Ed Miliband’s rising approval figures in the polls have led to some reassessment in the commentariat. A lot of it is predictably rather coloured by wishful thinking – with some honourable exceptions there are a lot of right-wing commentators who are still convinced that he is a leader who the voters do not see as up to the job and that this is an almost insurmountable obstacle to Labour, and a lot of people on the left who think he is either now the apple of the public’s eye, or that perceptions of the party leaders are at most a side issue, if not entirely irrelevant to how people vote.

At the extremes both are wrong – people who say that it is impossible for Labour to win with Ed Miliband are wrong, the leader is but one factor in voting intention and there are clearly many others. It is perfectly possible for a party to win despite having a duff leader. It would be equally wrong to say that leader perceptions are not a factor at all – we can be relatively certain from key driver analysis of recent British Election studies that perceptions of the party leaders are a major driver of voting intention. It is as much wishful thinking to dismiss the problem for Labour as it is to pretend the problem is insurmountable.

Let’s first try to identify the problem. It is easy to cherry pick good and bad results for party leaders, all politicians have strengths and weaknesses. For example on the up side Ed Miliband is seen as the most in touch of the party leaders, is far more likely than the other party leaders to care about the problems of ordinary people and often leads when people are asked how well the party leaders are currently doing at their jobs. On the downside, he is also seen as weak, not up to the job and people don’t think he looks like a potential Prime Minister. All of this adds colour and understanding to WHY a party leader is seen positively or negatively, but doesn’t get us to the core question of whether they are a positive or negative for their party. Let’s see if we can find some questions where we really can benchmark a leader against their party.

First there is the comparison between leader ratings and voting intention. Labour have a lead of around about 10 points in the polls, and yet David Cameron has a lead of around about 10 points as best Prime Minister. It is perfectly normal for the governing party to do better in Best PM than in voting intention because the sitting PM has the benefit of incumbency (it’s easier for people to see them as Prime Minister), but this is an unusually large gap. Below is the Conservative poll lead (or deficit) since YouGov started regularly tracking both figures in 2003, put alongside the Conservative leader’s lead (or deficit) on the measure of best PM.

You can see IDS lagged significantly behind his party (on average he was doing 16 points worse than his party). Things improved under Michael Howard, who only lagged 7 points behind his party. That shrunk to 5 point when David Cameron took over and once Gordon Brown replaced Blair the Conservative lead in voting intention was almost identical to the Conservative lead as best Prime Minister. Now look at what happens once Ed Miliband takes over as Labour leader. Labour are the opposition now so the lines are reversed as we would expect, but look at how far Miliband lags behind his party. The average gap is 18 points.

Let’s take another measure, Ipsos MORI have a great question on whether people like both the party and the leader, like the party but not its leader, like the leader but not his party, or don’t like either of them. Tracking data for it is here. If you look back to the last Parliament people consistently said they liked the Labour party more than Gordon Brown (in April 2010 43% said they liked Labour, but only 37% liked Brown) – he was a drag on his party. With David Cameron it was the other way round, in April 2010 53% said they liked Cameron, but only 38% liked the Tory party. Cameron was a positive for his party (note that even then more people liked the Labour party than the Conservatives!).

MORI have only asked the question once about Ed Miliband, well over a year ago, but ComRes asked an almost identical question this April. They found that David Cameron’s advantage over the Tory party had vanished, now 37% of people liked the Tories, 38% liked Cameron. Ed Miliband’s figures though looked worse than Brown’s – 45% of people said they liked Labour, but only 22% said they liked Miliband, so he trails his party by 23 points. A majority of people who said they liked Labour said they didn’t like Ed Miliband.

Another straw in the wind, back in May YouGov did a hypothetical poll asking how people would vote if Boris was Tory leader at the next election. As a control, they asked how people would vote at the next election if the party leaders remained David Cameron, Ed Miliband and Nick Clegg. The result was that a normal voting intention figure of CON 31%, LAB 43%, LD 9% became CON 32%, LAB 40%, LD 10% once you mentioned Cameron, Miliband and Clegg as the party leaders.

But even if Ed Miliband is less popular than the party he leads, does it matter? The first thing to note is that these opinions are already there, they are already factored into Labour’s price, and yet Labour are ahead by 10 points. Clearly Labour are perfectly capable of getting the most votes with Ed Miliband as their leader.

However, that doesn’t mean Ed isn’t costing the Labour party votes. Just because Labour have a good lead in the polls, doesn’t mean they couldn’t have a bigger one. Going back to that ComRes poll, 82% of people who say they like Labour AND like Ed Miliband who were asked their voting intention said they’d vote Labour tomorrow, amongst those who like Labour but do NOT like Ed Miliband that figure falls to 61% of people.

