UK Opinion Polls Thread

[Filuwaúrdjan]

ComRes

CON 40
LAB 27
LIB 18

Which leaves 15 for others. Which, ahem, stretches credibility somewhat. Almost as the fact that it was done by a polling company with such a dreadful track-record and with so little credibility that they had to change their name a while back.
 
While we're talking bilge polls, various useless polls were published on Sunday and can be found here: http://ukpollingreport.co.uk/blog/archives/1075 (who the hell are "BPIX"?). Other than to sell papers, I don't get the point in all the dodgy polls by dodgy firms seen over the past week. It isn't as though we'll learn anything.

[Filuwaúrdjan]

I'm not complaining that polls are showing the Tories leading (they would certainly win the popular vote if an election was held today; denying that would be nothing short of delusional).

I just don't like rubbish polls that tell us nothing interest/nothing likely to accurate (except by chance) and that seem to be churned out in a vain hope of bolstering flagging newspaper sales. A poll that, for example, shows "others" on 15% is clearly not credible, even if it is correct in showing the Tories ahead.

[Filuwaúrdjan]

High 'others' in opinion polls often occurs immediately after a 'big' event before the polls settle again and they shrink back to below 10%. It's very much part of the 'ouch' factor when many loyalists become disenfranchised with the party for a short period of time. I'd be worried if we weren't seeing it after the week Labour has had (and as a result in the next round of polls we may see a 'deadcat bounce' for Labour as they correct themselves)

Which is precisely why it's so bloody stupid to commision so many polls at the moment! (other than for the purposes of selling newspapers. Or trying to sell them).

O/c Comres (sounds like a disease of the foot) almost always shows a weirdly high "others" vote and the internals are often "strange" even by the stands of poll internals (which... certainly isn't a good sign).
You don't have to reject all polling firms to declare one to be worthless. I believe that Comres is worthless and I believe that this is shown by a quick check of the, frequently way out of line, numbers that it's been pumping out for the last two years or so.

No problem with accepting that the Tories are ahead. No problem accepting that a sizeable chunk of the Labour vote might currently be pretending that they won't vote out of protest. Big problem with accepting any value in the rubbish pumped out by certain polling companies.

---
Does this post read like one written by someone in a bad mood? If it does; apologies. I am in a bad mood, though its nothing to do with this sort of thing. Basically I'm not well at the moment, am tired and can't sleep and have overworked myself today. And my computer is playing up, because it always does that when I'm tired and irritable. Apologies if I accidentally snap at anyone.

[Filuwaúrdjan]

but I can't see large numbers of people switching from Lib Dem to the Tories,

Traditionally very few (relatively speaking...) people switch their vote directly from Labour to the Tories or vice versa. A vote for the Centre Party (whether it calls itself Liberal, Alliance or LibDem) is often a stepping-stone for a switch between the two big parties. Tory victories are usually accompanied by a drop in Centre support (see 1951, 1970 and 1979). Other movements (in real elections) are usually the result of shifts in turnout patterns. Usually; you do get an election like 1997 every few decades or so.

Never forget that the party system in Britain is founded upon the fault line(s) of Class.

You must be logged in to read this quote.

There is nowhere in England like Kansas or Nebraska.

[Filuwaúrdjan]

OK, does anyone actually think that these early 40's numbers are going to hold for the tories?

Could easily do in pre-election polling but, as I've said before, I don't think that matters a great deal. As to whether they'll be up there during an election campaign... I don't know and neither does anyone else. Too much is uncertain.

If Labour somehow won the next election, would they still lose Brighton to the Greenies?

It is possible that Labour could hold Brighton Pavilion even if it loses the next election. It's also possible that Labour could win the next election and lose Brighton Pavilion.

[Filuwaúrdjan]

Polls this late in the year? Normal people aren't thinking about politics right now, dearies. Wait until the New Year.

Just hope that none will be published next week; the numbers would be close to meaningless.

