Are we being misled by the Baxter seat calculation?
February 26th, 2005
-
Working the swing out from 1997 - not 2001
The standard way of working how many seats a party will get with a given share of the vote is to use one of the seat calculators, most commonly Martin Baxter’s, which makes its prediction based on applying the suggested uniform national swing to what happened in each of the seats in 2001.
But what happens if instead you compare today’s vote shares with what happened in 1997 when a much younger Tony Blair (above) led New Labour to its stunning landlside victory?
For 2001 was an extraordinary election for Labour which managed see its margin over the Tories decline by more than a quarter but still be returned with a majority of 160+. This was because the party did so much better in its marginals and the fact there was more tactical voting.
But what happens if these two factors do not happen again on the same scale which seem likely? Why not apply recent poll shares to the 1997 General Election result. This is possible thanks to the Hill & Knowlton 2001 calculator which is still available on-line.
When doing this you have to factor in the fact that there will be fewer seats this time because of the Scottish boundary changes.These are reckoned, on a notional basis, to have “lost” Labour 10 seats and the Tories, Lib Dems and the SNP one seat each.
So taking the February ICM poll of LAB 37: CON 34: LD 21 you get the following from Hill & Knowlton with comparisons on doing the same calculation with Baxter. LAB 336 (-33): CON 215 (+26) : LD 63 (+9).
Thus instead of Baxter’s prediction of a Labour majority of more than 90 Hill & Knowlton, adjusted for Scotland, gives you a Labour majority of just 25.
Which one’s right? Hard to say but the special factors that saw Labour get so many seats for its vote in 2001 won’t exist to the same extent as last time. Making the calculation, as H&N does, on the 1997 result puts today’s uniform national swing predictions into perspective - vital for all those spread-betting.
© Mike Smithson 2005
MessageSpace Advertising
Is there not another reason why Labou got so many seats for their vote share in 2001, namely the larger drops in turnout in Labour’s safe seats than in other seats of all kinds?
I can just about imagine a tactical unwind and Labour not doing so well in their marginals as last time, but I don’t think there will be a disproportionate rise in turnout in safe Labour seats.
For the ‘Hill & Knowlton’ results from 1997-2005 to work, there would be some very peculiar patterns. I myself would be pretty surprised if the overall swings and effects are very far from uniform.
You are probably right Robert, however there is probably some justification to this method for prediction purposes. The polls probably include, to some extent dependent on turnout models, all these (’non’) Labour voters in safe seats - voters who are effectively worthless with respect to the general election result. Using the 1997 calculator may be one way of filtering out the claimed ‘Labour bias’ in the polls - which have some have pointed out may not be a bias as such, just an inability to account effectively for Labour supporters failure to vote.
In practice though, however, you probably have to come up with some sort of average figure located between the underlying pollsters’ figures, and their turnout adjusted models.
I’d been toying with this idea too Mike. Its certainly attractive in that it deals with the problem of the Tories useless 2001 campaign in the ultra-marginals which wont be repeated this time.
Look at the actual % of the electorate that voted labour and conbservative since 1974 (from Keele.ac)
1974: L29.27 C29.76
1974: L28.57 C26.01
1979: L28.00 C33.33
1983: L20.04 C30.84
1987: L23.23 C31.81
1992: L26.73 C32.58
1997: L30.80 C21.85
2001: L24.15 C18.82
In 2001 labour had 1.6% less of the electorate vote for them than voted for them in 1992. (in fact in 2001 they only got 0.92% more of the electorate voting for them than voted for them in 1987). The reason they won in 2001 is a massive collapse in the conservative vote coinciding with a 18% drop in the turnout from 77% in 1992 to 59% in 2001.
The big story from 1992 to 2001 was 18% of the electorate going missing. In 1992 turnout was 77% in 2001 it was 59%. Comparing the % of the electorate voting for each party in 1992 and 2001 gives you a breakdown of the missing 18% of the electorate.
Cons have lost approx 13.76% of the electorate between 1992-2001
Labour have lost 2.58% of the electorate between 1992-2001
(there is a slight discrepancy with a similar article I posted on anthonys Blog because this time I used the Keele figures for electorate % rather than calculate my own)
This means that there could be a lot of tory voters out there. 41% of the electorate did not vote last time. It seems the people of Britain who have not spoken yet (or did not speak in 1997 and 2001) could have a very disproportionate influence.
