So was that the Labour recovery that was?

First of all – hands up – I got it wrong. My reading of the extraordinary events of the past fortnight was that this would give Labour a significant boost and that the crisis was a potential game-changer that could alter our view of the next election. This is how I’ve been betting.

We’ve all seen the incredible transformation that has come over Brown – the international acclaim that has given him an extra spring in his step, the newly recovered confidence and a smile that looks genuine if a little inappropriate for the times. All this has given heart to Labour MPs in marginal seats who had become reconciled to losing their jobs and salaries at the election.

Alas overnight we’ve had three new polls which must have gone down like a lead balloon in Downing Street. Surely Labour and Gordon could have expected better?

BPIX – the non-British Polling Council firm that uses YouGov for its fieldwork and never reveals its detailed data has this in the Mail on Sunday. The comparisons are with the pollster’s last survey just after the Labour conference – CON 46% (+3): LAB 30% (-1): LD 13% (-4). The fieldwork continued into Saturday.

ComRes for the Independent on Sunday has with comparisons on the firm’s last poll also straight after the Labour conference – CON 40%(-1): LAB 31%(+2): LD16% (-2). This was carried out on Wednesday and Thursday before David Cameron got back into the headlines with his big attack on Brown

ICM for the News of the World which might have had a voting intention question but if it did nobody is reporting the numbers. This had 13% saying they were more likely to vote Labour as a result of Brown’s handling of the crisis compared with 22% saying they were less likely.

All three polls have positive figures for Brown on his handling of the crisis: ICM had it by 54% to 36% and there were similar numbers from BPIX.

I’m far from convinced of the importance of non-voting intention findings. If they meant anything in electoral terms they would show up in how people say they would vote. They don’t and that is Gordon’s problem this morning.

We should get a clearer picture from other polls that are in the pipe-line. YouGov has a surveys of marginals out on Wednesday while fieldwork for Ipsos-MORI’s October monitor is taking place at the moment and we should have those numbers mid-week as well.

Before then I hope to look at the detail of the surveys – particularly the ICM poll which is reported in such an odd manner.

In the Glenrothes by election betting I expect that SNP will become an even tighter favourite.

Mike Smithson