Monday polls

It’s Monday, so we have the usual rush of polls – the daily YouGov, twice-weekly Populus, weekly Ashcroft and monthly ICM. I’ll update this post as they come in, and do a round up at the end of the day.

Populus have topline figures of CON 31%, LAB 33%, LDEM 10%, UKIP 15%, GRN 5%. Tables are here.

ICM for the Guardian have figures of CON 36%(+6), LAB 32%(-1), LDEM 10%(-1), UKIP 9%(-2), GRN 7%(-2). The six point jump for the Conservatives looks particularly sharp but usual caveats apply. While it tends to be the polls that show unusual results or big changes that get all the attention, they are actually the ones we should be most dubious about. If there has been a genuine surge in Conservative support, then we’ll see it across all the pollsters, and other polls so far this month have shown things pretty stable.

The weekly Ashcroft poll has topline figures of CON 30%(-4), LAB 31%(nc), LDEM 9%(nc), UKIP 16%(+2), GRN 8%(+2) (full details here). A one point Labour lead, and changes in the opposite direction to those in the ICM poll – I think we can be fairly confident that what we are seeing with these two is just random noise, the back and forth of normal sample error.

UPDATE: The last of today’s four polls, YouGov’s daily poll for the Sun has topline figures of CON 32%, LAB 32%, LD 6%, UKIP 16%, GRN 8%. After a couple of three point leads at the tail end of last week, it’s back to neck-and-neck from YouGov. Putting all four polls together I think it’s pretty much business as usual. One neck-and-neck, one Tory lead, two Labour leads. There is nothing here that’s incompatible with the steady picture we’ve had for the last six weeks: an extremely close race between Labour and Conservative, with Labour holding onto a tiny lead.


552 Responses to “Monday polls”

1 2 3 12
  1. As said on previous thread – if there has been a 6 point jump in Tory support, I can’t see why.

  2. Did ICM just post Peter Kellners prediction by mistake?

  3. Incredible Comedy Metrics

  4. Incurably Crazy Measurements.

  5. Or, for the right-leaning among us, Infallible, Credible Mindreaders.

  6. Tories squeeze all other parties – a very neat trick after the week we have just seen.

  7. Does seem a bit odd.

    Lib Dems way too high ( ^ __ – )

  8. “I’m cr.pping myself” – (Miliband)

    Whilst trying to keep party allegiance out of this, it doesn’t feel right – but I would like it to!

    Being serious though, is there any theory which supports the more successful a party is in the polls, it becomes a self fulfilling prophesy, or indeed at the other end, if a party starts doing badly, a vicious circle ensues as people desert a sinking ship?

    I am not linking this question to this ICM poll, but ask out of general curiosity.

  9. So the Conservatives pick up two points from the Greens? Hmmmmm…

  10. I’m not sure. I think there’s a view that if it looks close, one party’s supporters may come out of the woodwork to stop The Worst (in their view) from occurring.

  11. admittedly an odd poll… tories up 6 since last month hasn’t even been repeated by Ashcroft. i think 36 is the joint highest tory VI in any poll since March 2012.

  12. That ICM poll is wacky by any standard. Look at the issues polling:

    https://pbs.twimg.com/media/B9-CV2QIUAAhoSS.png

    So we have a very Tory sample that’s most concerned about… the NHS and cost of living. Okaaaaay.

    Possibly we can give the Tories “Jobs”, but those are very strange numbers.

  13. AberdeenAngus

    I think it depends on the margins. if it’s a small gap that might make voters on both sides more likely to come out. If there’s a pretty big lead then I think it’s more likely that voters on the losing side willbe less likely to bother voting.

  14. What we really need are the tables.

  15. Looking over ICM’s past results, this is a very high Tory score for them. 32% is within the normal realm of crap ICM Labour scores- 33% is typical from them- but the last time the Tories were on 36% was July 2013. They only had one 35% in 2014.

  16. Pink Van effect ?

    Keep receipts from your gardener effect ?

    Complete cobblers?

  17. Forget about the accuracy of this ICM poll, it will become the story. The narrative will be “Tories Surging Ahead in Polls!”

    And if the narrative moves onto that terrain then it can only be good news for the Tories.

    Cue to next question put to a Labour spokesman. “You say that, but you’re four points behind in the polls….”

  18. Tom Clark in the Guardian seems to at least acknowledge this time the existence of other polls over the weekend but his answer to the unusual +6 for the Tories is:

    “But having come closest to predicting three of the last four general elections, ICM’s regular phone poll for the Guardian is seen as the gold standard, and so Monday’s result will bring the Tories particular cheer.”

    I am really not sure whether he had a straight face when he wrote the article as he mentions the lead despite pressure over Tory donors and HSBC. Given all that and the other polls, surely it would be more probable that this poll is a little bit off! Still the poll was “ICM for the Guardian” so I presume one always backs one’s own horse.

    enigmajx

  19. Aberdeenangus

    There’s theories abound as to where the polls are going.