If say 5% of the electorate who voted Labour in 2001 abstained in 2005, say owing to the war, and 5% of the electorate who abstained in 2001 voted tory in 2005, the political map could change very dramatically, even if everyone else voted the same way as they did in 2001
For the record.
Id 5% of the electorate who voted labour last time abstain, and 5% of the electorate who abstained last time vote conservative, and everyone else votes exactly how they did in 2001 my calculations gives a voting % as follows:
C41.2, L33.55, LD 18.84
Baxter gives the tories with 312 seats, 12 short of a majority, H&K give the tories 318 seats (after scotland adjusted), 5 short of a majority.
There’s a YouGov poll in the Telegraph about Charles and Camilla. Would this likely also have asked about voting intentions, even if not published, and if so how could the resuls be accessed?
Paul your method seems to pre-suppose that people only ever vote for one party…or don’t vote. Labour have ‘lost’ far more than 1.6% of their voters since 1992 - it’s just been compensated for by the huge number of voters they’ve gained from the bTories.
Paul, all the evidence about the 2001 abstainers (eg the big discrepancy between Labour’s huge poll leads recorded in the campaign and their somewhat more modest 9% in the election, and follow-up surveys of non-voters) suggests that these were disproportionately more likely to have preferred Labour, rather than there being lots of Tories out there who abstained.
John O, what newspapers often do is to use different questions from the same survey in successive days, to spread the stories, so I would expect that the C & C questions were probably on the same survey as yesterday’s voting intention (someone might check the sample size details. They are mean so and sos and don’t like paying twice!
8-9 Alex, Robert.
The point I was trying to make is that with only 59% of the electorate voting last time relatively small shifts in absentions and returning abseentee voters can make a huge difference.
The figures I know were crude - assuming everyone else voted the same way in 2005 as 2001 for a start, however John major survived due to an unusually high turnout - a lot of people voted tory through gritted teeth to stop Labour.
This could happen again - especially if the gov’t keep doing things like proposing detention without trial. This alone has switched someone I know from wavering towards the Lib Dems to “I dont like the tories much but we’ve got to get this lot out”
Paul (10): it is very important to realise that around half of all voters change what they do between each General Election, as the academic British Election Study panel surveys have consistently shown over the last two decades. Less than half of the electorate are ‘Labour’ or ‘Conservative’ or ‘Liberal Democrat’.
However, while I pulled you up on the assumptions in your previous posts, you are quite right to say that there is much to play for because of the volatility of the electorate. Shifts in abstention and returning absentee voters can always make a huge difference and may do so this time, or next.
However, as the general perception of economic prosperity has not deteriorated significantly since 2001 and as the swell of ‘time for a change’ opinion has not reached anything like 1997 levels the probability is that the present government will be returned, with a smaller majority due to the loss of faith in the leadrship recorded by all polls itself partly a result of the circumstances of the intervention in Iraq, and its repercussions including its arrogance and authoritarian policies.
INterstingly in the YouGov details from their latest poll - in London the results are Cons 41% (-2), Lab 32% (+4) and Lib Dem 20% (-2). They have shown a consistent strong swing to the Conservatives.
Rik - doesn’t -2% suggest a swing *away* from the Conservatives?
Well this thread raises a very good point: why use a swing model for this election that didn’t work last time?
What evidence is there that THIS TIME, when much has changed, increased turnout helps Labour, as widely perceived?
When doing my calculations I use UK-Elect and temper it with a 14% tactical vote against Labour on a 57% turnout.
JohnO at 7. When the figures are known I expect them to show Charles voting Tory and Camilla UKIP!
Printz. Mori’s latest poll only counts those ABSOLUTELY certain to vote. And I think it’s widely accepted that those who MIGHT vote are disproportionately Labour supporters which is why high turn out favours Labour. All the polls seem to show the same thing which is that Labour are pretty much preferred to the Conservatives (which doesn’t say much). At the moment the Conservatives best chance is if Labour supporters lack of enthusiasm shows on the day. But all the polls have this well factored in already. And this thread seems to be a case of “William Rees-Mogg type” wishful thinking.
