Okay people heres a curved ball, which all you people that love to number crunch could have a go at.

What if at the next election the conservatives decided not to stand candidates in Scotland. It would be publically about english vvotes etc, but the real reson would be that iftheir supporters voted tactically it would cost labour more seats than they could possibly takke.

Peter.

I’m convinced that no party can win a majority at the next election. Labour only needs a 1% swing against it to lost its majority, which is virtually inevitable, and the Tories are still too unpopular outside London and the South East to be able to win the 116 seats they need for a majority in a single go.

Good arguement. The change in methodology does moves things on in our thinking.

The two things I would add are that firstly I sense that the duration of the lead has some impact on whether it is actually delivered on an election day. 1970 is the benchmark here - the swing back to Labour was strong but did not overcome 4 years of bad polls. So from your perspective as a Tory activist. 4 years of solid leads is more likely to convert into votes. I can’t back this with stats, I am not sure how you can.

Secondly, for all the talk of Tory failure in the north (and they do have a problem in Lancashire), they have to pick off the Bedford’s and Luton’s and Northampton’s of the south. There the “lumpen middle class” - I can think of no other expression - have a shallow conviction and can move en masse. The 1997 labour sweep of the south was only 14 years after they only had 2 seats outside London in the south. They only had 6 in 1987

I suspect this class are reserving judgement right now - and will do until Brown establishes himself. The person who is most watching this, apart from the politicians, is Rupert Murdoch.

I’d probably agree with Andy’s analysis that a hung parliament is probably the most likely result at the next general election.There’ll be an interesting tactical vote between people wanting Labour out and voters who don’t want the Tories back in.

Thanks Antony for this well argued and balanced analysis. It supports the view that I and a number of other political punters have held for a while that, subject to ‘events’ of course, a Hung Parliament is highly likely.

Rod Crosby has done some interesting statistical work on this and rates the probabilty as high as 90%. Intuitively, I feel this is a bit high but I wouldn’t quarrel with 75% - and you give some of the very good reasons why this might be so.

It will I think take the general public quite some time to realise that this is the direction in which we are heading. When it does so, it may shrink back in horror. On the other hand, it may say ’so what?’

Though interesting, it is idle to speculate on the long-term impact an HP would have, since much would depend on the strength of each Party, never mind the attitudes of their supporters and Leaders. It seems likely however that we could be in for a lengthy period of coalition Government and/or a change in the voting system, at the very least.

Better start thinking about it.

A neat analysis but it dos’nt take into consideration the boundary changes that have just been published which appear to be of benifit to the tories I think a small overall majority to them.

The article is here….

index.php/archives/2006/12/16/rod-crosby-is-a-hung-parliament-97-likely/

It’s based on Anthony’s data, with the following assumptions:
5 Sinn Fein abstentionists (effective majority is therefore 323, not 326)

2 SDLP (can’t see them holding Belfast South), who would vote with Labour in a tight vote.

Blaenau Gwent in the Labour column (either they win it back, or Dai Davies comes on board)

It therefore is based on a “practical” hung-parliament. The chances of a simple arithmetic hung parliament are consequently a little higher.

The maximum likelihood of a hp is at a Tory lead of about 6%, slap bang in the middle of what I have termed the HP-Grand-Canyon.

Another interesting feature of the data is the Lib Dems Kingmakers region, which is large, especially for LD shares over 19%.

One practical outcome of this is that it may raise the bar for the Tories from a 6% lead (largest party) to about 8%, the point where Labour and LD combined could not command a majority in the House, because I personally feel that if they had the numbers a deal would be inevitable.

Noel - the numbers are all based on the new boundary figures.

John - looking at the local election results last year, I think the Conservatives have more of a problem in West Yorkshire than in Lancashire. There’s a lot of key marginals in Lancashire, but in some councils there the Tories showed signs of life. West Yorkshire has some key marginals, but the Conservatives were going backwards last year.

On your other point, there are seats and demographic groups that swing more than others - seats that seem to produce larger swings than others. In the 2005 general election it was noticable that new towns, presumably for some demographic reason, tended to produce some of the biggest advances for the Conservatives.

Anthony,

I noticed that the SNP vote for westminster is up a bit, from around 19% to closer to 23%, is that enough to lose labour a seat or too, or even get the Tories a second scottish one.

Peter.

Peter Cairns,

In reply o your curve ball, is this serious or wishful thinking on your part ?

Judging by the 2005 figures, Conservatives should win far more seats by standing, than they could cause Labour to lose by not standing - not to mention what they should be expected to do in Dumfrieshire & Clydesdale.

If your point is that tactical voting could well work against Labour (both in May this year and at the next general election), then logically you should be urging all non-Labour voters to vote Conservative in those seats where they came second in 2005. There are at least five of these which could credibly be won from Labour in this way - more than the number of seats where the Conservative share of the vote was larger than the Labour majority and the Conservatives are not themselves in contention.