While Labour have a ten point lead in the polls this doesn’t matter that much. I doubt Prime Minister Miliband would arrive in 10 Downing Street, throw himself on the bed and cry himself to sleep because no one likes him and he only won by 10 percent points. However, mid-term opposition leads in the opinion polls have a tendency to fade as elections approach and a nine-or-ten point buffer mid-term may be far less comfortable come polling day.

What should be far more worrying to Labour is if things like leader perception becomes more important closer to elections. We all know the pattern of mid-term term blues, of oppositions doing better in the middle of Parliaments in the polls and in mid-term elections. Well, why is that? Part of it is due to the actions of political parties. Governments do unpopular things early in the Parliament and save nicer more populist things for the end of the Parliament when they need the votes. Oppositions are policy lite early in the Parliament and paint themselves as all-things-to-all-men, later on in the Parliament they must come off the fence and disappoint some people.

I suspect part of it though is how people think about voting intention and answer polls – right now I suspect a lot of voting intention is simply disapproval of the government, telling a pollster you’ll vote Labour is the way people indicate their unhappiness with the government. As we get closer to an actual general election it becomes more of a choice of alternative governments – which one would I prefer?

The reason that Ed Miliband lags behind the Labour party in polls is because there are a substantial number of people who say they’ll vote Labour, but on other questions say they aren’t sure who would be the best Prime Minister, or which party they’d trust most on the economy. It suggests that support may be pretty shallow and liable to fade once the general election approaches. Of course, there is plenty of time until the election, time to firm up that support, time for, as the vernacular used to be in the last Parliament, for Labour to “seal the deal”.

In short: is Ed Miliband a drag on Labour? Yes, he probably is. Can Labour win with him as leader despite that? Yes, it is certainly possible. Will he become even more of a drag as the election approaches and minds are focused on choice of government, rather than anti-government protest? The jury is still out.

Tonight’s YouGov poll for the Sun has topline figures of CON 33%, LAB 43%, LD 8%, UKIP 7% – very much a case of “polldrums”.

I’m not sure the tables have gone up as they should yet, but yesterday YouGov also did some questions on perceptions of David Cameron and Ed Miliband showing people a series of paired attributes such as strong or weak, trustworthy or untrustworthy, competent or incompetent, and asking which best applied to each man.

The large majority of ratings were negative, but David Cameron’s best (or least worst) ratings were on being seen as competent (37%), strong (35%), up to the job (35%) and having a clear sense of purpose (35%). His worst ratings by far were on caring about ordinary people (just 22%) and being in touch (25% – with 67% seeing him as out of touch).

Ed Miliband’s ratings were almost the exact opposite, his strongest attributes were being seen to care about ordinary people (46%) and being in touch (35%). However, his main weaknesses were on being strong (22%), decisive (21%) and being up to the job (22% – with 55% thinking he is not up to the job).

David Cameron’s is increasingly being seen as out of touch and uncaring towards ordinary people… but does better on being seen as a strong and competent leader. Ed Miliband is seen as much more in touch and caring, but has still not convinced people he is up to the job.

There are two new polls tonight. First YouGov’s daily poll for the Sun has topline figures of CON 33%, LAB 44%, LDEM 9%, UKIP 8%. Labour’s lead in back in double figures, but well within the normal margin of error of the Labour leads of 9-10 points that YouGov have been showing of late.

Secondly is Populus’s monthly poll for the Times which has topline figures of CON 34%(+1), LAB 40%(-1), LDEM 12%(+3), Others 15%(-2). Changes are from Populus’s last poll a month ago. Populus also asked their tracker questions on party image and questions on the reshuffle that is expected in coming months.

On party image there are some sharp drops for the Conservatives, most noticably in terms of competence and unity. Populus last asked the same questions back in March, around the start of the government’s “omnishambles” period. Back then 46% saw the Conservatives as “competent and capable”, this has now dropped twelve points to 34%, putting them marginally behind Labour. On being united the Conservative drop is even sharper – in March 42% of people saw the Tories as united, that has since dropped fourteen points to 28%.

For the reshuffle, Populus asked if each minister should stay or go. The public’s favourite for the boot was Andrew Lansley, with 53% of people who knew who he was thinking he should go. He was followed by Jeremy Hunt, of whom 48% thought he should go. More surprisingly given the lack of major scandal or hugely unpopular policies, Theresa May came third on 46%, followed by George Osborne on 45%. Much more popular were Ken Clarke (only 36% thought he should go) and William Hague on 20%.