The main focus of the poll is on Mr Brown now holding a 'dissatisfaction' rating of 60%. It was 27% in July and 48% in October.

He's not been seen to be doing anything while things that, while not exactly serious, are not good, have been going on, so that's no surprise. On a more general point, I worry that he's gone down the Wilson road and assumed that people like and will reward an appeal to managerialism. Would be nice to assume that events of the past few months might change that, but...

[Filuwaúrdjan]

And if the polls in the New Year are still bad, will we be recommended to wait until the spring?

I wasn't trying to make a partisan point; I just don't see the point in doing polls at a time of the year in which ordinary people are paying even less attention to politics than usual.

You must be logged in to read this quote.

All polls outside election time are worthless (almost by definition). If they are consistent then they are less worthless than normal. Polls done around Christmas or so are even more worthless than normal. And so on and so forth.

A slight exaggeration there, but I think that's a reasonable way of looking at things. Btw, I'm not "expecting" polling figures to improve next year; predicting polls really is a waste of time.

[Filuwaúrdjan]

I'd disagree there; polls during election time can magnify short term issues in the news and ignore long term trends.

Polls never have anything to do with long term trends; only General Elections do and some of those should probably be ignored as flukes.

You must be logged in to read this quote.

Yes, but that isn't the point. Polls taken during an election (by which I'm including the weeks leading up the official kickoff of the campaign) have a purpose and can be judged in an objective way after the election. Not true of other polls; they measure hypotheticals and not a lot else.

You must be logged in to read this quote.

Worthless is, I'll admit, an exaggeration. Except when polls are done around Christmas, in the middle of the Summer and so on. There was a local by-election last night with a turnout of 14%.

You must be logged in to read this quote.

True. But I'm not sure that that says much.

You must be logged in to read this quote.

Touché (though I'm not entirely sure about that; if more people paid attention to politics it's likely that the nature of political journalism and the like would be different in some way. But now I'm playing with hypotheticals and should stop before it's too late )

You must be logged in to read this quote.

Ah, but hast thee considered why conference polls are so useless?

[Filuwaúrdjan]

Quite a lot of them, actually ... but of course that way of looking at it has little to recommend itself to professional politicians whose jobs are on the line ... or even to anyone whose future may depend on the next election's outcome.

Such people (the first group anyway...) generally aren't too interested in long term trends so that's nothing to worry about

[Filuwaúrdjan]

ICM out with

CON 39
LAB 34
LIB 18

I take back all that stuff I said about end-of-year-polls being bunk.

Actually... I don't.

Will say something of interest (to me...) on regional breakdown lols (a general comment not specific to this poll) and the reason for it when I'm next online. Apparently I've nothing better to do.

[Filuwaúrdjan]

They released a poll on Christmas Eve?!?!!?

Civilisation is dead.

[Filuwaúrdjan]

Seems more-than-a-little implausible (and I don't buy it), except as confirmation that the government isn't exactly in a good shape right now... which is obviously true.

Fwiw another polling firm had something like 40, 29, 20 the other day (don't recall the exact figures, can't be bothered to check). Which is probably, more or less, what a General Election held today would look like if me readings of the local results is right.
I do hope we don't get another silly rush of rushed-polls in sunday papers... though I suspect that we will, just to piss me off.

[Filuwaúrdjan]

The Man From G.O.P.- I'd say the 1983 period, but it's a guess.

The answer is the early years of the Blair premiership. Except that that isn't the answer really; the methods YouGov use are totally different from those used by other firms *today* let alone a decade ago.

The answer is that there is no answer as there's nothing we can compare to.

[Filuwaúrdjan]

I'm wondering if YouGov were smoking something in that poll .

The Man From G.O.P.- I'd say the 1983 period, but it's a guess.

Late 1981 to early 1982, when the Alliance was over 50% in the polls, maybe, although some of the polls in 1995 might have had Labour that far ahead.

Labour last had a 26% lead in April 1997. The Labour showing; 23% is the lowest since late 1981. It is the Tories highest lead since 1968

Note that polls back then (and until quite recently) weren't weighted for turnout and so on, unlike now. No need to spell out the implication of that.

