than weaken in places where they’re third, presumably the mirror image of tactical unwind, making the LibDem decapitation strategy hard to achieve.
There is as always a fair number of people switching back and forth anyway, sometimes for bizarre reasons. I had a Tory->Labour switcher tonight because she works in Sainsbury and I paid at her checkout last week (so…?), and a Labour->Tory switcher because he’s unhappy that I’ve not stopped his neighbour banging the door (er…). A more serious reason why the Tories are losing some votes is the perception that their programme is a grumble-sheet rather than a serious plan for government: this irritates some of the Daily Telegraph type Tories.
Ste
People are already telling me they don’t believe the polls.
Andy
In your area are you targeting just v5,6,8 & 9’s in voter vault?
Nick - he can’t be “not out” … he lost Beaconsfield didn’t he?
Whether these polls are right or wrong might not matter. The effect will be to sap Conservative morale and possibly lead to internal feuding. It might be difficult to keep the lid on if activists think their work is being wasted on a useless campaign. How long will Maurice Saatchi and Lynton Crosby work to-gether for example? There was a program on Channel 4 about how Maggie started using advertising and focus groups in ‘79 but the others didn’t catch up till ‘87. And to this day there isn’t any better political advertising than the stuff the Saatchi brothers did for the Tories in the ’80’s
Darren - I am agent for a majority Labour seat so we are spending most of our time helping in target seats. Something that’s never really happened before. The campaign here is 100% about winning council seats in 2006
131 - Darren I would have thought that with annual electon in WH you have infomation on more than 60% of the electorate, whcih would mean that u would not need to use voter vault.
125 - that’s what i thought and why i got the impression of a lack of confidence - seemed a bit like straw clutching.
O/T - has anyone noticed a considerable difference in the way the media are covering this election compared to 2001. It may be related to the party strategies but aside from the news conferences there seems to be very little contact with the politicians and the campaign on the ground at all.
130 - not to mention leading Labour to third place in last year’s Euros with the lowest share of the vote this side of Ramsey MacDonald…
130 - That would depend on whether you count his by-election defeat, in 1981 I think, before winning Sedgefield in the 1983 general election.
Tabman [116] - what do you mean by Tory bankers? Individual seats? I thought there was a market on Conservative gains in London which might meet your needs… there’s a thread on the Other Side which might help too (it’s in “General Discussion”)…
I do not think there is A Yougov Telegraph poll tomorrow. Not covered on Sky review of the papers or on their poll tracker
133. “possibly lead to internal feuding” “How long will Maurice Saatchi and Lynton Crosby work to-gether for example?”
They could stay together for 3 weeks. If Blair and Brown managed to stay together for all these years, Saatchi and Crosby could stay for 3 weeks.
After the election (if the polls are right), it’ll be fun to see the conservatives internal debate.
IA - yes individual seats. Thanks, I had a look but one poster is quoting Orpington as a dead cert and a possible loss!
phillip
it is politicalnews.co.uk
u can get 1-8 on Tories taking welwyn hatfield from labour
Blair took another thrashing at the hands on Nick Robinson tonight, this time with regard to Mr Howard smugly signing the Labour petition- very effective. Any electoral advantage gained?
133. Interestingly on that point Martha Kearny on newsnight just mentioned ’senior conservatives’ in tomorrow’s paper complaining about the immigration strategy…
Seems to be valid. Con 36% Lab 35% LD 23% Other 6%
Source: YouGov / Daily Telegraph
Methodology: Online interviews to 1,483 British adults, conducted from Apr. 14 to Apr. 16, 2005. No margin of error was provided.
Oh hang on, misread Apr. 16, thought it said 18. Oops!
115 - Richard, can you point me to a bookie who’ll offer odds on you winning in S&C? ‘Cos I can’t see it on the markets anywhere. That’s going to be one hell of a swing if you can pull it off. Or maybe you’re just kidding yourself?
132, 138: True!
In looking at the polls, note that Populus is supposed (according to our guru Wells, who I reckon understands these things as well as any of us) to be relatively helpful to the Tories, while NOP is spuposed to be better for Labour. What I think is causing the crossover (apart from random variation) is the hardening of Labour determination to vote, which will be having other indirect effects on false recall etc.
