The squeeze is on the smaller players

That cloud looks like a tree. Or a bird. Or a flower. It’s human nature to see patterns in chaos and further, to try to rationalise and explain those patterns. Hence earthquakes are still archaically ‘acts of God’. Hence also the interpretation of the four point Labour lead in the YouGov poll published after the Cameron-Miliband interviews and Q&A as a Miliband ‘win’. Given that YouGov routinely publish over 20 polls a month, we should expect at least one outlier and several others a bit out-of-true simply due to statistical noise. It turned out that one of these followed the ‘debate’, so providing a misleading impression.

In fact, over the month, the Tories firmed up their slender lead in the PB poll average, though I’m more confident about the reported lead than I was last month. In February, it rested mainly on one ICM poll which gave the Blues a 4-point lead and itself looked a bit of an outlier. This month, the picture is much more consistent across the five pollsters in the index. The figures are:

Con 34.5 (+0.5)
Lab 34.1 (+0.5)
UKIP 12.2 (+0.1)
LD 7.9 (+0.3)
Grn 5.7 (-1.3)
Oth 5.7 (-0.1)

While the overall picture at the top of the list remains much the same, the big change comes at the bottom. The polling story of the winter was the Green surge; rather ironically, with the green shoots of spring, that’s now faded: Bennett’s party has lost nearly a quarter of their vote since it peaked in January, though still stands substantially higher than it did this time last year, never mind at the last general election.

That drop looks to be part of a bigger picture, with the big two squeezing out the lesser players (in England and Wales at least). In November and December last year, the combined Con+Lab share was about 62.5%; it’s now fully six points higher.

Most of that change has come not from the Greens but from UKIP, who are broadly flat on the month. Late last year they were consistently scoring in the mid-teens, peaking at 16.4% in October. Like the Greens, they too are down by around a quarter since then, which may go a long way to explaining the simultaneous rise in the Conservative vote.

In one sense, David Cameron ought to be reasonably happy with March’s polling figures. Apart from a second consecutive monthly lead, the Conservative share of 34.5% is his party’s highest since March 2012 and within touching distance of its 2010 result. Or it would be if there weren’t a very large fly in the ointment in the shape of Labour’s own share, which is several points higher than Gordon Brown managed on polling day; probably enough to hand Labour a narrow lead in seats were it to be realised on May 7. The collapse of the Lib Dem vote – still just a third of what it was in 2010 and their second-lowest ever in the series in March – remains the largest of the several critical swings in play.

March’s figures are realistically the last meaningful set before the election. Although one can (and will) be compiled for April, the reality is that so much can happen over the course of an election campaign that an average is of less value. Likewise, the purpose of the PB Average – to strip out the noise of short-lived events and statistical fluctuation and identify the true direction of travel in opinion – is of less use when the vote itself is only a few days away (and in the case of postal votes, mostly over and done with). So what can we expect of April?

The main movements now, as across the parliament, are not directly between Con and Lab but between each of them and their respective opponents in their area of the ideological landscape. This will be one of the most multi-faceted elections ever and anticipating its outcome is like trying to predict a game of chess played at sea on a heptagonal board where someone’s greased the pieces beforehand and where several players can move at once: tricky. Nonetheless, the debate this week may well prove critical.

For UKIP and the Greens, the debate is their best chance to make their mark in the campaign and reverse their recent declines. As was noted on PBC recently, the main effect of granting UKIP major party status seems to have been not to give them extra coverage but to give the Lib Dems less (not unreasonably, given that the Lib Dems are likely to behind UKIP in votes and the SNP in seats). If one minor party scores a big hit or makes a serious gaffe, that’s likely to have a knock-on effect on the top-line – as in 2010 when Clegg zapped Cameron’s lead. If not, it’s probable that coverage will continue to focus heavily on Con and Lab (outside Scotland and NI), leading to further falls in the purple-green share.

David Herdson

Harry Hayfield’s by election preview

Whyteleafe on Tandridge (Con defence)

Result of council at last election (2014): Conservatives 34, Liberal Democrats 6, Independent 2 (Conservative majority of 26)

Result of ward at last election (2012): Conservative 457 (48%), Liberal Democrats 376 (40%), United Kingdom Independence Party 115 (12%)

Candidates duly nominated: Martin Ferguson (United Kingdom Independence Party), David Lee (Liberal Democrat), Peter Sweeney (Conservative)

It shows how bad things were for the Conservatives when in 1995, Tandridge in deepest Surrey, popped up as a Liberal Democrat win, however since those days the Liberal Democrats have been knocked not only out of power but for six as well (as that’s how many councillors they have left). In 2003, the earliest data I have, there were ten of them, the following year they reached their peak of 11.

But have been falling ever since. But as you can see from the result at the last election in 2012, it’s not that they have been knocked out completely and with Tandridge being the sort of area that UKIP have the potential to do very well indeed, could we see something that has been discussed in many quarters (and give a chance for the Liberal Democrats to retain a substantial number of MP’s), UKIP taking votes from the Conservatives in sufficent numbers to allow the Liberal Democrats to HOLD or even GAIN seats that might not have gone that way if UKIP were not standing?

Harry Hayfield