[Filuwaúrdjan]

Oh it's definately an indication and that's it; it's similar to the crazy polls of the '94-'97 period. However, if it is history repeating itself then it tells us two things. Including 'swing back' to Labour the Tories are heading towards government. Secondly, they may be heading towards a landslide. Which gives us two years, at the most of a lumbering and flaking government.

Agree with most of that (you can probably work out which bits I don't ). Btw, I actually think a General Election next year is more likely than most people are assuming. If economic indicators, poll numbers and the like are at least moving in the right direction than the pressure to go for it would be immense and probably overpowering. But that's a long way off now (and yet isn't...). What it certainly is is idle speculation.

You must be logged in to read this quote.

I would not be happy if he were forced out as PM. I wouldn't mind if he chose (or "chose") not to be the Prime Minister candidate; there are some sorts of politics he's good at, elections don't seem to be one of them.

Personal opinion, nowt else.

[Filuwaúrdjan]

How long till Brown is gone?

No one knows for sure, not even him (unless he's a very strange surprise planned). A more interesting question would be how much it matters, really.

Anyway, he's not going to go because of a poll that look a little odd and has been published in the Sun, we can be sure of that.

*awakes next morning to discover that the opposite is in fact the case*

[Filuwaúrdjan]

Take it or leave it but YouGov report Labour on 23% which is apparently the lowest rating the party has had in the history of opinion polling.

You mean other than the one by the same company three weeks ago with the same number?

In any case, apples and oranges. Under the old ways of doing opinion polls it would have been quite literally impossible for a poll to come up with a Labour figure that low, newer methods mean that it is possible with the methods of some polling companies. If that makes sense.

You must be logged in to read this quote.

Well this is a post by-election poll (and they're always dodgy) so maybe not. But of course the current whining about petrol and house prices could cause all sorts of odd things to show up and the Silly Season is almost on us now so, who knows. Such numbers, if they happen, won't be "real" (in the same sense that this one isn't) except insofar as they would be (and this one is) an indication that the government is somewhat-less-than-popular.

[Filuwaúrdjan]

We only know that 30% was the low water mark for the Tories after the landslide election.

That's true, but then that isn't really saying a great deal; there was no chance that the Tory number in an actual General Election on that sort of turnout would be any lower than that. Easier to say with hindsight than at the time, perhaps, but still true.

What I'll note here is the stability of local election results over the course of the present Parliament; true there was a noticable swing to the Tories this year compared to 2006 and 2007 but it wasn't massive or in any way dramatic (now some of the results it produced were a little dramatic, but that's because in the world of low turnouts in local politics if you lose a few hundred votes in every ward you can easily end up losing most of your seats. It doesn't take much actual movement to produce huge changes in local elections) and the general voting patterns broadly fit those of 2006 and 2007. And yet over this period there have been huge swings in the opinion polls, almost every few months or so. The polls show an incredibly volatile electorate, something which does not fit in at all with the relative stability of local elections. I've come to the conclusion that most of the movement in polls over the past few years has not been "real" and would not have been reflected, much, in a hypothetical General Election held at just about any point over the course of the Parliament.

[Filuwaúrdjan]

So just out of interest, where do you believe things stand at the moment?

Tories likely to win, at least, a stable majority. I suspect that little now can change that.

You must be logged in to read this quote.

God, no. No, no, no, no, no, no.

You must be logged in to read this quote.

I don't mean that there has been no "real" movement. I just mean that the huge movements (the "s" is important!) that have occured all the time in polls over the past few years have, for the most part, not been "real". The polls have shown an extremely volatile electorate, capable of swinging in all kinds of directions, while local elections (and to an extent also local by-elections) over the past three years have been curiously stable. Swings have happend, yes (the consistency of the *pattern* of the Tory swing this year is very interesting) but nothing dramatic that can't be pinned on local factors of one sort or another.