In any case, the main beneficiary in the handful of cases where absence of a Tory could focus anti-Labour vote would mainly be LibDems, not SNP. Only in the case of the super marginal Dundee East and Ochil/Perthshire South is the Tory vote greater than Labour’s majority over the SNP, and I think that the SNP will win those two even if the Tory vote rose 5%.

Perhaps we can review the position after 3rd May, but I suspect that your case may then be weaker still.

Peter,

Looking at the headline figures for Scotland, only change I can see arising - assuming a pure uniform swing, Ho ho - is that Cons gain Dumfries & Galloway, LibDems gain Edinburgh South, and SNP gain Dundee East, Ochil / S Perth, and possibly Glasgow North.

I expect that the Holyrood result will be somewhat different, but you may find that the STV element at local council level throws up some interesting surprises in the councils, which could have an impact two years hence.

On the BBC Parliament 1987 replay, I remember Professor Ivor Crewe saying that the only way you’d get PR in this country was if there were 2 hung parliaments in a row. It looks like there’s a fair chance that we might get that at the next 2 elections. If the chance of one hung parliament is 75%-90%, the chance of 2 in a row must be around 50%, because one can’t expect a huge number of people to change their vote in the space of 12-18 months. This was borne out by the experience in 1974.

Linking to some of the points already made, the following should be heavily borne in mind too…

http://politicalbetting.com/index.php/archives/2006/06/30/guest-slot-rod-cosbys-by-election-trend-analysis/

No opposition party, repeat NO party, has ever exceeded it average parliamentary by-election performance in a General Election, and recently (the last 30 years at least), the Opposition has suffered a consistent 4% SWING-BACK to the Government from its BE performance.

The current Tory by-election performance is about a 3.9% average swing (very similar to their opinion poll position, btw).

Several conclusions appear to flow from this fact:

Unless there is a dramatic Tory improvement in by-election performance,- say swings of 10-15% (which would still be only modestly impressive by historical standards) we could safely say:-

The maximum GE swing could not exceed 3.9% (not enough even to make the Tories the largest party)

The likely swing (factoring in the historical swingback trend) could be as low as 0% !!!, leaving Labour with a clear majority.

To counterbalance that, Hague scored a 1.8% swing in 2001, and Howard got 3.1% in 2005. So a Hague swing would mean goodbye Labour majority, a Howard swing would mean hello LibDem Kingmakers(although Labour still largest party).

All very confusing!!! except none of it good news for the Tories…

Labour largest party looking overwhelmingly likely - hung parliament likely would be a fair conclusion…..

ergo, Tories best chance would be a rapprochement with the LibDems?

There seems to be a lot of nervous messages on here afraid to actually admit that Labour will NOT win the next general election - polls clearly show that not only is Labour behind in the general polling consistently - The Conservatives are playing the “Scottish” card on Brown , making him less and less likely to be acceptable to the English voters - the Conservatives will make massive gains in England , they already were the majority party in the number of votes cast at the last general election in England. They will also gain more Welsh and Scottish seats + the boundary changes will all play into their hands .
My forecast if there were a general election tomorrow would be a small Conservative majority of about 10 seats / by the time the next election comes which will be after a FULL 5 year term for Labour as they will hold out to the bitter end will a Conservative majority in excess of 50 to 100 seats . When Callahan left office in 1979 he faired worse in the election because the electorate were just wanting him to go - but insread he hung on for the full 5 year term - just as Blair is doing now & Brown will do after him !!
Nobody has mentioned the very poor polling results for the LibDems - their leadership is in shambles - their voters will turn to the Conservatives where they came from in the first place at the last 3 general elections .

Some of then postings imply that the voters have the ability to prevent an hp by changing their voting behavoir - see Peter the Punter above but I have seen others on PB.com

I’m not sure how a voter works out how best to achieve this, particularly if the perception is that both the Labour and Conservative are “safe”.

I think the key group is the “Anything but the Tories” faction, who will have a swing from them and toward the Tories, but we don’t know the net impact in seats

Your post Mike shows that you are today’s king of wishful thinkers . I would not say that Labour will win the next GE but will not at this stage say that they cannot . All the current polls suggest a hung parliament with some but not all saying the Conservatives will be the largest party but you come along and ignore them all and say there would be a small overall Conservative majority .
The Conservatives were NOT the majority party in votes cast in England at the last GE , they were the largest minority with 35.7% of the vote barely more than 1/3rd .
Unpopular as Labour are now the feeling is more of circa 1990 rather than 1994/1995 and many past governments have recovered from midterm unpopularity before and many have said that James Callaghan would have too if he had called the GE in Autumn 1978 .
There is all to play for and as Rod as shown the most likely result of the next GE is a hung parliament .

