When activists stagger home the night before polling day, it’s not just their sore feet and the thought that they’ll be up again to deliver more leaflets before dawn that makes them wish it was all over. By this point in a campaign most voters have already decided, so the desire to know the outcome is almost unbearable. It’s easy, then, to read opinion polls for a bit of ‘news’. Of course, when it comes to local elections you might as well read the tea leaves as national polls, but they’re still irresistible, aren’t they? And, this month, vaguely hopeful. I’ve written before that opinion polls showing a rise in the Lib Dem vote should be taken with as big a pinch of salt as those showing us in decline. They’re a very blunt instrument, and very few show a clear long-term trend instead of an hysterical margin-of-error blip – though the 24-hour news cycle’s desire for instant gratification (and the blogger’s, to be fair) means you rarely sit back and say, ‘Well, I’ll wait to see if that’s a trend over the next few months…’ So, with that health warning given, time to pretend they matter and pore over the entrails, but just remember that the time to celebrate or weep will only be Friday morning. In the meantime, good luck to Liberal Democrats working in local elections throughout most of the UK, and in the Scottish Parliamentary and Welsh Assembly campaigns.

Today’s most exciting polls are undoubtedly those suggesting the Scottish Parliamentary elections might put the SNP and Labour in a dead heat, while support for independence slips still further. Though both predict gains for the Liberal Democrats, they also suggest the party’s vote has been squeezed over the month, which must give Scottish Lib Dems mixed feelings. The most remarkable figure for me is the Guardian / Scotsman’s ICM poll on the gender difference in the Lib Dem vote – the most striking I’ve ever seen for any party – which suggests we have the support of 20 per cent of women in Scotland, but only 10 per cent of men. I wonder if the Scottish Liberal Democrat Party Election Broadcasts heavily promoted some of the party’s excellent women spokespeople, unlike the all-male ones we saw down in England? Without detailed breakdowns of Scottish voting intentions to hand, however, I’ll point you to Freethink for the figures, and take a look at changes in the party’s UK-wide support.

What did April’s Britain-wide opinion polls purport to show? Taking the mean average figures, it looks like both Labour and the Conservatives dropped a percentage point or so, with the proceeds split between the Liberal Democrats and the combined ‘Others’ (from 6.6% in March to 7.8%, though those figures measure so many different parties I don’t trust any samples of them, up or down).

In March, the Conservatives’ average showing across eight UK-wide polls was 38.9%, ranging between 35% and 41%. It may not sound that impressive, and it’s true that Labour was, incredibly, regularly in the 50s before they took power from the Conservatives in 1997 (with a significantly lower actual vote of 44%). However, March’s poll average was still the second-best the Tories had had since 1992. In April, an average of polls had them slipping back to 37.3%, ranged more narrowly between 36% and 39% – still far better than they’d been doing between 1992 and 2005, but no longer strikingly ahead of the average for the last year or so. So, possibly a loss of momentum, but qualified good news for the Tories.

The polls suggest unqualified bad news for the Labour Party at the moment. Their monthly average has dropped from an already low 31.1% to 30.2%, barely into the psychologically important 30s (however much opinion polls shouldn’t be trusted, it’s hard to resist the leap in the heart when Labour and Lib Dem figures begin with the same numeral). This is, without a single exception, Labour’s worst monthly polling performance for twenty years. The averages conceal a variation of between 29% and 33% in March, declining to between 27% and 32% in April. If I were a Labour supporter, I would be praying fervently that they will gain from a collective sigh of relief from the British people when they finally wave goodbye to the man whose principal legacy for 69% of people, according to yesterday’s poll in the Independent, is the bloody chaos in Iraq.

That leaves the Liberal Democrats. For a bit of perspective, since the party was formed the opinion polls have never showed us in poll position to win a general election, though a whole one month’s average since 1988 has put us in second place (that was August 1993, when a score of 27.1% was one of the three months in which we’ve struck 25%). At the other end of the pile, in July 1989 we rated just 5.4%, from which it took us nearly a year to reach again the dizzy heights of, er, 7%. The party is currently towards the upper end of that historic range, though not as well as during about three brief, hopeful years in which we tended to average over 20%. In the eight polls this March, we averaged 18.3%; in the six in April, it was up to 19.5%.

Splitting each month in half (taking the polls in a 4-4-3-3 pattern) suggests a gentle but sustained Lib Dem rise from 18.0%, to 18.5%, to 19.3%, to 19.7%. I may be an old curmudgeon about polls, but that at least looks mildly hopeful, even if taking an extra 0.5% from Labour would have put a bigger smile on my face. In both months, YouGov’s 16% dragged the average down, with the range in March otherwise from 17% to 21% and in April from 18% to 22%. Could it be that all those leaflets through doors have an effect? I’d assume that, but looking at previous years’ polls in the month before local (but not general) elections, last year they bounced all over the place, in 2002 and 2004 we seemed to dip slightly through April, and only in 2003 did there seem to be a slight rise during the month, though as part of a longer trend downwards. So, if these polls are in any way meaningful, I have no idea what’s prompted them.

The problem with using these figures as predictive ones (on top of the usual problems with polls), of course, is that they ask for voting intentions in hypothetical general elections. Most people voting tomorrow will be doing so in England, where only local elections are taking place. As a very rough rule, Liberal Democrats seem to do best the smaller the geographical contests being fought (best in locals, middling in generals, underperforming in European elections), so you would expect that the opinion polls will underestimate our actual local election performance. Our notional vote shares for the last four sets of local elections (2002-2006, excluding the general election year of 2005 with its massively different turnout and media coverage) have averaged about 8% higher than our average opinion poll share for April of each year, with estimated shares of real votes varying between 27% and 29% while opinion polls hypothesised between 17.3% and 21.0%. I’ll cross my fingers and suggest that we might get 27.5% across all the local elections tomorrow, with under 27% disappointing and over 28% worthy of celebration. Or, alternatively, you might just cut open a sheep and divine the results from that.

The opinion polls may soon have a new leader to consider – assuming the Labour Party at last succeeds in chopping off Mr Blair’s head and burying him at a crossroads (figuratively, Dr Reid; don’t send the boys round). This time next month, then, I’ll be digging out some opinion polling ‘greatest hits’ to see just what impact Tony Blair and David Cameron appeared to have on the Lib Dem vote, and speculate wildly on what Michael Meacher (only joking!) might do…