This week’s dash to Scotland by Westminster’s major parties has refocused attention on the men who lead them.
Earlier this week we took a look at how David Cameron is far more personally popular than Ed Miliband. When asked who would make the better prime minister, 35 per cent of voters say Cameron and 22 per cent choose Miliband. 
This is one of two great advantages the Tories appear to have ahead of the election, along with their lead on managing the economy.
It is also something Lord Ashcroft, the multimillionaire former Tory deputy chairman-turned-pollster, is fond of highlighting in his polls of the most closely fought election seats.
But Cameron’s lead may be less secure that it appears. While he has always been more popular than Miliband, scoops and speeches appear to have had a big impact on that popularity.
For instance, in 2012, his lead halved over one fortnight in late March, exactly when the Sunday Times broke its “cash for access” story.
Nor did the scandal just hurt Cameron. Our data on how voting intentions have changed since 1970 shows it carried over into how people felt about the parties. The Tories went from polling 40 per cent in early February 2012 to 32 per cent three months later.
The scoop showed how volatile our perceptions can be. Similarly, in October 2012 Miliband managed to reduce Cameron’s lead to just 4 per cent. The cause is hard to identify, but his polling peak did coincide with his most-favourably received conference speech, on the need for “One Nation” Labour.
Might the same thing happen again after this month’s conferences? And could Miliband become the more popular leader by May?
It is unlikely, but less so than it might appear. If we look further back, at how our perceptions of the leaders changed in the two years before the last election, the same story appears: major events can upend the way we view candidates.
The financial crisis may have wrecked the British economy for half a decade, but it also temporarily transformed Gordon Brown’s personal popularity. In September 2008 just 20 per cent of voters thought he was doing well as Prime Minister, while 73 per cent thought he wasn’t – a deficit of 53 per cent. Two months and a banking collapse later, that gap had closed to 12 per cent.
The sudden changes continued for six months. The gap widened and then narrowed and then finally settled after widening again. At the same time, Cameron’s popularity mirrored the stock market, diving and recovering as the UK was hit by recession.
Even more famously, Nick Clegg went from being fairly anonymous to fanatically popular thanks to the TV debates.
But by the end of the year, after taking his party into government, Clegg’s ratings had fallen from +63 to -30. His party had polled higher than Labour on the eve of the election. By the end of the year they were languishing on 10 per cent.
We can pick-up similarly drastic changes in specific traits, like how “decisive” voters think a leader is.
Gordon Brown suffered in late 2007 when he was seen to dither over whether to call an early general election. 37 per cent of voters had called him decisive when asked in October of that year – more than twice as many as Cameron – but just 14 per cent did six months later.
This shows how impactful events can be on this seemingly steady statistic. Cameron’s personal popularity is, and always has been, an advantage for the Tories, but the upcoming conferences and campaign could well change that.
If they do, the Tories could lose one of the few advantages they have left in a battle where few things favour them. They can ill-afford to see the British public take a grudging liking to Ed Miliband – they already have to win far more votes than Labour do to win a majority.
 Averaged figures for 2014 (January – August).