Looking purely at the polling figures, we’re firmly in hung parliament territory, finely balanced as to whether Labour or Tories would be the largest party. The Tories need a roughly five point lead to deny Labour a majority, which is basically where they are.

Whether they can break through to the 10 point lead they need to have an overall majority, however, is another matter.

The Communicate Research January poll compared voting intentions with how people generally saw themselves. Of people seeing themselves as Labour, only 64% currently intended to vote Labour, whereas the equivalent figure for the Tories was 82%. Now, there are different ways of looking at these data. On the one hand, it suggests that Labour is doing less well at winning over its supporters (which is how the Independent chose to interpret it). On the other, it indicates that the Tories have stretched their support as far as it will go and have little head room for improvement. To go further they will need to break significantly into other parties’ core support. Labour has more room to to grow just by winning back people who think of themselves as Labour, and there are more of them to begin with (29% of the total electorate, including don’t knows, other parties and refused to say, against 22% for the Tories).

Then there is the question of what happens when it comes down to a Brown-Cameron fight. There are obviously arguments both ways, but I don’t know a politician who wouldn’t rather fight it from Government rather than opposition. My view is that the advantage is with Brown.

Overall, I’d say the Tories are well short of where they need to be, and my gut feeling is that we’re heading for something between a Labour dominated hung parliament and a small overall Labour majority.

Although it is not really mentioned here, the Conservatives do not need at least a 10% lead in the final result as critics say they need, in fact if the Conservatives won 39%, Labour 31.5% and the Lib-Dems 18%; the Conservatives would win a majority of 8 seats.

The Conservatives will likely see their share of the vote rise quite substantially but I think this will be mainly due to big progress made in London (which has been shown quite recently giving them a 7% swing in this region) and East Anglia making a swing from Labour. It is likely also that the Lib-Dems will loose at least 10 seats in the South West and a couple in the South East as will Labour.

The idea about the Conseravtive’s progress being sustained to the South East is untrue too. If current opinion polls were the actual result at the next election then the Conservatives would pick up 5 seats in Wales and around 20 in the Midlands. As someone mentioned earlier there are quite a few marginals in Lancashire like Bolton West that could fall to the Conservatives too.

Nonetheless, I would agree that Scotland and the North of England excluding a few exceptions like Dumfries and Galloway and Westmorland and Lonsdale are places where the Conservative’s prospects are quite low. However, their new Northern strategy may be enough to gain them 10 in the next election.

The Conservatives need a lead of around 11 points on a uniform swing, although it depends to some extent on how well the Lib Dems do.

I suspect you’ve been playing with Martin Baxter’s swing calculator, which uses an adapted proportional swing, rather than a uniform one. Round these parts we are, for self evident reasons, somewhat biased toward Anthony Wells’ swing calculator

Phil, not on Anthony’s figures, they wouldn’t. They’d be 22 seats (or 25 depending on how you calculate it) SHORT of a majority….
Please explain?

Here’s where your argument unravels a bit Anthony: you’re of course entirely right to point out that polls used to report very high Labour leads when in mid-term local elections they (whilst doing very well) never came close to replicating them. And in the 1992 general election of course the same thing infamously occurred: polls reported a much stronger Labour performance than emerged.

The changes in methodology that you rightly refer to were all about tackling this inherent *Labour* bias in polls - so it logically follows that if it the methodology hadn’t been changed Labour would be reported as doing much better than they are - possibly to the extent of retaining a lead; and there would be far louder cries about the unimpressive performance of the Conservatives.

In other words, contrary to being a dismissal of any comparative arguments about the Tories’ performance your “methodology” argument actually adds weight to it: without Labour’s vote being incorrectly enhanced the Conservative lead should be even greater at mid-term than it actually is.

The other problem (accepting your point about unforeseen cataclysmic incidents transforming the political landscape) is that precedent shows that Governments regain support between their lowest mid-term position and polling day. For governments to go up, the share for the opposition parties (and/or undecideds) must go down.

It’s entirely plausible that next time any government gain in support will come principally from the minor parties and not the Conservatives, given the historically sizeable share these parties are currently being polled at.

My issue with Conservative supporters overconfident about anaemic leads for their party is not so much that they are overjoyed at being in front but their failure to grasp that political strength is cyclical within a parliamentary term and not on some ever-diverging lines.

It will be unprecedented if Labour does not pick up - probably a little, possibly a lot - between now and the next general election.

Adam,

I think you are missing the point, in that you seem to be suggesting a fixed relationship between the Tories and Labour, where as Anthony is talking about the relationship between government and opposition.

Polls in the eighties and nineties almost certainly overestimated mid term opposition leads because of the methodology used, and probably don’t now.

What we can’t say for sure is if there has been an over correction, in that it may have been that just as 20-30% leads were more like 10-15% leads, the 5% lead we are seeing now, could actually be a 10% lead.

This could be because of over correction so that when the election comes, the shift will actually be towards the opposition as opposed to the government.

