Yes, I know the monthly poll round-up here is sacred. But before we all chow down to our weekend Smeargate extravaganza, let’s briefly consider the latest exploits of BPIX.
As regular poll-watchers will know, we don’t incorporate BPIX’s figures into our monthly round-up at LDV because, uniquely among pollsters, their figures and filters are not published for scrutiny. It should be noted that BPIX commission Yougov to carry out the actual polling, but then apply their own undisclosed filters and methods to interpret the findings. BPIX only works for the Mail.
Last time they polled, in November, the Lib Dems scored a pretty awful 13%. Our range across all polls that month was 12-19%, so the BPIX result was merely at the lower end of what was confirmed, albeit in a volatile manner, elsewhere. The same poll put the Conservatives on 45%, in a monthly range of 37-45%.
Now, this weekend another poll was published by Marketing Sciences (ICM in disguise apparently) which put us on a healthy 21% but – far more importantly – only 5 points behind Labour. So, even with all caveats about BPIX in place, I was watching out for their final figures to see if this trend of a closing gap to Labour was confirmed, no matter what the actual figures.
BPIX’s poll in today’s Mail puts the Conservatives on 45%, Labour on 26%, Lib Dems on, er… To Be Confirmed? It’s a poll, not a Facebook event. Why are we still waiting, nearly twelve hours after the figures for the other two were released, for our figures? Has anyone spotted them elsewhere? We can’t look them up for ourselves, of course, because BPIX don’t publish detailed figures.
Cynical hat on. Maybe they were deemed too good?
Follow @libdemvoice.org on Bluesky
Like us on Facebook
Subscribe to our feed
Sign-up for our daily email digest
Lochee Park - 1973 #dundeewestend
Urgent action needed on Welsh rivers
Llandow, 1950: The worst air disaster the world had seen
Electoral Calculus's local election predictions
Are memories like a row of telegraph poles?
tom arms
Simon R
Alan Jelfs
Peter Hirst
13 Comments
Tim Montgomerie is claiming BPIX put us on 17%, here. Dunno where he gets that from, though.
Does that count as too good in the Mail’s eyes?
I have worked it out to be something similar too on the basis that ‘Others’ will have scored 12 per cent.
What does everyone else think?
It certainly counts as let’s-skip-that-one, I should think. They’re only happy when we’re in the low teens. I wonder where Tim got that figure from? Still not up on UK Polling. Called someone at BPIX, presumably.
I’m glad you drew my attention to that, though, as I see Tim is taking the most recent YouGov poll as his baseline, rather than the previous BPIX. Slightly questionable, I’d have thought, since we have no proof at all that the data is treated in the same way.
Possibly it’s relevant that comparing this BPIX result to the last BPIX result would show no net change in the Conservative rating, even after Smeargate, Porngate, Sistergate etc? Ooh, I am awful.
Not upto your usual standards Alix. You are you just factually inaccurate to say that we were still waiting over 12 hours later for the LD BPIX rating to be published. It was on politicalbetting.com within about an hour of the original Mail article yesterday evening and came via of all biased sources Tim Montgomerie. Your ( very poorly drafted ?) article at least suggests that this ( non existant) 12 hour delay was some how BPIX’s fault. Do you have any evidence of this ? Isn’t the simplier explaination for their non appearance in the Mail article that the Mail didn’t choose to include them ?
Overall what struck me about this post is the shrill, slightly paranoid tone that some how dark forces where conspiring to prevent news of a LD polling surge reaching the public. Last night saw two polling results. Con 45/43 Lab 26/26 and LD 17/21
The Lab/Con figures look remarkably consistant to me. The 4 point gap between the LD figures ( and remember that standard margin of error is 3% ) are simply what you get between internet and telephone pollsters but particualrly the way ICM structures its questions. There are reasons to argue that ts ICM’s question structure that prompts slightly higher LD figures.
Finally why the fuss. If you ignore BPIX as non BPC registered and lacking in transparency ( and wouldn’t ignoring them have been better than a half baked hatchet job ?) all we are left with is an ICM-by-another-name poll showing us up 3% ( within the margin of error) to 21% a heady height that Ming Campbell achieved.
Polls : report them all, report them monthly in a round up but don’t post specially only when they are good ones.
Eh what? Where is it on Pol Betting? I can see the ICM one but the BPIX post still says “no figure for the Lib Dems reported” at the moment. Still not sure when Tim M got that 17 figure from; happy to be pointed in the right direction.
“Isn’t the simplier explaination for their non appearance in the Mail article that the Mail didn’t choose to include them ?”
Yup, could well be. I’m not sure why that’s any better? It’s a political poll, right? I don’t think there’s anything wrong with registering the fact that I would have been interested to see it (and as I explained in the post, there is a particular reason why the normally unreliable BPIX figures might have been interesting here to shed light on another contemporaneous poll).
“Polls : report them all, report them monthly in a round up but don’t post specially only when they are good ones.”
Yes, I know. Hence my disclaimer at the start. My point here was not that we’ve had a good poll so much as we’ve had a non-reported poll. I’d be interested to know why. If you’re getting shrillness and paranoia then that must be poor drafting on my part. I was actually doing poll-geek with a touch of cynicism.
The BPIX LD figure is not mentioned on Politicalbetting, at least not in any main article, quite the opposite. The first time it is mentioned is comment 104. Who, apart from the greatest of obsessives, can be bothered to scroll down that far (I find it intriguing – and significant – that Jury Team seems to be run by a bunch of pb.com regulars, i.e. people for who politics is no more than sport who feel the need to lecture everyone who takes it a bit more seriously as being the ‘problem’)? I used quicksearch to find that factoid out incidentally.
Having said that, the fact that BPIX is not BPC registered tells me all I need to know about them. Reporting their figures is no more meaningful than reporting the results of those visitor polls you get on local newspaper websites. They’re Mickey Mouse.
Reported here as 21%
http://uk.news.yahoo.com/4/20090419/tuk-labour-support-slumps-after-emails-dba1618.html
James,
As you righly said it is reported on http://www.politicalbetting.com at comment 104 on one thread however they accurately cite sky news as the source which is a little more mainstream. The Lib Dem figure was thus available 2 hours 5 mins ( I said within about an hour in my original post – sorry) after the mail article went live. thats 12 hours before this published article alleges the figure was still not available.
Anyway I’m being far to critical of poor alix who does sterling, unpaid and I imagibe largely thankless work keeping this site going. My main targets are
1. BPIX. whoes website has been under contruction for over a year. they don’t deserve the publicity that even a hatchet job gives them. Though its worth pointing out that lack of transparency doesn’t mean inacuracy
2. Poll Obsession ( say he writing this) We have been through all of this with the “snow and strikes” bouncelet a few months ago.
Nope, absolutely fair dos, I didn’t see the Sky item (note to self: really must check mainstream media; occasionally they have news). I hereby withdraw my post.
David: the BPIX website has been under construction for several years and I’ve never had a reply from the email address on the site either, so criticise away…
17% – it was reported in the paper version at least – but you have to have the shame of looking through the Mail on Sunday in WHSmith to see it.
Interesting to note – despite SmearPornSistergate nobody appears to be registering a bump in approval for the Tories.
If we can get a few more percent in the polls, we’ll be at 1983 all over again, which is a bit creepy.
Tommorrows Guardian ICM : Con 40 Lab 30 LD 19 Others 11