Four new polls came out last night, published today:

YouGov in the Sun … CON 39%(nc), LAB 31%(-2), LIB DEM 20%(nc)
Populus in the Times … CON 36%(-3), LAB 33%(+1), LIB DEM 21%(nc)
ComRes for the Independent/ITV … CON 36%(-1), LAB 31%(+1), LIB DEM 19%(-1)
Angus Reid for Political Betting … CON 38%(+1), LAB 28%(+2), LIB DEM 22%(nc)

What does all this mean for Anthony Wells’ UK Polling Report ‘poll of polls’ average? Well, there’s an ever-so-slight narrowing of the gap between Labour and the Tories – or (looked at another way) a slight lengthening of Labour’s lead over the Lib Dems:

Con 38% (n/c), Lab 31% (+1%), Lib Dem 20% (n/c)

I thought I’d take a quick look at how the Lib Dems’ current poll ratings compare with previous general elections – what follows are the averaged figures for the polls conducted during the first week of the election campaign preceding the last four elections:

1992: 16%
1997: 12%
2001: 13%
2005: 21%

What’s clear is the party is performing much better at the equivalent stage in the general election campaign than was the case in 1992, ’97 or 2001; and on a par with how we were doing in 2005 (even minus the ‘Iraq factor’).

A few people have already expressed the worry, “Where’s our general election poll bounce?” – referring to the fact that the Lib Dems nearly always increase our poll ratings during the course of an election campaign owing to greater media exposure. It’s true this happens, but not normally until the last 10 days or fortnight of the campaign.

Of course what we don’t know this time is (i) whether pollsters’ methods have improved, so that our ratings are more accurate all year round, rather than only at election time, or (ii) whether the televised leaders’ debates and the perceived closeness of the contest are going to upset the usual calculations.