Craig Paterson poll analysis: Could the undecided and tactical voters really tip the balance?

CommonSpace poll analyst Craig Paterson explores how important the undecided and tactical voters will be in the #GE2015 outcome

THERE has been a lot of talk recently about the 'tactical vote' up here in Scotland. Indeed, some have even called this the tactical election. We Scots are used to voting tactically so it's not really anything new, but it's the potential sheer scale of tactical voting in this election that makes things so interesting.

First, a quick round up of where things stand in the polls at the moment. The last five polls out (including the Sun/YouGov poll for today) have been all over the place. Two leads for Labour, two leads for the Conservatives and one tie.

The only thing to say is there has been slight (very slight) movement towards the Tories overall.

The average of last five polls (five different pollsters) is as follows:

CON: 34 per cent
LAB: 33.6 per cent
LIB DEM: 8 per cent
UKIP: 14.6 per cent
GREEN: 4.2 per cent
OTHER: 5.6 per cent

Not much in it at all really. Over the next few weeks expect to see a lot more talk about seat projections and what one coalition looks like compared to another, the media love to guess.

It's worth at this point a little trip down memory lane back to the 2010 eve of election polls. Out of the seven polls conducted on the eve of the election none of them accurately predicted the outcome; granted, some came close but in an election this hotly contested a percentage point here or there could swing multiple seats.

In 2010, out of the seven pollsters' predictions on the three main parties, two accurately predicted the correct percentage of Conservatives, five underestimated them. All seven underestimated Labour and all seven overestimated the Lib-Dems, Cleggmaina at play here.

Caution is the watch word I would put out there. Everyone was talking about massive gains for the Lib-Dems in 2010 - some even predicting second biggest party - yet in the end their vote share was only one per cent up from 2005, and they lost seats.

Out of the seven polls conducted on the eve of the 2010 election none of them accurately predicted the outcome.

The BBC's editor of political research, David Cowling, had this to say after the 2010 result: "Were the polls wrong throughout the campaign? I doubt it, but then how do I explain the stark contrast between them and the hard reality of 6 May? At the moment I cannot.

"For me and for many others, I imagine, there is considerable work in progress as we try to reconcile the past four weeks of campaigning with the actual outcome."

This being said, there is no denying the sheer scale of the SNP lead and whatever happens they are set to make huge gains: 15, 20, 30 seats, maybe more. We won't know until early on Friday 8 May.

This brings us on nicely to a couple of groups that may tip the balance in some seats across Scotland, maybe even in the rest of the UK, too: tactical and undecided Voters.

Today the Scottish Daily Mail is running a spread on tactical voting, but can it really make a difference?

First, the undecided voters: the Ashcroft polls during the week give us a bit of an insight into this. I won't go into detail on every one but in the eight seats polled, the voters who either refused or were undecided ranged from between 12 per cent in Berwickshire, Roxburgh & Selkirk, to 21 per cent in East Renfrewshire. The average of these eight seats is 16.25 per cent.

Let's take a look at East Renfrewshire as an example. When asked "Who would you vote for?" the results were:

SNP: 31 per cent
LAB: 20 per cent
CON: 20 per cent
LIB DEM: 2 per cent
GREEN: 1 per cent
UKIP: 1 per cent
UNDECIDED/REFUSED: 21 per cent

As we can see from this there is a bigger constituency of unknown voters than there are Labour or Tory voters. These voters will have huge part to play in how this seat pans out.

What Ashcroft then does with a proportion of these undecided and refusals is reallocate them and put them into the party they voted for in 2010 to help extrapolate his final results. Therefore, before this is done it means that those still not committed or those refusing to say could well tip the balance and stop the SNP from winning some of these seats.

SNP vote share vs second place party vote share vs undecided and refused (UD+R) vote share is as follows:

Berwickshire, Roxburgh & Selkirk
SNP: 25 per cent
CON: 22 per cent
UD+R: 13 per cent

Dumfries, Clydesdale & Tweeddale
SNP: 30 per cent
CON: 24 per cent
UD+R: 13 per cent

East Dumbartonshire
SNP: 35 per cent
LIB DEM: 17 per cent
UD+R: 14 per cent

Glasgow South West
SNP: 41 per cent
LAB: 24 per cent
UD+R: 20 per cent

North East Fife
SNP: 32 per cent
LIB DEM: 17 per cent
UD+R: 17 per cent

Paisley and Renfrewshire South
SNP: 35 per cent
LAB: 21 per cent
UD+R: 19 per cent

Ross, Skye, Lochaber
SNP: 40 per cent
LIB DEM: 20 per cent
UD+R: 13 per cent

For me we should take these voters at face value. I would hazard a guess here and say the vast majority of people who are going to vote SNP have made their minds up already, so I do expect a lot of these undecided voters to break for other parties.

The question is, will they vote tactically? There is a whole lot more of the tactical vote than people think. Is it going to play a huge part in this election? When I say there is a whole lot more of it than people think, I say it because everyone is overlooking the number of people who will be voting SNP in this election who would not normally; a sizable chunk of them are doing so as a tactical vote to get rid of Labour.

Yes, undoubtedly, the SNP has increased its base vote but they have not secured the loyalty of all of their voters. People are hoping if given the chance they don't go all Lib-Dem on them.

There is a whole lot more of the tactical vote than people think.

Will it play a huge part? The Ashcroft polls tell us that between 76 per cent and 90 per cent of Tory voters rule out voting SNP (no real surprise), between 56 per cent and 74 per cent of Labour voters rule out SNP, and 73 per cent to 85 per cent of Lib Dems rule out voting for the SNP.

On the flip side, 17 per cent to 31.7 per cent of voters would consider voting SNP (not including, of course, all the ones who have switched already). In the Labour-held seats, between 26 per cent and 36 per cent of Tory voters would vote Labour.

In the Lib Dem/Conservative seats, 25 per cent to 46 per cent of Labour voters would vote Lib Dem, while 39 per cent to 55 per cent of Lib-Dems would vote Tory. Potentially unseating their own party candidate in favour of the conservatives - how far the Lib-Dems have fallen!

It would appear from the Ashcroft numbers that tactical voting could well make the difference in these seats. Tactical voting is hard to predict, or control, but make no mistake, the right wing press and the parties themselves are attempting to make the choice a little clearer.

The unknown factor is whether or not it will be enough to affect the outcome. This is clearly more difficult to answer, but all three parties are singing the same tune: "Vote for us to stop the SNP."

You think they would all just stop the pretence and become one party. Still, all to play for with 17 days to go.

Picture courtesy of Keith Ivey