Has anyone else been having difficulty getting pasts on to the site recently?
Paul (10): it is very important to realise that around half of all voters change what they do between each General Election, as the academic British Election Study panel surveys have consistently shown over the last two decades. Less than half of the electorate are ‘Labour’ or ‘Conservative’ or ‘Liberal Democrat’.
Paul (10) Continued: However, while I pulled you up on the assumptions in your previous posts, you are quite right to say that there is much to play for because of the volatility of the electorate. Shifts in abstention and returning absentee voters can always make a huge difference and may do so this time, or next.
Roger, I’ve always thought of him as a ‘greenie’ albeit one with a highly reactionary tinge! You’re probably right about her.
Paul 5. The interestingn thing is that the main reson labour got in was that tories just stopped voting between 1992, and so now its quite possible with a more ‘realisitic’ looking opposistion (which i think is the one thing that micheal howard has acheived) to appeal to core voters then they might do a lot better thsn people expect.
13 Tabman - swing Since the GE not since the last poll!!
Rik why are you not out their canvassing on a saturday afternoon in Sutton and Cheam. Are you hemmed in, in the office?
Andrew at 24: well, I’ve just come in for a meal after 2 hours on the High Road - it’s freeeezing! The Tories had been there but had given up an hour earlier. Anecdotes: some signs of real election fever - people queuing up to talk with me and the County candidate, passers-by calling out pro and con, people asking the exact date to check they’d be around, etc. Three separate people said they’d been “wavering Labour” but had returned to the fold - one because we’d finally banned fox-hunting, the other two because they disliked the populist Tory posters (’they pretend they’re superior and then they talk like drunks outside a pub’, said one elderly woman). Two pensioners torn between the Tory council tax offer and the Labour ID card policy - they wanted both, and walked away, still debating with each other. A Greenham Common veteran bitterly criticising me over Iraq, and cursing the fact that she would still have to vote for me to keep the Tories out. Lively! - high turnout stuff.
Nick
Still you won’t have go the turnout that Boris Johnson got last night. He was up in Durham at the University. He managed to get over 600 people listening to him, some via videolink. It was extraordinary. Whatever one’s thinks of the man’s political skills, there is no doubt about his popularity. I think after the next election, the Tories should think seriously about providing him with a role in the party.
17 - Roger. But what about all those that are staying at home because they are loathed to vote Labour after Iraq?
If they are encouraged to come out and vote, in a drive to increase turnout, is that really good for Labour?
Surely they are more likely to vote against Labour or for an anti-war party.
Aren’t these voters being picked up in polls as the ones less likely to vote?
I’m not convinced at all that this time increased turnout is good for Labour.
24 - Andrew - dont worry I was out early this morning then back for lunch when I posted. Out again this afternoon delivering and only just back now. Had another new helper turn up unexpectedly this morning. We now have over 150 deliverers across the constituency. Over 30,000 of my latest newspaper (see my website) have gone out in the last 3 weeks!
Results of latest telephone canvassing are very encouraging with lots of new pledges and offers of help. We are motoring!
The down side of this game is days like this week when I was up early to do delivery before work of some flats that I cant normally get into - it was snowing!!! Delivering the snow isnt the best idea!
Printz at 27: The poll evidence supports Roger pretty unambiguously: if one doesn’t discount lesser voting intention, there is a huge Labour lead in all polls except YouGov. The problem is that all pollsters acknowledge that it’s hard to predict voting intention accurately - people who swear they’ll vote sometimes don’t, and people who think they won’t bother sometimes do. Humans! So unpredictable…
Nick
29 - Nick what “huge” Labour lead? Would that be the 2% in MORIs last poll or the 3% in ICM?
Labour supporters armuch more likely, as well, to exaggerate when they answer turnout questions. I ve long argued that Labour turnout projections should be weighted in different manner to take this into account
5 Paul - I think you have to remember that in 1992 Labour was still a “Clause 4 Socialist” party. I reckon the “Tory” vote was boosted because of that. That doesn’t apply any more.
