I wrote about this in my last post – exploring what, if anything, we could tell from the polling about whether Boris Johnson would get the blame if Brexit did indeed end up being delayed past the 31st October.

With the government now pushing for an election in December the issue has now arisen again, with lots of people dragging out a ComRes poll from the 16-17th October that asked how people would vote in an election if Britain had NOT left the European Union on 31st October, showing Labour one point ahead. Some people are sharing it with excitement, others with dismay. Both should probably calm themselves.

As a general rule, you can only usefully ask people a polling question if they actually know the answer… and most of us aren’t actually very good at predicting how we will respond to hypothetical situations. If you take this specific question, it was asking people to imagine quite a lot. How had the delay come about? Had the government fought it, or gone along with it? How had the government explained and reacted to the delay? Given the dates of the fieldwork, many respondents wouldn’t even have known about the deal. All of these things will impact how the public react and whether they blame the Conservatives or not… but were impossible for respondents to know.

In short, polls measure current public opinion. They can’t predict the future. While you can ask respondents to predict their own future opinions, they aren’t necessarily very good at it.


There were two polls in the Sunday papers. ComRes had a poll conducted for BrexitExpress (a pro-Brexit pressure group) prominently but poorly reported in the Sunday Telegraph. The voting intention question included The Independent Group as an option, producing topline figures of CON 36%(-2), LAB 34%(-3), LDEM 8%(-2), TIG 8%(+8), UKIP 6%(nc). Most polling companies are not, at present, including the Independent Group in polls – something that will presumably change as they take steps towards actually forming a party and clarifying their future intentions. The tables for the poll are here.

The Sunday Telegraph headlined on a finding that 44% of people agreed with a statement that “If the EU refuses to make any more concessions, the UK should leave without a deal”, suggesting rather more support for no deal than almost all other polls. I would advise a lot of scepticism here – agree/disagree statements are a rather suboptimal approach towards asking polling questions (I’ve written about them before here) that tend to produce a bias in the direction of the statement. The problem is they give only a single side of the argument – the question only asked people if they agreed with a statement supporting leaving with no deal in those circumstances. It did not offer people alternative options like a delay, or accepting the deal, or having a referendum. One can imagine that a poll asking “In the event that the EU does not agree to further changes to the deal, what do you think should happen?” would have produced rather different answers. Indeed, later on the survey asked which outcomes people thought would be best for the UK economy and best for UK democracy, which produced rather more typical results.

Note also that the Sunday Telegraph’s claim that the poll showed an increase in support for No Deal is not accurate – the poll back in January asked a differently worded question (it was structured as support/oppose, rather than an agree/disagree statement, and was in a grid along with other options) so they are not directly comparable.

As well as the ComRes poll there was a BMG poll hidden away in the Independent. The figures were unusually reported without excluding don’t knows or won’t votes, with the Conservatives on 31%, Labour on 27% and 8% for the Liberal Democrats. According to the Independent the Conservative lead is five points once don’t knows are excluded – that implies something along the lines of Con 40%, Lab 35% and Lib Dem 10% – though the full figures are yet to appear on the BMG website.


-->

Ipsos MORI’s monthly political monitor poll for the Standard was published yesterday. Topline voting intentions were CON 39%, LAB 37%, LDEM 10%, UKIP 5%. The two point lead is unchanged from MORI’s previous poll in September, and are very much in line with the other recent voting intention polls since my last update. YouGov and Kantar polls last week both showed 5 point Conservative leads, a Survation poll a one point Tory lead.

While voting intention polls continue to show a small Tory lead, the underlying figures remain poor. People don’t rate the government or the Prime Minister (net satisfaction for the government is minus 48, for Theresa May it’s minus 32), economic optimism is low (61% expect the general economic condition of the country to get worse over the next year) and confidence in May’s ability to get a good Brexit deal continues to trickle downwards, this latest poll has 19% saying they are confident, 78% saying they are not.

Full tabs for the MORI poll are here.


Today’s Observer has, inevitably, news of someone else who wants to set up a New Centrist Party (NCP). Hence, to save some of the phone calls I’ll get tomorrow asking whether the polls suggest a NCP would actually do well, here’s my answer.

First, one should not assume that there is actually a big appetite for a new party in the place some in the commentariat seem to think there is. While most of the public consider themselves to be at or around the political centre, this does not mean there is a consensus about what that centre is, or that those people who consider themselves to be in the “centre” necessarily share their views with what the Westminster/media think the “centre” means. Public opinion tends to the left on economics, and is quite right-wing on more cultural issues like immigration and crime. There may well be a gap for a political party putting forward that combination of views, but it doesn’t seem to be the same gap that most of the proposed centrist parties are seeking to fill, which often seem to be aimed at a far more liberal worldview.

Even assuming there is a gap to be filled, can we tell how well it would do? Could you do a poll asking how many people would support a new party is one of the most depressing questions I get asked. It is one of those things that opinion polling simply can’t do very well. If you were setting up a NCP there is certainly lots of useful things market research could tell you about which demographic groups are most open to considering it, which messages would chime, what obstacles it would face – but in terms of predicting how well it would actually do? No, it can’t be done in any sort of useful way.

Imagine someone set up a new NCP in the UK. How would you ask a question measuring its support? Well, you can’t just say would you vote Conservative, Labour, Lib Dem or NCP, as no one would know what the hell NCP was or what it stood for. But if you tell people in the question all about this new party, then you are essentially giving it a little advert, pushing all its positive attributes in the question and putting it at the forefront of people’s minds, so that would be hideously leading. Also, what about all the other things that drive voting intention apart from policies? What about existing party loyalties? Who the party leader is? What about about how seriously the media take it? What about whether your friends and collegues are all talking about it, or have never heard of it? What about whether it is seen as a serious contender, or as a wasted vote in a tight Con -vs- Lab battle? Whether people would vote for a new party is dependent on so many unknown hypothetical questions it is impossible to expect people to give any sort of useful answer.

