Don't be fooled: Labour may be ahead in opinion polls, but this election is not over

Many Labour voters are trying to decide which they hate less: Tony or the Tories

In a British general election, around 50,000 people work flat out to mobilise 45 million voters; just over 0.1 per cent of the electorate trying to influence the other 99.9 per cent. There is a further factor which increases the disparity; much of the activists' effort is wasted. A large percentage of all political propaganda leads to glazed eyes and closed minds.

In a British general election, around 50,000 people work flat out to mobilise 45 million voters; just over 0.1 per cent of the electorate trying to influence the other 99.9 per cent. There is a further factor which increases the disparity; much of the activists' effort is wasted. A large percentage of all political propaganda leads to glazed eyes and closed minds.

All the parties try to refine their techniques, without being certain that they are succeeding. Don Rumsfeld is right. "There are a lot of unknowns out there.''

Not if you believe the latest opinion polls, however. They may be all over the place, showing everything from a 10-point Labour lead to a single-point margin, but given the current electoral boundaries, even the lower figure would comfortably re-elect Tony Blair.

Yet there are grounds for caution. The first is the opinion polls' recent record. Over the past few years, they have consistently understated the Conservative vote. It may be that they too have now refined their techniques. It may also be that Conservatives are no longer so embarrassed at admitting their allegiance. But until we see the pollsters getting the Tory figure right at an election, a measure of scepticism is justified.

There is also the mood on the ground. Here I have to be wary. So far, I have only been in the Thames Valley, and there are likely to be considerable regional variations in this election. But there do seem to be a lot of undecideds, and genuinely so.

In previous elections, when voters told canvassers that they had still not made up their minds, there was often a subtext: "When I do, it won't be in your favour.'' This time there is much less of that. There is a demand for information, especially among younger voters. I have come across a number of people who said that they had not previously taken much interest in politics. They would now, and they intended to study the parties' programmes in detail. It was almost as if they were buying a car or a mobile phone and were comparing the small print to make sure that they found the best deal.

I have also encountered much less strident apathy than in 2001. I would be surprised if Oxfordshire and Berkshire have a further fall in turnout.

Both the main parties had problems with their own tribal supporters. The Labour one is easy to summarise: Tony Blair. As the PM himself recognises, he has lost his electoral appeal.

That ought to be worse news for Labour than the polls are reflecting. When the firm's principal asset turns into its principal liability, it is usually time to call in the receivers. The polls would suggest that Gordon Brown has made good the deficiency, but how deep is this pro-Brown sentiment? Do the voters really warm to Gordon, or do they merely prefer him to Tony?

Whatever the answer, the Labour Party can hardly hide its leader for the next two and a half weeks, and every time he appears on television he irritates a lot of viewers. Many Labour voters are now trying to decide which they hate less: Tony or the Tories.

The Tories also have problems with their leader. Not many people are crossing the road to tell Conservative candidates that they adore Michael Howard. But the party has a more basic tribal difficulty, with voters who think that the country is so far gone to the dogs that there is no longer any point in voting.

A lot of these are older people on whom the Tories could once have depended. Now they might well be disillusioned enough to abstain. Although they like the party's messages on crime and immigration, they think that they have heard it all before, and nothing ever seems to happen.

Apropos of crime, one would have thought that there ought to be a number of high-minded, fastidious Tories who are unhappy about the emphasis on law and order. Yet they are surprisingly hard to find, even where the sweet Thames runs softly through the gentle and affluent meadows and manor houses of south Oxfordshire. Far from objecting to the emphasis on crime, comfortably off Tories are much more likely to applaud it, because they feel threatened; a view which unites all social classes. It is almost impossible to discover anyone who is happy with the way that the police and the courts currently operate.

The Tories' basic problem is the one identified by all political parties when they feel that they are not doing as well as they deserve: how to get their message across. There does not seem to be anything wrong with the message itself. Cleaner hospitals, discipline in schools, more police, secure frontiers, controlled immigration; they are all pledges which lend themselves to a common-sense simplicity of exposition.

I have also heard a number of Tory candidates impress voters on the doorstep by drawing attention to the small number of promises which the party is making. "None of that is rocket science,'' the candidates will say. "None of it is impossible to deliver, and deliver it we will.'' That seems to work.

There is a final reason for believing that this election is still interesting. Time behaves in a curious way during general elections. In the pre-election period, it appears to foreshorten; once campaigning starts, however, it seems to stretch out. In January it seemed as if the election was imminent. Now, with 17 days until the polls open, a lot could still happen. People will be watching television, reading literature, meeting candidates; only then will they decide. I suspect there will be more late decisions this time than ever before in a British election. Late decisions could produce late swings.

So could the final opinion polls, and the resulting headlines. The late Robert McKenzie, who invented the swingometer, always insisted that the newspaper headlines had influenced the outcome of both the 1970 and the February 1974 elections. On polling day in 1970, the papers were making a firm prediction: "Wilson sweeps back. Five more years of Harold''. In February 1974, they said similar things about Ted Heath. In both cases, a lot of voters were put off. "Sweeping back - don't fancy that much. Limping back chastened might be more like it. If he's sweeping back, he doesn't need my vote."

If the final polls show Tony Blair sweeping back this time, there could be a similar effect. In 1997, many voters would have been happy at the thought of a Blair win by a huge majority. To a lesser extent, that would also have been true last time. Now, the mood has changed. If the polls show another large Blair majority, that could drive the Tories to the polling booths in order to prevent it. In 1992, the Tories won 14 million votes, the highest figure for any party in a general election. Many of them turned out because the opinion polls were claiming that Neil Kinnock was about to become Prime Minister.

A lot is still in doubt, and two and a half weeks really is a long time in British politics. This election will not be over until the undecideds decide.