30 - Rik I think Nick is referring to the raw data pre weighting.
Sorry Nick - one voter is going to be pissed off - They were hunting in South leicestershire again today
33 - Even those leads are not huge! Huge was pre 1997 when Labour were often 20%+ ahead. 10-12% is comfortable but not huge and 2-3% is effectively neck and neck. Labour’s lead is about halved on the unadjusted polls from 1996-7 and even from some of the pre 2001 polls.
This whole election is open in a way that some of us never thought possible when IDS was leader. The hugely gratifying message that I see is that the Lib Dems are down considerably on their polling figures of recent years. I see no reason why they should get a “campaign boost” this time and I would be surprised if they get more than 20% on polling day. My personal prediction is for them to get 18-20%, with a fall in London and SE, but a boost in Midlands and the NE.I think they will be static in Wales and Scotland. They are reverting to their 80’s pattern of more widely and thinly spread support and I think therefore they will see a net loss of seats this time. Other than in a few University or Moslem concentration seats, I dont think there will be much “Iraq effect”.
29 - Nick, I’m not convinced that out of those that don’t vote a greater percentage of them would vote Labour than those that do vote (the argument that higher turnout helps Labour most) especially considering the bitter disappointment and betrayal Blair’s second term has been to perhaps ten million people.
If you had taken an opinion poll of those who marched against the war, you’d likely find a disproportionate number were from real Labour, Liberal, Green and the stay at home vote. They won’t be voting for you will they? Likewise millions of others are loathed to vote Labour because of the lies told over Iraq and tuition fees etc.
I’m yet to see a shred of evidence that in this election a higher turnout will favour Labour. There is a big difference between those that say they WOULD vote Labour and those that say they WILL to polling organisations.
Labour may spend an unprecentended sum trying to manipulate turnout as they did in areas that suited them in the local/Euro elections against Electoral Commission advice, but I very much expect this won’t help them or you.
Rik. Mori’s basic poll has Labour 43% Con 31%. That’s pretty huge. If it happened I think William hague might be the last Tory standing (or should that be sitting?)
http://www.mori.com/polls/trends/voting-all.shtml
Andrew at 26. I wouldn’t advertize that on a Friday night 600 Conservative students preferred to listen to an MP than doing what students usually do on a Friday night. It’s bad for the image. Anyway I thought all that Old Fogey stuff was dead? I mean he might be entertaining but on a Friday night!
Friday evening at 5:30 PM. You could go and see Boris and then get merry in the pub.
37 - Roger - 13% (unadjusted)lead is not huge - it is comfortable. Have a look at http://www.mori.com/polls/trends/voting-allpub-trends.shtml and you will see that in 1994 Labour’s lead reached 39% and was regularly over 30%. Now that’s huge!
13% can become 6 or 7% in one poll shift. The Lib Dems have seen movements of about 10% in their poll rating in the last year or so. When you have been studying polls for as long as I have you know what a huge lead looks like!
For all you doomsayers out there I can also recall when the Tories under Margaret Thatcher polled just 24%! We went on to win the next General Election. So lets have some realism and some perspective here!
Of course another unconsidered factor is the reasons people answer polls as they do - especially relevant with Mori as they don’t just ask “how will you vote?” but also “for whom would you be most likely to vote?”. Now it may have diminished somewhat recently but traditionally Labour for many people would, I hypothesise, have been the ’socially acceptable’ answer to give if pushed. Hence you don’t just have the problem of people refusing to admit they support the Conservatives but also people, who possibly might not even vote, claiming to be Labour supporters. Mori’s methods give much less room for people to cop out and say “don’t know”, exacerbated by their ‘face-to-face’ interview techniques.
