The value of polls can go down as well as up

Contrary post of the day: the massive poll leads recorded in recent days make me nervous. It's good to have a lead, of course, and despite protests from various quarters, Cameron has in fact always had one apart from Brown's honeymoon bounce. When the history of this failing, tired government is written, I hope it will record the important fact that ICM's poll of a 6% lead for us in fifty marginal seats came before Brown's Yellow Saturday moment; caused it, indeed, and was not merely an effect of it.

But....

These shifts seem to me to be very fast. And Ben Brogan records that Lord Ashcroft is talking about a still impressive seven percent lead.

Labour, surely, is bound to recover from these lows. We all need to keep steady when strange polls are recorded (only a few weeks back one firm had Labour with a one point lead). I hope the hares will not start calling for the tortoises to get their skates on if these massive leads stabilise. You can almost hear it now: "Cameron couldn't sustain the momentum! Give me a real Conservative!"

My suggestion: forget the polls and campaign as if we were, say, three points behind; hard work in front but everything to fight for.

Comments

[Deborah]

Absolutely right, Louise. Neil Kinnock learned the hard way that taking too much notice of polls can be dangerous

[Robert McIlveen]

It's almost impossible to imagine the election result could be any better than the YouGov one in the Sunday Times - while it would be nice, it will be very easy to get disappointed when the lead (inevitably) ebbs and flows.

"If you can meet with Triumph and Disaster
And treat those two impostors just the same" then we'll have a much happier time of it!

[Christina Speight]

Leads MAY move fast - but this misses the nature of publicity for polls and the media that carry them. A ONE percent shift from Lab to Tory double to TWO percent when the lead figure comes up. This makes a better headline

ALL polls in advance of an election being declared are a valid snapshot of undercurrents but must never be taken as predictions. Once polling day is announced, they DO mean something electorally.
(end of this ex-researcher's lecture!)

[Letters From A Tory]

I refuse to believe that Labour will be anywhere other than several points higher than this at the next election. The Conservatives should be encouraged but not overly excited by this poll.

[Tony Makara]

Louise you are so right. We must not suffer AC MIlan syndrome and start kissing the trophy just yet. In the champions league final AC Milan were leading Liverpool 3-0 at half time and many of their players kissed the trophy as they passed it on their way to the changing rooms for the halftime break. As many will know Liverpool came back and won the game. Many of the AC Milan players now believe they lost that game because they got too complacent because their lead was so big. So AC Milan syndrome must not happen to the Conservative party. A 16% lead in the polls is enough to make anyone think "Game over" but that would be a mistake. Labour have won three elections back-to-back and they must be respected as skilled campaigners. Your cautious warning is very wise Louise.

[Northernhousewife]

Agreed Louise.
The danger is people will complain when this lead inevitably receeds and demand a change of course.
Probably the same people who voted for IDS or would have voted for DD.
I am not seeking to be disrespectful to these men who are subtantial politicans, but they were not the right course for the leadership.
All I am saying is you haven't always judged correctly.
The current guys have done OK. Trust them

[Man in a Shed]

What the poll volatility perhaps show is how worried the public is. Trust and faith have been lost in Labour, but aren't yet transferred to us.

We must explain why Labour have failed to show that we understand the problem. Proposing the solution should perhaps wait till an election date is known as currently the Conservative Party is Gordon Brown's favourite think tank.

[David Belchamber]

Louise is quite right in what she says, so the all-important thing is to put up people and policies that are clearly superior to Nulab's; then we will deserve to win.

I suggest that we have some way to go in both regards but we are definitely making significant progress in the right direction.

[David]

A serious plan to reclaim powers from Brussels after an election is needed. Without that what's the point of the rest?

[Buckinghamshire Tory]

Indeed, there's much work that needs to be done before the victory is secured.

[Yet Another Anon]

Not only can polls go down as well as up, but they can be well out and remain well out for a long time, because the whole premis they are based on is flawed.

They assume a margin of error within + or - 3%, this may be true so far as what the population would say to someone asking them about their voting intention, people though lie to people conducting polls or hold one opinion, but are embarrassed to admit it, there is also a tendency to favour incumbents and especially where there are doubts about the alternatives and the incumbents have been in place for some time - so opinion polls could show a governing party being behind throughout an entire parliament and yet the government could win an election.

After a party has done well in an election there is a tendency at least briefly for opinion polls to register a sharp swing to that party after that event, as happened after the 2005 General Election for example, Local Elections for some years now have not correlated much with General Election results, nor have European Election results, in parliamentary by-elections Labour has actually done quite well since 1997 compared with past governments and the Conservatives have still not taken a single seat off another party in a by-election since the early 1980s - the opinion polls are most useful to the media to sell papers, magazines etc... and justify their existence, otherwise in the main their actual usefulness is questionable unless they can more closely approximate to the actual conditions of the elections they are polls of. They will always be limited by the fact that people know that they are not a real election and so even if they try honestly to give what they think they would say they could not be sure that that would be the same in an actual election.

[Tony Makara]

There are still a number of things that I'm not happy about. Europe, lack of a promise of tax-cuts, single mums being forced to work, etc but I'm going to be supporting every Conservative candidate in every constituency at the next election. Let's get the change of government that our country needs and we can we have contention on specific issues afterwards. Our country cannot afford a fourth Labour term, it would be a fast-track to statism!

[Andy Stidwill]

I agree that polls which exaggerate the situation are not particularly helpful, since they can always lead to complacency in certain quarters. In some ways it's better to have a whole series of polls showing the party doing fairly well rather than a few giving spectacular findings.

[peterbuss]

Hang on - this seriously good news - even with all the above caveats on Opinion Polls.
Generaly the Govt of the day gets a boost post budget - these polls show the reverse.
Of course there's a load of hard work we must do before the election which is still up for grabs but for goodness sake until Cameron came on the scene we never got above 33-34% in ANY Opinion Poll. Now we sniff at 40%.That figure for the Tories has now become pretty much routine and its just amazing that we are (rightly) demanding highe support - for years we would have died for a poll which gave us 37%.Interesting also that Tory suppot increases the more Cameron sticks to the middle gound on tax and the fading again of Europe from the public debate.

[peterbuss]

Hang on - this seriously good news - even with all the above caveats on Opinion Polls.
Generaly the Govt of the day gets a boost post budget - these polls show the reverse.
Of course there's a load of hard work we must do before the election which is still up for grabs but for goodness sake until Cameron came on the scene we never got above 33-34% in ANY Opinion Poll. Now we sniff at 40%.That figure for the Tories has now become pretty much routine and its just amazing that we are (rightly) demanding highe support - for years we would have died for a poll which gave us 37%.Interesting also that Tory suppot increases the more Cameron sticks to the middle gound on tax and the fading again of Europe from the public debate.

[Yet Another Anon]

but for goodness sake until Cameron came on the scene we never got above 33-34% in ANY Opinion Poll.
The Conservative Party did get above 35% showing in Opinion Polls under every leader it had in opposition, they even got above 34% in one or two in the latter half of the 1992-97 parliament - there was the fuel crisis in 2000 when the polls briefly showed the Conservatives at 38% leading Labour, and under Michael Howard at 40% and under IDS there were numerous ones showing the Conservative Party at 35%+.

In Local Elections and European Elections from 1999 on the Conservative Party has consistently been getting 35%+ (the only exception being the 2004 European Election when the Conservatives were the largest party but only on 28%).

[Chad Noble]

It's going to take an uber effort to fark up this kind of lead Louise.