The hardening is certainly real: it is partly due to the general increase in interest as the election approaches, partly to that helpful poll a few weeks back that showed the Tories 5% ahead, but mainly reaction to the Tory campaign: I’ve lost count of how many voters have told me “I was hesitating because of Iraq, but Howard has made me remember what they’re like”. Had a lifelong LibDem offer to cancel her holiday tonight to stay at home to vote Labour for the first time - part personal vote, part “we must stop the barbarians”. (Fortunately she can still get a PV!)
The question is whether part 2 of Palmer’s Paradox will now kick in and the determination will weaken if the Tories are seen to be hopelessly behind. I think it will, a bit, but on the whole people now know what they plan to do.
Posters seem a matter of local strategy - there are no window posters whatever in Broxtowe for the other Parliamentary candidates (compared with numerous Labour ones), so far as I know, and that certainly doesn’t reflect an absence of support for them. I think that parties tend to judge that if they can’t be competitive in a poster war it’s better not to fight it at all. Or perhaps they’ll try to come with a coordinated late rush - we’ll see.
To hear Nick talk there will be no swing to the Tories. To hear Richard talk the LDs will be lucky to get a third of the vote in Sutton and Cheam. Then we had a Lib Dem saying that the Tory vote was squishy and that there was a 5% swing to Lib Dems in Tory seats. Enough guys! This is serious candidateitis. As things stand Labour will lose quite a few seats to the Tories including Welwyn but there’s no sign of the sort of swing which would cause a hung Parliament. Remember that Blair won’t get much more than a third of the vote. On a 60% turnout that means he’ll get a comfortable majority with around 20% of the electorate. A heck of a lot of people may be sick of Blair but not many need to continue voting for him to win the election. As for the pollsters I think we should recognize their difficulties. People do forget who they voted for last time so you need to adjust but how and how much? There is a problem of shy Labour voters according to most pollsters but shy Tory voters according to You Gov. Finally there’s the crucial question of differential turnout which is crucial. Mike says Labour 10s aren’t as certain as Tory tens. Really? And if so how do you take account of it. The point is the raw figures aren’t accurate but noone has a clue which are the right adjustments. In the light of all this I will continue to look at how contests are shaping up locally in as much as I can get the relevant information.
The one opinion poll where one can make comparisons over a long period is MORI’s “all those naming a party” poll.
For the record - at at this stage in previous campaigns the Labour lead was
1992 - 3%
1997 - 17%
2001 - 26%
On the vexed question of how well are the Tories really doing, which seems to be the main bone of contention at themoment, the corresponding percentages were
1992 - 38%
1997 - 32/34% (depending on which poll you choose)
2001 - 28%
Those figures should be seen as “who do you support” and includes armchair fans.
Figures for other parties/ previous years at
http://www.mori.com/polls/trends/voting-all-trends.shtml#1986
If I was a Tory, I would seize on the point that each time the real result was much more favourable to the Conservatives.
If I was Labour/LibDem, I would be hoping that the Conservative switchers on canvass returns were just in comparison with the exceptionally low point of 2001, and not yet enough to bring them back to 1997 levels of support.
Tabman [142] - if you’re looking at those seats where the Tories are favourites, I think Labour have written off Hornchurch, Bexleyheath & Crayford and Enfield North; also Hammersmith & Fulham where the sitting MP pulled out for health reasons last month; most people think the Tories will win Ilford North and I would add Croydon Central, in both of these Labour rely on the “council estate” vote which seems to be their Achilles heel.
146. Both Guardian and Times are running grumbling Tory complain at Howard headlines - but then they would wouldn’t they…?
149 - I dont know of one (sadly)! But who last time would have given great odds on Romford being taken, or Taunton. I honestly dont yet know if there will be enough of a swing to win the seat. It LOOKS like it and the returns are very good. But it is quite possible to get a very good swing and still not quite take the seat. I am not so naive as to go to bed each night assuming victory. But I do honestly believe that there is something significant happening here and it is reflected in other Lib Dem seats in this area.
