Screen shot 2015-04-10 at 07.30.49

Lord Ashcroft’s latest batch of polling in Lab/Con marginals included an interesting note on how the campaigns are being carried out:

“Most people said they had had literature, direct mail, phone calls or visits from both main parties, but Labour have the edge in these local campaigns on the basis of this evidence. More than 70 per cent had heard from Labour in half the seats polled.”

This stuck out, because in the previous polls in these seats the Conservatives had mostly had the edge. Here’s the percentage of people in each seat who, in the most recent available 2014 poll, reported having been contacted by Labour or the Conservatives “in the last few weeks”:

2014 Contact Rates

It was a good start – the Conservative ground campaign had reached more people than Labour in nine of the ten seats.

Yesterday’s battleground polls found a very different picture, though:

2015 Contact Data

Over recent months, the lead has reversed – more people reported being contacted by Labour than the Conservatives in all ten of the constituencies.

Finally, here’s a table of the change from 2014 to 2015 in contact rates for each party in each seat:

Comparison Data

That makes for sobering reading. As you would expect, both sides have stepped up their campaigning as the election approaches. Bluntly, in these marginal seats our step-up simply hasn’t been as big as Labour’s. Despite some Tory campaigns doubling or even tripling their contact rates, Labour have made some huge strides to overtake us.

There’s a lot of variation between the ten seats, but the average increase in the share of voters contacted by Tory campaigns is 35 percentage points, while the same figure for Labour campaigns is almost 55 percentage points.

It’s a turnaround from last year – in the 2014 polls the Conservatives led in the average contact rate by over 8 percentage points. These are Conservative-held seats, so the incumbents had evidently been out and about early. But the differential increase in recent months has overwhelmed even that, leaving the average Labour campaign now 12 percentage points ahead. In short, they’re knocking on more doors, and delivering more leaflets.

Are the numbers reliable?

This type of constituency polling, on such a large scale, is unprecedented. As such, it’s relatively untested, and you can never reliably test something like people’s reports of being contacted or not. There’s always the potential for shyness, tactical dishonesty or plain forgetfulness.

We all know that you can push a leaflet through someone’s door regularly, or even talk to them face to face, only to be told next time you come round that they never hear anything from the political parties. It’s human nature – with busy lives and a good dose of cynicism about politics, people merrily ignore or forget the contact they do have (raising an interesting question about how much leaflets matter, but that’s for another day).

Even if people are forgetful, though, there’s no obvious reason why that might apply differently to different parties. So if the figures are distorted, the likelihood is that distortion affects both Labour and the Conservatives equally – which means the figures are viable to compare with each other at least.

The question of whether, having been contacted, people are actually persuaded by the message is a totally separate matter. We aren’t measuring the effectiveness of Tory messaging here, we’re looking at the performance of the campaign machine on the ground in reaching voters.

One final consideration is the question Lord Ashcroft asks. It’s quite specific – this isn’t a measure of whether you’ve been contacted ever, it’s a question of “whether any of the main political parties have contacted you over the last few weeks”. As such, while the poll shows a burst of Labour activity recently, it doesn’t reflect the potential impact of longer term campaigning – something which we know the Conservative incumbents were ahead on, given their lead in 2014. Again, we’re interested in these figures as a snapshot of the performance of the campaign machine right now, not in using them to predict election outcomes.

Those caveats aside, there’s clearly something going on. So what’s happening? In reality, I suspect these trends are symptoms of a complex mixture of factors on both sides of the political divide.


Labour have a larger party membership, which obviously makes it easier to knock on more doors and deliver more material. Like ours, it’s fallen over the years but it remains sizeable. Within that, of course, we don’t know the activism rates, but we know they have a bigger pool of people to draw on.

Backing that up is the added power of the unions. Unite’s leaked political strategy is explicit about providing teams of activists for their favoured candidates, while I’m told that in some areas the unions have funded hi-tech canvassing equipment to allow canvassers to file responses instantly and even to register people for postal votes on the spot. Labour sources also suggest that the Government’s rhetoric on the unions has spurred them into greater levels of activity than they displayed in 2010 - though to an extent they would say that.

There’s also the shape of the Labour campaign to bear in mind. In recent years they have focused very heavily on registering their supporters as postal voters. Converting that effort into votes requires a lot of activity in precisely this stage of the campaign – might Labour have effectively front-loaded their campaigning as a result? We’re also told that Labour’s election addresses have gone out earlier than those of Conservative candidates, presumably to ensure they catch postal voters in time – that’s bound to have had an added effect on the contact rates.

The Conservatives

By contrast, the Conservative Party faces some challenges on the ground. While the decline in membership has at last been reversed, we are still fewer in number and more advanced in years than Labour. In some parts of the country, long-serving activists have also decamped to UKIP, further weakening the machine. Many associations have struggled to replace long-serving stalwarts as they age, die or leave to follow Nigel Farage. As I argued after the European elections last year, that’s bound to show in the ground war.

There have been efforts to plug the gaps. Team 2015, and its partner Road Trip 2015, are designed to focus our activists in the target and marginal seats where they are most needed. The former aims to do so on a regular basis, diverting people from their home constituency to a nearby 40/40 battleground, while the latter descends en masse for a day on one seat or another, delivering huge amounts of literature as they do so. Both are good ideas, but neither is a substitute for a true grassroots movement. Focused into a by-election, like Newark, they can make all the difference, but with 80 battlegrounds to fight these initiatives can’t cover them all, all of the time.

Thirdly, there’s the question of campaign technology. VoteSource has arrived, Merlin is mercifully on its way out, and I gather that after its latest update many of the new system’s early teething problems have been solved. However, it’s far from ideal to receive and have to adapt to a whole new platform so shortly before polling day. Grant Shapps would, I’m sure, have introduced it earlier had he been appointed Party Chairman earlier, and previous Party Chairmen should have acted on it years ago, but the practical impact remains the same regardless of intentions. Most associations still only have a small number of people trained in using VoteSource, while recent weeks have been spent battling with and then ironing out problems (either of the system or of its users’ abilities or both).

Fourth, some Conservative candidates also argue that overall contact figures only provide part of the picture. While some candidates take the approach of campaigning to the whole electorate, others are using better data to identify and intensively target crucial groups of voters - be that their core supporters, undecideds, supporters of other parties whom they hope to squeeze and so on. That would be good campaigning, but it wouldn’t reflect well in this poll question – fewer voters would report being contacted, but they would be the right voters to contact and they would be contacted a lot. By the same token, some Tory campaigners report Labour blanket-leafleting seats with generic national literature, presumably to avoid tipping over the spending cap on local campaigns – a practice which would produce artificially high contact rates without necessarily equating to an effective campaign.

This question of targeting may explain some of the contact rate deficit in some seats, but it doesn’t explain the dramatic reversal of contact figures since last year. Nor does it stop these figures being somewhat worrying. If we had all the people we might want, and the perfect campaign setup, it would be possible to contact target voters and others – that evidently isn’t happening.

One thing is indisputable: as this site has long argued, while speeches, policies and strategies at the top are valuable they cannot stand alone. A successful Conservative Party will always require a mass membership, grassroots movement to fight and win elections. Some progress has been made in reversing the decline in membership, and new tactics offer ways to make up for numerical weakness, but we are still a long way from the machine we need.

132 comments for: Is there a problem with the Tory ground war in the marginals? And if so, how big is it?

Leave a Reply

You must be logged in to post a comment.