ComRes: Tories 7% ahead

More tomorrow.

8.15am, 1/4: "Only 49 per cent of Labour supporters are "absolutely certain" to vote in a general election, compared to 71 per cent of Tory supporters and 55 per cent of those who intend to back the Liberal Democrats." (The Independent).

Comments

[Kevin Davis]

These Comres polls are very volatile, considering back in November they were telling us the Conservatives were 13 points ahead - the largest in 20 years!

That said, they appear to be in line (if not the same) as the Conhome poll of polls.

[Afleitch]

Pretty much to be expected now the post-Budget rucous has died down. Worth keeping an eye on other polls this week and next.

[StevenAdams]

The polls were bound to spike at some point after the budget - the key is to ensure that they keep trending upwards, and that means a good balance of positive policies as well as capitalising on Brown's ever-more shifty leadership.

[Opinicus]

The lesson is still the same

We are not safe from the natural volatilities of politics until there is a reason to vote Conservative.

Despising Gordon Brown in particular and MPs in general is a reason not to vote, not to vote Conservative.

I think the message is getting through but it is terribly hard pounding. We've wasted most of the last six months since the IHT announcement. If Brown's government wasn't being pursued by the Furies we wouldn't be even this far ahead.

[Jonathan M. Scott]

It's not the ComRes poll (or any other poll) that is volatile at the moment - it is the electorate!

[Sammy Finn]

Why is it that Conservativehome only gives prominence to polls that show the Tories way ahead?

[Conand]

Sammy@1:14

ConservativeHome reports all the polls.

The reason we are way ahead on most of the recent (prominent) polls is because ..er.. WE ARE!

[Jack Stone]

Sammy its the other way round normally. This site is made up of people who view themselves as semi-detached Conservatives!

[Northernhousewife]

Anthony Wells' comments on Uk polling are worth a read. Apparently this poll was not 'very Tory', but then it was not 'a very Tory sample'.
Rather than paraphrase and be accused of seeing what I want, I will copy and paste the man himself.

'This is not, necessarily, a change in public opinion though. Long time readers will remember that we used to coment on ComRes’s lack of political weighting. Well, they do now weight by past vote and have done for several months, but the way they do it seems very strange since their target weights change significantly from month to month. ComRes aren’t unusually pro-Labour, which was one result of not weighting politically, but they haven’t seemed to have made their samples much less volatile, which is normally seen as the benefit of political weighting.

In last months’ poll that showed an 11 point Tory lead, 19% of the total sample (including non-voters) said they had voted Conservative back in 2005. This month, only 17% of the sample said they voted Conservative back in 2005 - so, the Tories are down, but it’s a less Tory sample.

[Northernhousewife]

More from Anthony...

- when ComRes first started weighting by past vote they produced the figures that were easily the most generous to the Conservatives.

But - because they use a figure based on a rolling average of only 4 polls, rather than 10 polls like Populus or 20 polls like ICM - their target weights are very variable. Recently their weighting has not been very favourable to the Conservatives at all.

[gezmond 007]

Oh , that expailns it Northernhousewife , you are typical of the people on here who dont see the wood for the trees as you are so blinded by Dave and his mates.

When Comres polls were ok for the Tories these were great polls , now they are not you have found a reason why not.

Keep walking into oblivion !! Beliving your own publicity !

[Peter Harrison]

Gezmond - Northerhhousewife hasn't found a reason to disbelieve the polls. She has highlighted comments by Anthony Wells, an expert pollster, on why ComRes are still one of the most volatile pollsters and why they are one of the pollsters least favourable to the Conservatives at the moment. She could also have added that the fieldwork was carried out over the Easter holidays. There is plenty of evidence that the Conservatives do worse in polls taken over holiday periods.

It may be that this poll is right. However, it is so out of line with other recent polls that it may be a rogue. We'll only know when further polls appear.

And if you think a 7% lead is somehow a disaster, you really need to think again. That is a comfortable election winning lead. I now that isn't what the seat calculators say, but they are based on uniform national swing. There is plenty of evidence that this breaks down when the swing is large. Since WW2, a lead of 6% in general election votes has always been enough to give the party concerned an overall majority.