    So far the only statistally significant predictor is a reversion to prior election figures (which is a fairly weak effect so can only be said “It happened more often than not”).

    The “jump to the winning side” probably exists to some extent, whether it’s greater or not than the “don’t let that lot win effect”.

    It shouldn’t be hard to see if a lead in the polls leads to an increase in the lead in the polls in the leadup to elections. Gut feeling is there won’t be enough data to make it statistically significant.

    Alternatively one could look for increases in the polls (as opposed to the polls themselves) creating either a momentum or regression to mean effect.

    I suspect either of these effects will only be noticable very close to the election, within a week or so.

  20. Perhaps just reversion after a series of above-average Labour results. A 1-2 point Labour lead still seems to be the trend.

    (Except that Labour are obviously still in the 36-40% range, barring outliers.)

  21. @ Spearmint

    According to the commissioning newspaper ICM’s telephone poll is THE gold standard (which failed between every country between 1932-1935 by the way) of polling. There is some words about swing back and bread and butter of elections (very apt as food prices are coming down, especially in Morrison’s. They don’t say, but as it is a telephone poll, so it doesn’t even reflect the shy Tory effect.

    Now, Labour said exactly 6 things over the last week, which resulted in an effect of 3% Labour, 20% Green, etc move to Conservatives. Ergo, it is Labour that squeezes smaller parties, and blows them to the Conservatives, without significant effects on their own VI. Now in Labour HQ there are debates how to repeat it in Scotland.

    That’s done. Analysis is solid and supported by data.

  22. Typing posts on mobile phones is worse than doing the same on a tablet …. especially on a train. Apologies for the numerous errors above.

  23. Bit of a laugh though-you must admit :-)

  24. Remember the 1987 GE. There was a poll that had the Tory lead at only 4%

    “A week before election day, on what was to become known as “Wobbly Thursday”, a rogue poll for The Daily Telegraph showed the Tory lead down to 4%”

    We all know what happened a week later!

    Given the huge increase in the number of polls since then I expect a few more rouge polls. Of course it could be the first of a trend but I doubt it – the Tories had a back week last week imho.

  25. @ Bill Patrick

    On what basis can you say, ” Labour are obviously still in the 36-40% range, barring outliers” ? they have been >35% once in last 25 polls I think, or am I missing something?

  26. Forget about the accuracy of this ICM poll, it will become the story. The narrative will be “Tories Surging Ahead in Polls!”

    The tory press will hype this no end. good for them. labour will panic…

    but does it make a tory hold in ipswich, hastings or hove any more likely…hence i prefer ground up analysis.

  27. Crossbat11
    “The narrative will be “Tories Surging Ahead in Polls!”

    Yep, already been mentioned (with graphics just to reinforce the point) on Sky news.

    To be fair, Sophy Ridge did point out at the end that none of the gain has come from Labour & that there are other polls out point to a rather different picture.

    She added that this last week has been good for Labour but that this poll is something that could worry them.

    (Yes Sophy, because nobody in Labour understands anything about polling & outliers do they…..bangs head on desk)

  28. ‘Remember the 1987 GE. There was a poll that had the Tory lead at only 4%’

    I recall the 1997 election . A week before Polling Day ICM showed the Labour lead falling to 5% leading to fears of a repeat of 1992 in Labour ranks. Their final poll put Labour’s lead back at 10% – still 3% lower than the actual outcome.

  29. @Crossbat11

    “Tories Surging Ahead in Polls!”
    “You say that, but you’re four points behind in the polls….”

    Guaranteed to spark outrage not only from UKPR regulars but also grammar pedants. Is there still a rule that there have to be at least two of something to justify putting an extra s on the end?

  30. Spearmint

    Is 31% high for “cares about the NHS”? Did they poll the same issues in their last poll?

    Seems a bit strange to dress “the economy ” up as “jobs, prices and wages” yes they do amount to the same thing. I suspect the choice of wording creates a less significant effect than the word “economy” particularly to Tory ears who move to NHS once they don’t hear the word “Economy” as they were expecting.

    House effect or not, the differences poll to poll (assuming they used the same list of issues) might be informative.

    In isolation the poll tells us little (as with any poll) combined with the few recent labour leads I’d suggest it’s more a case of a few polls varying by more than average and we are still neck and neck. It’s not hugely important if this poll is outside the 95% confidence level, it might be a Con 33 and lie just inside or lie just outside, it’s pretty much impossible this poll is from a Con 30 state of the world.

  31. It can’t be right because, as has been explained by experts here, it is in defiance of all psephological rules & indeed those of the Universe itself for Cons to get more that 35%.

    As someone has just observed on pb, not only is this ICM poll an outlier, the GE itself could be too.

  32. Colin

    Clearly.

    The meaning of outlier means take 3 points off them, add it to us and I’m still unhappy!

  33. @ Colin

    GEs are outliers as the next GE often shows this. YouGov polls 4 times a week, ICM once a month, the HoC once in every 5 Yeats (they use to do it more frequently). The validity of outliers doesn’t depend of frequency (and sample size).

