Could Nick Palmer be right about the next round of surveys?

In a post last night Nick Palmer – the Labour MP for Broxtowe and the first of any party to contribute to Politicalbetting under his own name – made an extraordinary prediction about the next round of polls.

This is what he wrote: “I think we can all agree the media stories this week are mostly helpful to the Tories, so do Tories here expect to be back in front in the next poll? And if so, do they expect it to last? My own guess is that it’ll be pretty even in the next poll, but the Tories shouldn’t get too excited yet as they won’t get that much useful coverage every week.”

If Nick is right then we could see big changes on several betting markets including the timing of the general election and the spreads on how many seats each party will win. There seems to be a lot of money about ready to go on the Tories at the first sign of good news for them and my reading is that we could see a market over-reaction.

If you think that Nick’s analysis is correct then there could be short-term profits to be made.

There were some signs in the most recent poll – Monday’s ICM survey on attitudes to an EU treaty referendum – that the Labour position might just have eased up.

For although not all the pollster’s normal voting intention questions were asked some were and the detail has been published on ICM’s website. What ICM did ask was how people intended to vote and what they did last time. But no past vote weighting calculation was carried out and there was not the usual filtering by certainty to vote.

But – and I must warn that we are entering very dangerous territory here – the raw data showed that fewer respondents said they planned to vote Labour next time than at the last general election. The split, and remembering that this is raw data, was 270 saying they’d voted Labour in 2005 compared with 254 saying that they would next time.

Looking back at other ICM polls since Brown became Prime Minister and you find more saying they would vote Labour next time than in 2005. Thus in the ICM poll for the Sunday Mirror earlier in the month the raw data had 233 saying they had voted Labour last time with 257 say they planned to at the next election. There was a similar pattern in the July Guardian poll which like the one in the Sunday Mirror reported in its headline figures a Labour margin of 6%.

In fact in almost every single ICM poll for the year or so before Gordon became Prime Minister the raw data showed Labour with more declared 2005 voters than those saying they would vote for the party at the next general election. Things changed when he arrived at Number 10.

Now before you risk money based on this data be very careful. You should know that I am not confident enough to re-enter the Commons seat spread markets although my position on the election timing remains.

Anyway we should not have to wait too long before we see some real polls. Within the next eight days there should be at least four.

Mike Smithson