Today’s Observer has a news story about another NCP that may or may not be launched. Except, wasn’t there another one of those a month or two ago? And another one a few months before that? Didn’t George Osborne’s old Spad set one up? If you look throught the Electoral Commission’s list of registered parties there are a fair few examples of people setting up new Centrist, pro-European parties over the last couple of years, none of which have made any impact at all. This is not because of their political positioning (I expect all have espoused very similar views), but because no one really noticed them or considered them a serious electoral contender.

A party with £50 million to spend on publicity should at least get noticed, but a lot more will depend on how seriously it is taken. Do established MPs who bring credibility and a voice in Parliament defect to it? Is it reported along the main parties in media? Does it make a leader who is known to the public and seen as a viable, potential Prime Minister? How, under FPTP, does it propose to deal with the Liberal Democrats who are fishing for the same vote? Those are the actual obstacles facing a new party getting off the starting line (let alone actually being a success at elections), and opinion polls can’t predict how well they’d deal with them.


A new centrist party?

With little political news over the Summer the media have entertained themselves with talk of new political parties. I have awaited the first poll to ask how people would vote if there was such a party with some trepidation. thus far it hasn’t turned up. Depending on how it is worded a poll question could either suggest triumph or disaster for such a venture. Either case should be ignored – polls asking about how people would vote in hypothetical situations aren’t particularly useful.

Back before the election YouGov asked a couple of questions asking how people would vote if the Labour party split into a centrist party and a Corbynite Labour party. That found Labour voters splitting fairly evenly between the two parties, with little impact elsewhere (a result that under FPTP would likely have delivered a Tory landslide). Of course that was a new party explicitly framed as a split within Labour. It it had been presented as a split from the Tory party, I expect it would have taken most support from them. A new party might actually seek to present itself as being made up of the centrists within both Labour and the Conservatives (though more important is how it would be seen by the public – how a party describes itself is not necessarily the same as how the public sees it), in which case it would have ambitions to take support from a wider pool.

As an explicit anti-Brexit party the first place to look for what support an anti-Brexit might receive is the EU referendum vote. 48% of people who voted in 2016 wanted to Remain. In more recent polls that group splits pretty evenly between Remainers who still think Brexit is a bad idea but that it should go ahead now the people have spoken, and Remainers who think that Brexit should be resisted and overturned. Some have suggested that this means the pool an anti-Brexit party is fishing in is only about 25%. I’d be less sure – at the moment we’re in a political situation where the political class has largely accepted the principle of Brexit and is arguing about the form it will take. Were that to be shaken up, were there a significant political force arguing for changing our minds, perhaps more of those who voted Remain would see it as something to be fought rather than accepted. Who knows?

A more negative consideration is what one thinks a new anti-Brexit party could offer that the Liberal Democrats aren’t already offering. Normally when there is speculation about new political parties it’s because there is a chunk of the electorate who support a political viewpoint that no party is representing – UKIP wanted to leave the EU when no other party did, the Greens offered an emphasis on the environment and anti-austerity that the other parties weren’t. We don’t have to ask hypothetical polling questions about how people would vote if there was a centrist, liberal, pro-European party standing…we already have a perfectly serviceable party of that description and they got 8% of the vote at the general election.

Ah, you might say, but this new party wouldn’t have the baggage of coalition that the Lib Dems have. Or it would have a better known and more substantial leader than Tim Farron. That may or may not be true, depending on who ended up being involved -serious political figures like Tony Blair or George Osborne would bring their own baggage. On the other hand, a new party wouldn’t have the local government or organisational base that the Liberal Democrats do.

The real difference between a new anti-Brexit party and the Liberal Democrats would be the political context and narrative. It is this that makes it impossible to predict from polling how any such party would do. If a party was set up by a couple of whohe’s it would likely sink without trace – if one looks through the register of political parties at the Electoral Commission you’ll find several new parties set up as pro-EU vehicles, and that none have had any impact. In contrast were twenty Conservative MPs and twenty Labour MPs to defect and form a new party, it would create a huge media buzz, there would be a lot of fuss and attention (needless to say, it would also deprive the government of a majority) and that would give it the potential to get a fair amount of support.

In judging these sort of hypothetical questions, I always look back to the polls we used to see in the final months of the Blair government, asking people how they would vote if Gordon Brown was leader. They would invariably show that Labour would perform less well under Gordon Brown. In the fullness of time Brown did take over, and Labour shot into a double digit lead as all the newspapers treated Brown like the second coming. The problem with those pre-Brown polls was that people couldn’t predict that wave of excitement and positive media coverage, couldn’t predict how they would react to it. Given the right people and media coverage, a new party could succeed to some degree (certainly the currently arithmetic in the Commons would make it comparatively easy for a party with Conservative defectors within it to make an impact). Whether it could be successful enough to actually retain or win seats and have a long term future is an entirely different matter – FPTP does not forgive smaller parties without concentrated support, the anti-Conservative vote is already split and the most pro-remain areas tend to be held by Labour.

In short, it could work in terms of upsetting the current narrative if not necessarily in electoral terms… or it could fall flat, but treat any polling questions asking how you would vote if X party existed with a huge pinch of salt. Without the context of the people involved and the political narrative around it, they simply aren’t good predictors.