  34. There were two rogue ICM polls for Labour a couple of weeks before the 1997 election as well (one much too low, another much too high in terms of lead).

    On the subject of Peter Kellner, this might be a useful commentary which applies to this poll too.

    https://yougov.co.uk/news/2013/07/16/are-labour-and-tories-really-level-pegging/

    To summarise in a sentence, there is a turnout filter, plus a “shy Tory” filter. Seeing the raw crosstabs here might be enlightening.

    And before anyone accuses me of wishful thinking, I don’t think for one moment the Conservatives were really as low as 28% in December or 30% in January either.

  35. Aberdeen Angus,

    It’s a reference back to the period when people were claiming that Labour were guaranteed a result in the 36-40% range. After all, where else would the Lib Dem voters go?

  36. I was very undecided between the Greens and Tories. I then thought about the Queen and Phil the Greek, cramped into 563 Steve Biko Buildings Poplar and decided to go with the Tories. Clearly many others made the same decision.

  37. @ Alan,

    They have changes from their last poll, so I assume they ask about the same issues each time.

    NHS static, “Jobs, Prices & Wages” up, Immigration down, Education up.

    Labour ought to be pretty happy with that issue polling, actually, if it didn’t come coupled with that topline VI. Although this seems consistent with the overall trend that public concern about the NHS is increasing without boosting Labour.

    @ Colin,

    As someone has just observed on pb, not only is this ICM poll an outlier, the GE itself could be too.

    Lol!

  38. All the papers/blogs are going mad about this poll. What’s the matter with these people – Lab ahead 2/3 points last few days and then this outlier.
    truth is Lab probably ahead 2 points – I remenber a couple of ICM polls ago the blues on 28!!!
    Calm down all – no wonder non politicos get confused with polls when this is how they are reported….

  39. GRAHAM>
    That ICM poll in 1997 led me to shout at my young children to hurry up and get in the car: I had to get them to school, and then to College to teach. Memories.

    An Oxygen and Magnesium moment OMG

  40. What were the dates of polling (last Yougov was Thur-Fri)?

  41. @ Jasper22

    “All the papers/blogs are going mad about this poll. What’s the matter with these people ”

    I don’t know, but I guess it’s more about the 6% than anything else. How many more words can be written about polldrums (TM Amber)?

    If it had been for any other party, you would see the same indulgence.

  42. Certainly a weird day…..Daily Telegraph now saying Ant and Dec don’t rate Ed Miliband.

    That’s the end of him then !!!!!

    Now Ashcroft will come out with Lab 6 points ahead and we’ll know it’s just one of those days..

  43. @ JP

    13-15 Feb

  44. OK, lets accept that until we have corroborating data in the form of other polls, this is an outlier. We have 57 minutes I think until Lord Ashcroft’s poll. What will Lord Ashcroft’s’ poll have to show to give some credence to the ICM poll?

  45. I think it is ludicrous to talk of swing back and I will use actual voting numbers to make my point:

    LD 2010 Election 6,838,824
    LD 2014 European 1,087,632
    LD 10% 2010 2,972,532

    UKIP 2010 Election 919,546
    UKIP 2014 EU 4,352,051
    UKIP 15% 2010 4,453,707

    Green 2010 285,616
    Green 2014 European 1,244,975
    Green 7% 2010 2,078,397

    From this I observe:

    1. at 10% support there are 3,864,292 former LD votes in play if they vote

    2. All those who voted UKIP in the 2014 European election could vote UKIP again and still not match 15% of the 2010 GE. It is therefore quite possible that no swingback from UKIP is going on.

    3. For the Green party to achieve 7% their actual vote will have to increase by 833,422 votes over the 2014 European election, in the number of people who vote stayed the same as 2010.

    4. i question the veracity of polls that use 2010 as the base line, when voting patterns have changed so radically since then. Just look at the change in actual election numbers for LD, UKIP and Green between 2010 and 2014, rembering that unlike 2010 both UKIP and Green will likely have or come close to having fiull slates.

  46. In polling news that may actually reflect reality – have the Lib Dems hit their floor and begun climbing again?

  47. LASZLO

    @” The validity of outliers doesn’t depend of frequency (and sample size).

    So what’s the MOE at a UK GE ?

  48. @AberdeenAngus

    I’d say we need more than an Ascroft Con lead to make us believe there’d been a blue surge. This is one poll and that would be just two.

    As I said last night, this is the weekly rhythm of polling now. Monday is spurious Tory surge day. The rest of the week is factitious Labour recovery. Week on week, nothing has changed. The “Postage Poll Median” is still 2 points in Labours favour, and will be even if Ashcroft and YG show the same as ICM. At least here we should retain some equanimity.

  49. I do not see any sign of panic here just people quite rightly questioning the
    veracity of a poll which Anthony suggests we should be dubious about until it
    is confirmed by others.

  50. I’ve just thought of another ICM acronym:

    Inherent Conservative Methodology.

    :-)

    We need Angus Reid to start polling again and then the fun would really begin.

1 2 3 12