When asked to place politicians on the political spectrum in May 2005, just 21 per cent of voters put Michael Howard at or near the political centre, and the Tories remained stuck with fewer than 200 MPs. This year, 42 per cent thought David Cameron lived close to the centre, and he ended up as prime minister.

The lesson seems obvious: to win elections, party leaders need to be widely seen as centrists. If most voters place you clearly on the left or right, you are doomed. After all, only 12 per cent of voters at this year’s election described themselves as ‘very’ or ‘fairly’ leftwing, while 11 per cent regarded themselves as ‘very’ or ‘fairly’ rightwing. Fully 54 per cent said they were in the centre, or ‘slightly’ left- or right-of-centre.

Readers will not be surprised by what comes next: a big, juicy, combative ‘but’. The centre is unquestionably a seductive location, but it is also a treacherous and deceptive one. As with the mirage of an oasis in the desert, the journey towards it can end in grave disappointment. To reach the cool, refreshing waters of electoral success, we must separate illusion from reality.

Let’s start with those voters who place themselves bang in the centre. They comprise one voter in four: any party that can win them over is well on the way to victory. YouGov questioned more than 32,000 electors at the time of last year’s Euro-elections. Eight thousand of these described themselves as ‘centre’ rather than right or left. However, when we examine their attitudes we find nothing that can be remotely described as a single view of the world. They are all over the place – as are those who place themselves on the right or left.

Forced at gunpoint, I would describe the average centre-voter as someone who is financially insecure, thinks most politicians are personally corrupt and regards the three main parties as essentially similar. They would like to nationalise the banks, stop all new immigration and retain Britain’s nuclear deterrent. That said, there are plenty of centre-voters who are not like that – and also plenty of left- and right-of-centre voters who do fit that description. In short, a politician who looks to the electorate to define a distinctive centre ground will look in vain.

What, then, about the electoral record? Surely the last three elections prove the importance of fighting from the centre, for when the Tories were considered to have rightwing leaders, in 2001 and 2005, they lost heavily, whereas when they paraded their centrist outlook this year, they won? Indeed, the same argument could be applied to the previous 20 years. Labour’s fortunes in the 1980s and 1990s correlated with their perceived ideological location – from leftwing disaster in 1983 to centrist landslide in 1997.

Again, an impertinent ‘but’ leaps up to shatter the thesis. The three previous postwar landmark elections tell a different story. In both 1945 and 1979 the victor was plainly radical. Clement Attlee defeated Winston Churchill with a manifesto that promised a different kind of Britain from that of the pre-war era, while Margaret Thatcher sought unashamedly to reduce the role of the state. If anti-centrist politics were doomed always to fail, Attlee and Thatcher should have lost. The 1964 election provides a more modest example of the same point. Harold Wilson’s election-winning manifesto was more radical than Hugh Gaitskell’s election-losing manifesto five years earlier, even though Wilson’s (in many ways rather good) government was not, and was never going to be, as transformative as those of Attlee or Thatcher.

So, can we draw no general conclusions from the five landmark elections of the past 65 years? I believe that is going too far, for there is a common thread that runs through them. It is simply that ‘centrist’ is the wrong concept to describe the five winners.

A more accurate term is ‘in tune’. Attlee, Wilson, Thatcher, Tony Blair and Cameron were all in tune with the swing voters of their time. At the end of the second world war, the last thing voters wanted was a return to the economic horrors and social divisions of pre-war Britain. They demanded a new approach to housing, health, schools and welfare. Labour alone offered this. The Conservatives had to accept the post-war settlement in order to become electable again.

In 1964, Wilson’s appeal was rooted in his clever contrast between ‘modern’ Labour – the party of science and sixties popular culture – and the ‘old-fashioned’ Conservative party, which had allowed a patrician earl to ‘emerge’ by some mysterious process as their leader just a year earlier.

Move on another 15 years and we come to the greatest insurgent victory of all. In the summer of 1978, twelve months before her victory, Thatcher looked as if she would fail. The far more popular James Callaghan had steered Britain through two horrible years of economic turmoil. The country appeared to be on the way to recovery. Then came the winter of discontent – the public sector strikes that left petrol in short supply, rubbish uncollected and (in just a few, but lethally iconic, cases) bodies unburied. Thatcher’s radicalism caught the mood of the time. As in 1945, the status quo was utterly discredited.

And so to the two ‘centrist’ victories of 1997 and 2010. The real point is not that these were triumphs over extremism: few people regarded John Major or Gordon Brown as raving ideologues. Rather, the two defeated governments were seen as tired and sleazy administrations that had screwed up the economy. Blair, 13 years ago, and Cameron, this year, were seen as fresh, clean and more capable than their rivals or, to return to the leitmotif of all five landmark elections, more ‘in tune’ with Britain’s swing voters.

That still leaves us to explain the evolution of the Tory party from extremism and defeat in 2001 and 2005 to centrism and victory in 2010, and the similar journey of the Labour party between 1983 and 1997. I would argue that the real failing of the Tory and Labour parties in their nightmare years was that voters regarded them as hopelessly incompetent and out of touch. ‘Extremism’, with its negative connotations, was a useful label, but it stuck because Michael Foot, William Hague and Howard were already perceived to inhabit a different political planet from most swing voters.

The ‘extremist’ tag doesn’t always work. In a notorious radio broadcast during the 1945 election, Churchill accused Attlee of wanting to introduce a Gestapo-style tyranny to Britain. He failed utterly, because Attlee’s radicalism matched that of the electorate. Attlee was no centrist, but he was in tune with the public mood. Which is why he, like Thatcher a generation later, changed the face of Britain, and also moved the centre ground onto new territory.

The lesson for Ed Miliband and his new shadow cabinet is clear. Do not passively accept the current definition of ‘centrism’ as the lodestar for devising new policies. If you succumb to that temptation, you risk being condemned as weak and indecisive. Instead, give swing voters solid reasons to hope their lives will get better if you win. Then they will vote for you. You need to design a strategy and create an image that will persuade voters you are competent, clear-sighted and in tune with today’s Britain. Sometimes these virtues will be found on what voters regard as the centre ground; but not reliably enough for you to plug that into your political SatNav as your sole destination.

Cartoon: Adrian Teal