“I go down to speaker’s corner I’m thunderstruck
They got free speech, tourists, police in trucks
Two men say they’re jesus one of them must be wrong
There’s a protest singer singing a protest song - he says
’they wanna have a war to keep us on our knees
They wanna have a war to keep their factories
They wanna have a war to stop us buying japanese
They wanna have a war to stop industrial disease
They’re pointing out the enemy to keep you deaf and blind
They wanna sap your energy incarcerate your mind
They give you rule brittania, gassy beer, page three
Two weeks in espana and sunday striptease’
Meanwhile the first jesus says ’I’d cure it soon
Abolish monday mornings and friday afternoons’
The other one’s on a hunger strike he’s dying by degrees
How come jesus gets industrial disease”
M Knopfler c1982
Printz. As one who went on the anti-war march I would agree with you that Labour supporters (probably an overwhelming majority) were seriously angry with Mr Blair and his government. From my anecdotal evidence almost 100% swore they would never vote New-Lab again. My own feeling is that time has healed somewhat and perhaps 75% will be back if the election looks close. If Labour stay well ahead in the polls then I think maybe 50% of the serious anti-war brigade will not come back. For myself if anyone had asked me in December I would have said Lib/Dem but now I’m starting to feel more forgiving and I really fancy seeing what Gordon Brown will make of the job when Tony goes. Also I can’t help but find Charlie Kennedy completely hopeless which doesn’t leave me many options.
32 Alan. Point accepted, clause 4 has gone and Labour have economically reformed. But I think many, particularily of the older generation fear a creeping authoritarianism, epitomised by Clarkes detention without trial proposals.
While some of the voters not voting in 2001 were core Labour voters who didn’t bother, this time I suspect it will be post ‘97 Labour converts and Lib Dem supporting tactical voters who sit on their hands (or vote Lib Dem)instead of voting Labour.
With Labour achieving a 160 majority on the back of the support of only 24.15 of the electorate, it does not need many of the missing 41% of the Electorate to return and vote tory to wipe Labours majority out.
I think Labour’s position is a lot more shaky than generally thought.
If labours majority is cut to less than 40, they will have real trouble controlling the parliamentary party and Michael Howard may well employ the same guerrila tactics that John Smith did (ie abolish Pairing etc.) to wear down and destroy them in the same way that Majors administration was destroyed.
Good to see that Rik is getting the same feeling as our team in Angus. LD’s actually prepared to vote Tory to defeat the SNP locally and totally fed up with Labour. Many lost Tory voters returning not only to vote but actually work on the campaign. It’s going to be a very interesting couple of months.
In which case, Paul [43], Blair will go sooner rather than later and his successor will have the option to seek his own mandate at a time of his choosing.
Thanks for that link Rik. Extaordinary in many ways that no pollster even got the right winning party in ‘93 let alone that they won by over 7%. Also looking from ‘92 on it’s cleat that ICM is much the most accurate pollster
Roger - Listen carefully: Labour’s lead IS-NOT-13%-BECAUSE-THEY-ALL-WON’T-VOTE….COMPRENDEZ LOVE?
I’m a newbee on this site - so sorry if this has already been discussed - has anyone considered the possible electoral ramifications of a ghastly terrorist strike in the UK a few days before the election - a la Madrid??
42 - Roger, you are right, time heals and the opposition is weak. Michael Howard should have maximised this week’s news about the legal advice for war and got it on all the front pages to dig up memories, but he is too weak.
Most people won’t vote on the war. Most people only have a very superficial view of politics and never took too much notice of who said what about dossiers and intelligence. Blair has fooled most people with all his wriggling and spinning, but there’s still a sizable chunk who will not bring themselves to vote Labour until Blair has gone, because they have seen through him and don’t trust him as far as they could throw him. If that is only one Labour voter in ten, which I believe is a reasonable estimate, it would have a major impact on the election result.
I couldn’t improve on it Sarah (words by Bob Dylan)
I laid on a dune, I looked at the sky,
When the children were babies and played on the beach.
You came up behind me, I saw you go by,
You were always so close and still within reach.
Sarah, Sarah,
Whatever made you want to change your mind?
Sarah, Sarah,
So easy to look at, so hard to define.
I can still see them playin’ with their pails in the sand,
They run to the water their buckets to fill.
I can still see the shells fallin’ out of their hands
As they follow each other back up the hill.
Sarah, Sarah,
Sweet virgin angel, sweet love of my life,
Sarah, Sarah,
Radiant jewel, mystical wife.
Sleepin’ in the woods by a fire in the night,
Drinkin’ white rum in a Portugal bar,
Them playin’ leapfrog and hearin’ about Snow White,
You in the marketplace in Savanna-la-Mar.