When polls didn’t correct for intention to vote a lot of people who said they would vote Tory didn’t vote, But we could be in a situation now where a percentage of those who right now are saying they are unlikely to vote might actually come out and vote Tory (or indeed Labour).

In effect we know and understand the mistakes we used to make and have tried to correct them. What we don’t know is if we have done so successfully or if we are making alll together different mistakes now.

Peter.

Forget polls. REAL votes tell a better story….
http://politicalbetting.com/index.php/archives/2006/06/30/guest-slot-rod-cosbys-by-election-trend-analysis/

Well done, Rod. I’d not seen that article before (must have been away). It certainly cuts through a lot of the verbiage and puts the current state of the polls in perspective.

Peter, I am indeed asserting that this isn’t a government vs. opposition question but a Conservative vs. Labour question. I’ve seen no evidence - or until this discussion even heard the argument advanced - that it’s the former.

What possible intrinsic reason can there be for the high Labour leads seen in the 1990s being an opposition bias rather than a Labour bias? In opposition to such a theory let’s consider that:

1) Final polls since 1997 have been spot on in forecasting the results; that suggests that polls throughout the term have been more or less accurate, does it not? Why should they suddenly be wrong again?

2) Even when polls have been out, the problem has been far more acute in overstating Labour’s share than understating the Conservatives’ - Tory share has actually tracked rather well whilst Labour’s has ricocheted all over the place.

3) Comparing council election share with opinion poll share we find a far closer (though not accurate) track: Labour has polled at or around 28% in the three most recent sets of elections excluding 2005, while it’s national share has been between 31% and 36%. That’s not inconsiderable, but compared to the 1995 council elections when (from memory) Labour’s actual vote share lead was something like 16% but the opinion polls were showing 30-40% Labour leads.

Again, in these elections, Conservative share has tracked very closely to their opinion poll share; I’d also advance a theory that Labour always does worse in council elections than in national elections so that even the slight lag between its actual and polled shares can be explained, but not something to go into now.

So the only remaining issue is how the growing pool of undecideds (and indeed support for minor parties) might “corrupt” the precedent argument I’m making. It’s certainly a factor, but again, it’s surely for you to come up with convincing arguments to explain why, unusually, undecideds may break in proportions any different from the ways they normally break.

Conventional thinking is that most undecideds who do actually vote plump for the opposition, but not from where they are at mid-term: instead from the start of a campaign. By the start of the next campaign I think the undecided block (and the minor party block) will be substantially smaller than it is now. But I am highly sceptical that it will be substantial enough to explain an “over-correction” in polling.

“Some of then postings imply that the voters have the ability to prevent an hp by changing their voting behavoir.” Yes, John. How else would they prevent it?

“…The key group is the “Anything but the Tories” faction, who will have a swing from them and toward the Tories…” and presumably not towards Labour, LibDems, Greens, BNP, Flatearthers etc. Evidence?

From the link: “The difference is that Labour often get in because the opposition vote is split between the Conservatives and Lib Dems, while the Conservatives normally don’t get in because their opponents vote tactically for the party best placed to stop them.”

It used to be the other way round in most of Scotland outwith Glasgow. The change can’t entirely be blamed on Thatcher, because it started well before her time, though she undoubtedly has done more to further the cause of the SNP than Alex Salmond, and I think even he (modest and self-effacing soul that he is) would agree with that.

It remains to be seen what the outcome of the SP election will produce in the way of loss of confidence and morale within the Scottish Labour Party, if they do badly. That could reduce the contribution to Labour’s majority from North of the border.

They may well sustain significant losses in May, including the psycological and financial loss of office, and it may be that some of the same FPTP constituencies will be lost in the Westminster election.

Independence may well remove it altogether before the following election. The SNP allways fall back a little in the polls just before the vote, but then that doesn’t matter if they are far enough ahead to start with, does it? How far is far enough?

Can someone please explain to me why the assumption that Labour will improve between now and the general election? Why the assumption that governments “always” recover after the middle of the Parliament?

Yes, I understand that the Tories did when in government, but that doesn’t prove that a Labour government would.

Labour has been re-elected twice already, as far as I can read the figures it appears to me that the governments figures come election day were *worse* than most polls at the middle of the Parliament, not better.

To continue my post, in every set of figures on this site it appears to me the mistakes in the polls are not biased towards the opposition. Regardless of whether the Tories are in government, or Labour is, the bias has always been towards Labour. Come election day the government may have done better in the eighties. But post-’97 come election day the opposition party has done better.

Might the bias not be a party bias, rather than a government/opposition bias?

Philip

Try following Rod Crosby’s link. I think it might help you.

RodCrosby Says:

February 13th, 2007 at 2:16 am Tories best chance would be a rapprochement with the LibDems?

Not at all, the SNP not the LibDems.

Then they could be the minority partner the SNP needs if the Greens don’t get enough seats.