Not sure how clear that is; will elaborate more if it's not clear. Actually, I'll dig up a few examples.

You must be logged in to read this quote.

Of course. But *what* are the trends that they show; do they say more about party motivation than how people would actually vote [qm]. Their volatility, in contrast to actual elections, is the thing that makes such a question legitimate.

[Filuwaúrdjan]

Some Coventry wards...

Westwood (suburban area, fairly typical in most respects if a little more working class than average, includes part of Warwick Uni)

2006: Con 1566, Lab 1374, BNP 473, LDem 383, Ind 120
2007: Con 1426, Lab 1326, BNP 456, LDem 374, Ind 161
2008: Con 1713, Lab 1389, BNP 425, LDem 277, Green 165

Holbrook (almost the very stereotype of a blue collar inner suburb, pity about the lack of a LibDem candidate in 2008)

2006: Lab 1673, Con 833, LDem 481, BNP 451
2007: Lab 1658, Con 871, LDem 544, BNP 469
2008: Lab 1821, Con 1077, BNP 509

Earlsdon (middle class (esp. professionals) urban area)

2006: Con 2520, Lab 1073, Green 805, LDem 670
2007: Con 2364, Lab 1241, Green 652, LDem 427, BNP 189
2008: Con 2515, Lab 1155, Green 698, LDem 440

Edited-in notes: wards chosen on a semi-random basis, the choice of Coventry not random at all; there have been city-wide swings in all directions there, the same basic set of candidates tend to contest each ward each year and while local politics there can be odd, it isn't as quirky and personality-driven as other places. And of course the point isn't the swings (whatever they may be) but the relative stability of voting patterns. More soon from other places.

[Filuwaúrdjan]

Ipsos Mori show Tories breaking 50%

CON: 52%
LAB: 24%
LIB: 15%

'Tis a silly poll but...

Seat forecast on a uniform swing

CON 488
LAB 121
LIB 13

CON MAJ 326

Desperate Spinner: our vote is holding up! This is the first Mori poll in months that it's not fallen further off a cliff! Recovery!!!!!

A Different Desperate Spinner: 121 seats is a whole 69 more than we managed in 1931; and boundaries then were more favourable to us than they are now!!!!

But, seriously...

I heard that the LibDem figure was actually 12%. Not that it matters; this isn't a poll done to measure public opinion so much as it is to change the opinions (in a newsworthy direction, whatever that is) of a small group of people next week or so. I know, I know, I'm a dreadful (to paranoid) cynic about that sort of thing, but this time the chances of me being wrong are rather small. Especially given the source, so to speak. I guess we'll see more polls like this over the next few days, but hopefully not quite so comical.

[Filuwaúrdjan]

It has the Tories winning Glasgow North for example.

Now that is rather amusing

However...

You must be logged in to read this quote.

Meh. I won't criticise thee for fantasising this time

Though I'll say that you're always better off being happy with what you have now than looking for better and better news in the years to come (an interesting flaw in modern society is the latter trait). Best way to avoid disillusionment really.

(fwiw I still expect a Tory majority of between 40 and 60 at the next election. Though I'm not stubborn enough to avoid lowering or raising that if the facts do change. I'll also stick my neck out and say that the swing (whatever it is) will be highest in suburban areas, especially post 1970's ones and especially ones with a lot of growth in the past decade).

[Filuwaúrdjan]

Elections like 1931 are only possible in an atmosphere of mass panic.

[Filuwaúrdjan]

Every time we beat the Enemy in the Six Nations a shit poll is out the same night. I'm calling this a conspiracy to (mildly) deflate my mood. I think the evidence is clear enough.

Labour are effectively in freefall shedding 9 points since the last ComRes poll of 2008.

Meh. That says almost as much about the hilarious volatility of ComRes as much else. Sort of obvious that Labour support has been falling again, though (same sort of reasons as last year as well).

[Filuwaúrdjan]

I believe that this is one of those "lol@thepollingindustry" moments.