You lefties are so sentimental….BUT YOU’RE STILL W-R-O-N-G!
I’ve just had a look at that mori site linked by Rik in 39.
At one point in 95, there was a Labour lead of 43.5% reported. Not vote share - lead (Con 18.5%/Lab 62%/LD 14%).
Blimey.
More seriously, scanning down it casually, you’re struck by the invariable double digit (not just teens - twenties and thirties) leads from 1993 on (aside from an isolated ICM poll in 97) - a major dip can be easily correlated with the Fuel protest - until after the 2001 election.
Since mid 2003, however, double digit leads have been few and far between. It does look a lot closer.
That said™, another thing that stands out is the Lib Dem increase - the last time they scored this consistently in the 20s was 1993-1994. Could the Blair effect have stolen Lib Dems, who are now returning?
Spam-trap’s caught another longish comment.
Remind me to copy them in future ready to paste and try again …
Sarah J is back
Does Andy or anyone else know how to avoid the spam-trap, if that’s what’s done for a couple of my ponderous ponderings recently? I did indeed copy, but the attempts to paste took much more time, and eventually I only got the first two-thirds of one through this morning - breaking it down seemed partially to work.
It really is a disincentive to formulate developed comments.
One way that occurred to me after losing three detailed ones in a row was that if a series of DDoS attacks were organised against every porn-spam and online bloody poker spam merchant, we wouldn’t have that problem.
But that’s almost certainly illegal (although satisfying to consider)
Trying again …
I followed the link that Rik put at 39 about the polling trends hosted on the MORI page.
Blimey. At one point in 95, Labour had a 43.5% lead. That’s lead, not vote share (Con 18.5, Lab 62, LD 14).
Scrolling down and tracking the Labour lead column is good for highlighting the trend. One thing stands out - from mid-93 onwards, the Labour lead was almost invariably in double figures - and not just in the teens. Twenties and thirties were common. Exceptions look like rogue polls or can be linked to the Fuel Protests.
From mid 2003 onwards, they look a lot lower - double figure leads are fewer and farther between. It does look an awful lot closer than at any time since Black Wednesday in October 92.
Another thing leaps out - from mid 94 onwards, the Lib Dem vote share collapsed from mid twenties to mid-teens and didn’t recover to being consistently in the twenties until 2002. Could it be that the Blair effect robbed the Lib Dems of a chunk of support, who are starting to trickle back? Will the fear of a Tory resurgence pull them back into the Labour fold?
Trying again …
I followed the link that Rik put at 39 about the polling trends hosted on the MORI page.
Blimey. At one point in 95, Labour had a 43.5% lead. That’s lead, not vote share (Con 18.5, Lab 62, LD 14).
Scrolling down and tracking the Labour lead column is good for highlighting the trend. One thing stands out - from mid-93 onwards, the Labour lead was almost invariably in double figures - and not just in the teens. Twenties and thirties were common. Exceptions look like rogue polls or can be linked to the Fuel Protests.
From mid 2003 onwards, they look a lot lower - double figure leads are fewer and farther between. It does look an awful lot closer than at any time since Black Wednesday in October 92.
Another thing leaps out - from mid 94 onwards, the Lib Dem vote share collapsed from mid twenties to mid-teens and didn’t recover to being consistently in the twenties until 2002. Could it be that the Blair effect robbed the Lib Dems of a chunk of support, who are starting to trickle back? Will the fear of a Tory resurgence pull them back into the Labour fold?
Trying again:
I followed the link that Rik put at 39 about the polling trends hosted on the MORI page.
Blimey. At one point in 95, Labour had a 43.5 percent lead. That’s lead, not vote share (Con 18.5, Lab 62, LD 14).
Scrolling down and tracking the Labour lead column is good for highlighting the trend. One thing stands out - from mid-93 onwards, the Labour lead was almost invariably in double figures - and not just in the teens. Twenties and thirties were common. Exceptions look like rogue polls or can be linked to the Fuel Protests.
From mid 2003 onwards, they look a lot lower - double figure leads are fewer and farther between. It does look an awful lot closer than at any time since Black Wednesday in October 92.