As I indicated before I have on good authority that the Lib Dems are now calling in resource to defend Richmond. An informant locally also tells me that the Libs here are struggling with a lack of deliverers (as I said on here some time ago when I found paid for dleiverers being used).
We have canvass returns back to 1999 at least to compare to, so I know where swings are happening. Plus I have conducted a rolling survey since I was selected and it is interesting to see the movement in the last 18 months. As recently as 6 months ago there was no sign of a swing at all. Now it is very noticeable.
Tabman - “The first door I knocked on, I was told to burn in hell. I felt things could only get better after that ”
Quote by
Lucille Nicholson, Tory candidate in Easington, Co Durham, where Labour is defending a 21,949 majority .
[130] Well Nick, one Labour mate did suggest to me once that Tony Blair HAD been kidnapped by aliens and replaced by a Tory Doppleganger..
Think you are right though- this Victor Meldrew stuff of Howard’s has not played well with quite a few Daily Telegraph types (I got a blazered Rotarian switching over the weekend- “That bloody man Howard is the absolute end”)
Quick update. For no good reason (other than a hunch of mine from having been reading this site) we canvassed a !very! run down housing association estate tonight that none of us have ever canvassed before.
There were two surprises. 1) Labour vote is totally divided - either very strong or non-existant, i.e. voters here have either decided they will stick with Labour through anything or they have deserted them and seem to have bypassed the Lib dems and come directly to us and 2) the reason is over tax, and not -as I was anticipating- immigration.
Don’t get me wrong, we were still very much the minority party in this area but it’s the low level of support for Lib dems that is -for me- reassuring, suggesting that they won’t be tactical voting in large numbers.
We have a significant number of voters in an area I would have guessed we would have had none.
We are going to blitz the area for the rest of the week and see if it sustains.
Nick. Could almost agree with your remarks on Blair’s holding his bat - except that you think he will win the European Referendum? Duh? How so? How can he possibly win this? When the vote will give everyone a chance to kick him without letting the Tories in? When two thirds of the electorate are ideologically opposed? Whenmost of the press will be viscerally against? Do you know something I don’t? Or are you just whistling in the dark. If you are, I must say it’s a very quiet whistle, and it’s very dark…
So all the polls are wrong now are they? - MH is actually sweeping to power on the back of overwhelming popularity.
I think the spin factor has gone into overdrive tonight. Although a constant refrain of Mike S (seized on by many on the site) is that ‘it is in the polls DNA to overstate Labour’. A constant refrain could have been that ‘the polls are normally about right with the Tories’, but have usually understated the Liberal Democrats.
The underlying figures for the Tories are awful. Michael Howard is not popular, they are behind on key issues, the haven’t convinced people that their tax plans are workable. The only area where they appear to be scoring well is immigration - but this does not appear to be the ‘clincher’.
Polls are rarely ‘bang on’ - but they are not designed to be (anyone who knows anything about sampling theory knows that they cannot be), but they do show trends. And the trends aren’t looking good for the Conservative Party.
143 Is this correct I wonder? The poll in Mondays Telegraph had sample taken up to 17/4. The poll you are reporting for the Telegraph tomorrow has sample ending on 16/4.
A lot of the “story from the street” comments on here are starting to smell pretty fishy. It’s clear this has become an important forum for anecdotal reading, and that means canvass stories here are unlikely to be 100% honest.
I very much hope that on May 6th our regulars will come back here to explain how they were so wrong - because they cant all be right…
143 161 Figures and sample size are sam for the YouGov poll in S Times.
162 - I really do resent the constant insinuations on here that “canvassing stories are fishy”. WHat possible benefit do you think we gain from taking the trouble to post on here? If you really do persist with this line of attack I will just carry on merrily getting posters up, finding new pledges at the rate I have, and not bother the readers of this august site with my experiences. YOu are living in fantasy land if you think that Marcus and I can be bothered to invent canvassing stories together and post them on here.
But if you chose to dream on and meanwhile we gain support - so be it!
Will someone look at my post at 80 and please tell me if the bookmakers have made a mistake? Thanks