The Scottish Party is almost independent now.

The English Conservative party should split completely from it and become the English National Party; collect the votes of racist former non-voters; insult the Scots; campaign for an English Parliament and/or independence for Scotland, keep mentioning the West Lothian Question and the fact that health, education and local government policies for Gordon Brown’s constituents are determined in Edinburgh.

I think the key group is the “Anything but the Tories” faction, who will have a swing from them and toward the Tories, but we don’t know the net impact in seats

There are many who vote against Labour too.

The old class/council house/culture/political division is gone. There is a danger that party managers and pollsters if they fail to recognise that elections are no longer won or lost by persuading your “natural supporters” to come out and vote for you. The last nine elections at least have not been won by any party. They have been lost.

Denis Canavan’s results show that it is no longer true that most people identify with and vote for a party. In Falkirk West, that is even true of those who go to the polls. A “good” candidate used to be worth no more than 1,000 votes, but in Tatton a “bad” candidate was worth 10,000 Conservative [sic!] votes.

Many vote for the least worst option as they perceive it at the time, and few except the tiny number of party members actually believe as a matter of faith, the political philosophies of the parties they cast their votes for. This is not unlike the decline in religious belief and church attendance. If it wasn’t for the Apathy Party getting the majority vote, voter churn would result in frightening degree of instability.

Peter: Rod’s figures refer to by-elections and not polling data. That is comparing apples to oranges.

Look at the real polling data. In 1998 virtually every poll gave Labour a lead over 20%. Throughout 1999 the polls gave Labour typical leads between 20-30% still. In 2000, barring the “fuel crisis” abherration, Labour has double digit leads. While in 2001 EVERY SINGLE POLL except for just one gives Labour double digit leads. Many polls over 20%

The real result was a Labour lead of “just” 9.0%

So where is this mythical swing back to the government from the mid-term polls?

——-

In 2005 Labour won with a lead of 3.1%

In 2002, not even one poll ever gave Labour a lead of 3 or less.

Yet most of you guys are saying the 2006/early 2007 results should be viewed from a fact that historically there will be a swing to the government. That’s not just happened in either Labour Parliament, there have been massive swings from those early poll results to the opposition by election day since Labour came to power.

——-

John: I don’t want my party to appeal to “racist non-voters”. I agree on everything else though.

Another thing that’s not been factored in here is changes in the leader of the opposition.
The 1993/4 polls that Anthony quotes were with John Smith as Labour Leader. After his death, the Labour poll leads increased hugely once Tony Blair became leader.
Similarly, in 2002, the year that Philip mentions, IDS was still leader of the Conservative party and appealing to absolutely no-one, least of all the Conservative party. When he was ousted and replaced by Michael Howard, again the Tories recovered in the polls.
But in 2007, there is no such trump card to be played. The Tories are not about to change their leader and cannot expect a boost such as they enjoyed in 2003 and as Labour did in 1994. Indeed, it is the party of government that look set to unveil a new leader. Therefore on the face of it, nothing seems too likely to make their poll leads greater than they currently are.

Philip, I completely disagree: comparing polls with REAL elections really IS comparing apples to oranges.

Remember, by-elections are REAL votes cast by REAL voters in REAL (identically comparable) constituencies, in elections that are both touted and understood by electors to be REAL verdicts on the incumbent central government (unlike both local elections and Euro elections, for example, which are capable of being approached and viewed entirely differently by the electorate).

Parliamentary by-election results are perhaps the best indicators that we have of future general election performance.

Labour has been in power for almost 10 years. It is a stark fact that during this time, the principal opposition Conservative party has failed to register a single by-election gain, the longest such hiatus -by far- in recorded electoral history. Moreover, it is hard to find a more universally dismal set of results ever to be recorded by an opposition party. For the record…..

Leader:HAGUE
Beckenham by-election, 1997 - Conservatives narrowly hang on to seat after member resigns amid scandal. -

Winchester by-election, 1997 - Conservatives turn a 2-vote deficit into a 22,000-vote deficit after appealing original result. -

Romsey by-election, 2000 - Conservatives lose a seat while in opposition for only the second time in 70-odd years. –

Leader:IDS
Ipswich by-election, 2001 - Conservatives fail to gain a seat they have held in recent memory. They lose votes, rather than gain. –

Leader:HOWARD
Leicester South by-election, 2004 - Conservatives fail to gain a seat they have held in recent memory. They drop to third place. -

Hartlepool by-election, 2004 - Conservatives plunge from second place to fourth place for only the second time in history (and for the first time when in opposition). -

Cheadle by-election, 2005 - Conservatives fail to regain a seat they lost by only 33-votes four years previously. –

Leader:CAMERON
Blaenau Gwent by-election, 2006 - Conservatives finish in fifth-place for the first time in history. -

Bromley and Chislehurst by-election, 2006 - Conservative vote plummets in own seat by largest amount in history. Seat almost lost……

There is a pattern there, isn’t there…… Brutally put, the Tories are not real contenders for government. Their realistic short-term goal is survival and consolidation…..