Another thing leaps out - from mid 94 onwards, the Lib Dem vote share collapsed from mid twenties to mid-teens and didn’t recover to being consistently in the twenties until 2002. Could it be that the Blair effect robbed the Lib Dems of a chunk of support, who are starting to trickle back? Will the fear of a Tory resurgence pull them back into the Labour fold?
Trying again …
I followed the link that Rik put at 39 about the polling trends hosted on the MORI page.
Blimey. At one point in 95, Labour had a 43.5% lead. That’s lead, not vote share (Con 18.5, Lab 62, LD 14).
Scrolling down and tracking the Labour lead column is good for highlighting the trend. One thing stands out - from mid-93 onwards, the Labour lead was almost invariably in double figures - and not just by a bit. Twenties and thirties were common. Exceptions look like rogue polls or can be linked to the Fuel Protests.
From mid 2003 onwards, they look a lot lower - double figure leads are fewer and farther between. It does look an awful lot closer than at any time since Black Wednesday in October 92.
(Pt 1)
Another thing leaps out - from mid 94 onwards, the Lib Dem vote share collapsed from mid twenties by about 10 points and didn’t recover to being consistently in the twenties until 2002. Could it be that the Blair effect robbed the Lib Dems of a chunk of support, who are starting to trickle back? Will the fear of a Tory resurgence pull them back into the Labour fold?
Yay, it worked! (10th attempt)
Avoid using a word that starts with a “t” and ends with “eens” when referring to numbers between twelve and twenty (eg when referring to leads in the polls). It seems that it is one word that triggers the spamtrap.
Andy - yes I followed Rik’s link and one of the other things that stands out about the Lib Dem vote is that it always seems to dip a bit in the couple of months before the election to then recover during the campaign.
At this point in Feb ‘97 they were at just 11% in a MORI poll and in 2001 they were at 14%. So even if you take thelower end of their current figures - 17-18% - they are still up compared to the same point in the cycle at the last two elections.
Off topic slightly, but does anyone think the numerous and growing allegations about electoral fraud in Birmingham could have negative effects on the Labour vote there?
What allegations ? Please do tell !
thomthumb (64): I think this probably refres to the alleged postal voting fraud in predominantly Asian/Muslim wards such as Bordesley Green in which Labour council candidates are accused of falsely producing several hundred votes, the complaints being on behalf of local Liberal Democrats who had hoped to win these council seats.
Nick, conventional pollsters’ raw figures, unadjusted for voting intention, are virtually worthless, IMO. In the months following the 9/11 bombing, MORI was giving Labour a 30% lead on its unadjusted figures. I doubt that even the most optimistic Labour supporter believed Labour were really 30% then.
And without wishing to prejudge the outcome of the case, Robert, I think the evidence so far seems pretty damning. I think it’s also significant that Labour has stopped financing the accused councillors’ legal expenses.
In Bordesley Green they are accused by the People’s Justice Party, while in Aston by the Libdems. In total 6 councillors are accused of rigging ballots.
I was in Birmingham just before the May 2004 elections, and I was somewhat surprised by the results. The turnout in Bordesley Green was recorded as 54.3%, compared with, say Edgbaston 33.0% or King’s Norton (a split-won marginal) 36.3% or inner city marginal Nechells 25.5%.
It was clearly a hard-fought contest.
Robert@69 - Some Birmingham wards have had higher turnouts for sometime, the old Moseley ward has been over 40% for a while, the old Sparkhill ward has been just under 40%. Also, in my opinion, there has always been a certain amount of fraudulent voting in the area since contests became close, there was an election petition rejected on the grounds of mis-service in 2000 in Sparkhill which was otherwise clear cut. However, this years antics were on a more spectacular scale and it is best to remember that these two petitions were brought as examples, there could easily have been more had local parties had the time and more importantly the finance to do this.
Anti-war lefty Ken Macleod is getting worried: An uneasy feeling, nothing more: This country is sleepwalking towards a Tory government. A new Tory government would…
Thanks! get out of debt
vigrx…
Quite vigrx ?…
free bingo…
Please hand free bingo ….
free bingo…
Lots of free bingo !…