After three general elections they are still stuck at sub-200 seats. i.e. worse than Labour was in 1983, and it took Labour 14 years to come back from that (NB, with an electoral system stacked in their favour!).

However, unlike Labour’s long road back, which was characterised by successive leaders of increasing stature, the Tories have continued to select weirdos and political pygmies of whom Cameron is but the latest ridiculous example.

The current swing to the Tories in opinion polls is about 4%, almost exactly the same as that achieved in by-elections since 2005. In other words, Puny and Pathetic….

NO opposition has ever gone on to do better than its by-election performance. So on current form, Labour are in absolutely NO DANGER of being supplanted as largest party by the Tories at the next election. The betting market is seriously out of kilter with reality.

Until the polls show the Tories consistently 10-15% ahead of Labour, and they are obtaining by-election swings consistently in excess of 10%, politics in this country will remain distinctly uncompetitive.

However, current opinion polls have consistently indicated a hung parliament, usually with Labour as the largest party. Perhaps the swingback to Labour at the end of the third term will be modest…

Tory leader Hague got a 1.8% swing in 2001: Howard achieved 3.1% in 2005. If the Tories get a Hague swing again, it’s goodbye Labour majority, if they get a Howard swing it’s hello LibDem Kingmakers (although with Labour the largest party by far)

So, in conclusion, all the current evidence suggests the Tories cannot possibly become the largest party, let alone win a majority, at the next election. However, they would have to be doing appallingly badly not to deprive Labour of THEIR majority. So, at the moment, a hung parliament with Labour the largest party is the most likely outcome.

The Tories could of course improve their position over the coming year or so. However, an obstacle in their way is an inevitable change of face of the Labour government. I would not bet against that alone giving Labour some kind of a boost….

Forget it, Rod. Philip sees only what he wants to see. We should really take pleasure that there is another mug punter out there rather than try to enlighten him.

Peter the Punter
“Some of then postings imply that the voters have the ability to prevent an hp by changing their voting behavoir.” Yes, John. How else would they prevent it?

Imagine I am a voter and I want to use my vote to stop a hung parliament. Please advise me what I should do? Vote for the party of my choice, vote tactically, not vote at all, persude other people to vote. I am but one vote in millions.

I can avoid a an hp by electing either labour or conservative outright, am I going to vote for the party I most oppose in order to avoid an hp.

The only variable that does come into play (imho) is turn out - see 1992. BUT highest turnout in recent times was Feb 1974 (I remember it well waiting for the eleventeenth recount at Bodmin), when we had, you guessed it - a hung parliament.

Oh and another thing Peter the Punter

““…The key group is the “Anything but the Tories” faction, who will have a swing from them and toward the Tories…” and presumably not towards Labour, LibDems, Greens, BNP, Flatearthers etc. Evidence?”

Fairly obvious really, the Conservative majority of the 80’s has been picked off from four sides - Labour, Lib Dem and Nats x 2. We have had less of the side to side swing of the 50’s and 60’s. This implies an “anything but the Tories” vote and meant they lost safe-ish Tory seats like Argyll and Bute, which Labour would never win. Before 1974, Others/Libs was a dozen seats or less.

To call it tactical is probably incorrect, as it is (again imho) demographic reflecting our nation’s greater diversity. Cameron’s strategy is surely to pull these guys back as the least wedded to another party which he seems to have had some modest success. It was also the first step in Kinnock’s strategy in the 80’s - get back the guys that went to the SDP.

I note that a report today suggests the SNP wants a deal with the LibDems. Labour out is what they want!

Will the same type of alliances (e.g. Con/LibDem) gain momentum for the next General Election?

Philip

“Imagine I am a voter and I want to use my vote to stop a hung parliament. Please advise me what I should do?”

OK - Let’s imagine you have a vote in, say, Guildford - a highly marginal constituency. You favour the LDs but only slightly. You don’t like the idea of a hung parliament though. You believe that the more seats the LDs get, the more likely a hung parliament is. What’s more, you think there’s a chance that the Conservatives, who are your ‘next best’ option’, might just win an absolute majority in the House, and the polls suggest it’s going to be close. Now, you think to yourself ‘Hmmm..I would vote LD but it kind of helps towards an HP, so I think I’ll vote Conservative instead.”

That’s how it works, Philip. Of course, it’s only one vote but then if there’s a genuine widespread concern about an HP, enough other people might think similarly for an HP to be avoided. It’s not wholly satisfactory, I know, but until such time as the constitution is changed to allow you personally to select your own Government, it will have to do.

Your other point is baffling. If I am an ‘Anything but the Tories’ voter, surely I work my way through the whole spectrum of alternative possibilities before, in my desperation, I vote blue?

Maybe all you mean is that some people who voted Labour last time because they were sick of the Tories are now sick of Labour and will vote Tory again. This is true, if somewhat unenlightening.

Peter

(”John” not Philip by the way - but hey what’s in a name, they are all made up anyway)

I think the complexity of your response makes my point. That is how us anoraks speak - not John and Jean Voter who are not at that level of sophistication in sufficient numbers.

Fear of the Opposition party is a basic enough motive to influence - 1992 I assume is the measure. Fear of an hp?? Not when the parties are so ideologically similar and there is no obvious threat to stability (poll tax riots or miners strike)

Sorry if the other point is beneath your dignity. I’ll go off to the children’s guide to elections website, sorry to intrude on your time

John

Apolgies for the wrong name, made up or not.

Most voters may not be at ‘that level of sophistication’, but some are and enough to make a difference. Without them, Anthony and his fellow professionals would be out of work.

Anyway, I am pleased to note we agree on something. A while back I argued, on politicalbetting.com , that the conventional wisdom that the LDs get squeezed in a tight GE contest might not apply next time round. I took a fair bit of flak for doing so, even though it was tentatively suggested as no more than a possibility. You state succinctly one reason why that conventional wisdom may have to be ditched next time round. If it does, it increases of course the probability of a Hung Parliament - and that’s without Rod’s fancy maths which already suggest something approaching 90%.

The ‘other point’ might be beneath my dignity if I could understand it. I’m still struggling to see why an ‘Anything but the Tories’ voter would vote for them in preference to any of the alternatives on offer. Maybe it’s me that needs to consult the children’s guide to elections website.

“There seems to be a lot of nervous messages on here afraid to actually admit that Labour will NOT win the next general election”

Err I think most people had stated that was the most likely outcome before your statement.

I’m disappointed the discussion has progressed this far without any discussion of differential swing. There will not be a uniform swing, or a proportional swing. If - and it remains a big if, hopeful as I am - the tide does stay with David Cameron and the Conservative Party, then there is likely to be a disproportionate swing in seats with high numbers of swing voters. A lead of 6% or 7% might be enough for a Conservative majority, whatever the uniform swing calculator suggests.

Firstly, I am Peter Cairns, and have been all my life….. I’ve never been that confortable with the whole idea of not being open about who you are and what you believe in.

Secondly, I think in a close election the LD’s will be squeezed. There may well be a core of LD supporters who see a HP as a great opportunity for real power and electoral reform, but as with Labour and the Tories, the majority of LD voters do so out of allegience or habit, more than as a result of political tactics.

The issue for the majority of the public will be “Who will be in power, who will be PM”, and that will be good for both Labour and the Tories at the expense of the LD’s.

The Tories seem to be benefitting ( as the SNP are) from a “Time for a Change” mood and the LD’s aren’t. I also expect the BNP and UKIP to get squeezed too.

Whether it is enough to get the Tories in is another matter.

Of course depending on the state of Labour and Tory finances come the next election, the prospect of a hung parliament and an autumn election just after a spring one (if there has been no party funding reform) may make Labour in particular pretty amenable to a deal with the LD’s.

Peter.

‘Parliamentary by-election results are perhaps the best indicators that we have of future general election performance’.

As the result of a by-election is unlikely to determine which party runs the country logic would suggest that people are more likely, one not to vote, and two to vote abnormally. How can you then rely on them to indicate anything?

Peter Cairns,

I find myself agreeing with you yet again. As election day approaches the LDs will be squeezed. This will be particularly true in England where voters will see one Scottish leader keen to prop another Scottish leader in power. It is obvious that an LD vote will produce a Lab-LD coalition - led by .. Brown.

This is the main reason why any tactical voting is likely to produce very different responses at next election. In 1997 - and possibly to a certain extent in 2001 - it was understood that the LDs were more likely to support Labour in a hung parliament than the Conservatives, so voting LD to get/keep Tories out made sense. On the other hand, this does not work if the aim is to get Labour out - hence why LDs made net losses to Conservatives in 2005, and have done so consistenly at local level since about 2001.

Rod Crosby suggested some way back that the best hope for the Tories was a rapprochement with the LDs, but I think that John Dick’s idea of an alliance with the SNP might be more productive - especially in the short term.

Only snag is, although the electoral arithmetic appears attractive, is it remotely feasible politically ?

Ralph,

Spot on,

On the basis of Dumfermline East in the Scottish Parliament we can confidently predict a LibDem landslide at the next general election when Ming Campbells troops sweep Labour aside.

Equally the Moray By-election shows that come May the SNP will pick up 75 seats at Holyrood…..

As we all know, neither of these is going to happen.

Peter.

Ralph / Peter,

Absolutely ! Do you remember Peter Snow’s custard spills during the last years under Major ?

I disagree with Rod Crosby’s analysis / prediction simply because the by-election data since 1997 has actually been too sparse and unrepresentative. This may have been a by-product of the large number of relatively young MPs first elected in 1997, and Blair’s ritual clothing in ermine of the old guard subsequently.

The message from Council elections on the other hand shows a clear patten of LDs supplanting Tories as key challenger to Labour in the former industrial cities, but losing ground elsewhere as Tories steadily build up their English council base. It will be interesting to see whether there is any change to the former pattern in English Metropolitan councils this May. If so, that spells seriously bad news for Ming. (Though I bet the spin machine will focus on LDs winning more votes than Labour in England)

Ralph/Peter: I did say: “Parliamentary by-election resultS….”, meaning, if you haven’t already read the article, taken COLLECTIVELY….

No-one in their right mind would forecast the general election on the basis of a single result, let alone a LibDem “surprise.”

However, you either accept the FACTS shown in the graph, or you don’t.

NO party has ever done better than its AVERAGE by-election performance. In FACT, ALL have done worse….

So, what’s yout point?

Toby: If the changes from 2002 and 1993 were changes due to changing leader of the opposition (a very valid point), then how come everything else I wrote?

How come every single poll in the election year of 2001 over-estimated the Labour lead? How come in the first Labour term, where there was no change in the opposition leader after the start, virtually every poll in the entire life of the Parliament (barring the fuel crisis) over-estimated the Labour lead?

RodCrosby: “Philip, I completely disagree: comparing polls with REAL elections really IS comparing apples to oranges.” Maybe, but this is a Polling Report blog, this thread was about how far ahead in the polls the Tories should be, and that is what I am talking about. The Tories in every election out of the last three decades have done better than was predicted by the polls in the mid-term, whether they were in government or opposition. So being ahead now is quite significant I think, given that historically the Tories (and not the government) improve from the poll results.

I haven’t touched your obsession with by-elections, but if you must know I find your “rule” to be a clear case of “post hoc ergo proctor hoc” and as believable as the idea that the winning party at the election is simply the party who’s colours match that on the shirts of the FA Cup holders. Two by-elections does not represent an entire nation.

Rod: “However, current opinion polls have consistently indicated a hung parliament, usually with Labour as the largest party. Perhaps the swingback to Labour at the end of the third term will be modest…”
You conflate the ideas here of the poll results and a “swingback to Labour”. Which was the last election in which there was a “swingback to Labour” from where the POLLS stood at this point of the Parliament, and the actual election results? The Tories have done better than what the polls have said at every election for decades.

‘I did say: “Parliamentary by-election resultS….”’

Which show results that are the reverse of the poll data. And that’s including Blaenau Gwent and Bromley et al which are clearly strange results.

I think we’ll have a better idea of where things are going once Blair leaves Gordon Brown (or whoever) takes over. There are substantial numbers of natural labour supporters (like me) who are heartily sick of Blair, who want to see Labour in power but with sensible policies. If his successor drops things like ID cards, doesn’t do another Iraq, maybe think again about road pricing, sort out the home office shambles and make sure taxes at least don’t go any higher I’d be happy to start voting labour again.

We have all yet to see a public analysis of just how votes for minority parties might work out in some of the large conurbations. There was a little noticed election in Bede ward of Nuneaton last week. It WAS a local government result, but it gave all three big parties food for thought. I don’t think the past is much guide to the future in similar areas simply because of the local effect of the presence of large numbers of non-white residents in some areas. That may not be very palatable, but it is the case.

Yes, a poll for the Daily Telegraph on 9th January 1995 gave these laughable figures:

Labour: 62%
Conservative: 18%
Lib Dem: 14%
Others: 6%

How they could even print such a poll is mindblowing. Were they so incredibly ignorant about the real parametres of voting behaviour, that a winning party’s glass ceiling is 45%, and a losing party’s ‘glass floor’ is 25%, that they could not see the unlikelihood of their poll being accurate?

To compare pre-1997 polls to modern polls is just completely absurd, and quite frankly pre-1997 polls shouldn’t be given any attention because they were so laughably ridiculous.

comparing by-election results to polls is also stupid. by election results are votes cast in demographically very narrow places. A by-election in blaenau gwent, which is a very deprived Welsh seat who’s local sport is burning Tories, does not have the same validity as a properly weighted opinion poll.

Looking at any by-election result, votes will never be cast in the same way as in a nationwide general election.

I don’t think anybody is trying to compare by-election results with what might happen at a parliamentary election - General or otherwise. What is stupid, however, is not to recognise that polling patterns that existed up to th 1990s may no longer hold true. The point that I am making is that while it is unlikely that such as the BNP will win parliamenary seats in the foreseeable future, although watch East London, they do get noticable scores in some seats - eg just short of 10% in Keighley in 2005. It is the effect that this has on individual seats that is interesting and as yet unknown.