Archive for May, 2017


NEW PB/Polling Matters podcast: That latest YouGov forecast, pollster wars & TMay’s lost momentum

Wednesday, May 31st, 2017

On this week’s episode of the PB/Polling Matters podcast Keiran is joined by Chris Curtis from YouGov and Adam Drummond of Opinium to discuss the latest polling and General Election campaign developments.

Chris Curtis explains the background to this week’s controversial YouGov model seen in the Times that has suggested we might be facing a hung parliament. Curtis explains what is behind the model, its limitations and how we should read it.

Later in the show, the panel discuss why pollsters have such differing opinions on the state of the campaign and what May must do now to regain momentum and avoid the keys to Number 10 slipping through her fingers. Meanwhile, the team also unveil new polling on who voters trust most to negotiate Brexit and who they think will win.

Listen to the show here

Follow this week’s guests





Theresa’s Tories drop to their lowest level yet on the Commons seats spread markets

Wednesday, May 31st, 2017

And LAB buy level now above 200 seats

There’s been a big shake-up in the betting following the publication by the Times of YouGov’s new election model that suggests that Team Theresa could be net losers of seats a week tomorrow and not have a majority.

Latest CON seats spreads SportingIndex 365-371 Spreadex 365-371

Latest LAB spreads both SportingIndex 195-201 and Spreadex 196-202

Latest LD spreads both SportingIndex 12.5-14-5 and Spreadex 12-14.5

To put this into context. Just after the local and mayoral election earlier in the month the CON buy level was at 400 seats plus while LAB was in the 150s.

All this has happened on the day of the big all party debate which Corbyn has announced that he will be representing LAB. Mrs. May won’t be going and is sending Amber Rudd instead. Corbyn’s presence will certainly add to the occasion and the PM’s absence has become the big political story of the day.

In retrospect it looks like not going to the event might have een a serious mistake by Mrs. May. It could enhance Amber Rudd’s chances of being her successor.

Events from Cambridge are on BBC1 and start at 7.30pm.

Mike Smithson


What if this latest from YouGov proves to be correct?

Wednesday, May 31st, 2017

Alastair Meeks looks at the possibilities & the bets

All the sensible people had decided.  The Conservatives’ epic leads of the early part of the election campaign may have dissipated in part, but they remained set for a hefty overall majority. Then YouGov published their first seat-by-seat estimates, which to the consternation of many showed a hung Parliament.

The reaction of far too many has been that this cannot possibly be right.  Even before YouGov’s analysis came out, I explained why I thought the idea of a surge in support from the young and previous non-voters was less likely than the alternative picture painted by pollsters such as ICM and ComRes.  This is all hypothesis though.  YouGov and Survation are reputable pollsters.  We need to consider seriously the possibility that they’re along the right lines.

This is not going to be another thread about methodologies.   For one, I’m not a pollster and for two, such threads are of what I shall politely call niche interest.  On 9 June 2017, only a few hundred people are going to be concentrating on sampling techniques, filters and such like.  Everyone else is going to be looking at the actual result.

With that in mind, let’s look at what YouGov’s estimates would mean.  First things first, it would mean that Britain would have a hung Parliament.  As I write, you can back that at 8/1 with Betfair Sportsbook.  It’s just one poll from one pollster at present, so that’s probably a marginal bet at present, though I wouldn’t criticise anyone for taking it.  Keep an eye on this – if we see similar polls, you should ensure you have this possibility covered.

The question that many have reasonably asked is where Labour would get their gains from.  The answer is to be found in the seats where Labour are getting their increased support from.  Look for Conservative seats which are younger than average, less white than average and where voters have more capital value in their property than average.  London, its suburbs and exurbs could prove positive territory for them – Theresa May has set out her pitch to the nation founded on a flinty provincialism so she could hardly complain if the spurned capital in turn rejects her.  Battersea (6/1 with Betfair Sportsbook and Paddy Power), Bedford (11/2 with Betfair Sportsbook and Paddy Power) and Brighton Kemptown (5/2 with Betfair Sportsbook and Paddy Power) might conceivably turn from blue to red, even as Halifax and North East Derbyshire might head in the opposite direction.  I’m on all of these.

So about this hung Parliament.  The Conservative and Labour figures would be strikingly similar to the tallies each got in 2010.  That resulted in a durable Conservative/Lib Dem coalition.  That option would not be available this time because the seat tallies of the Lib Dems and the SNP would be almost exactly reversed.

The numbers would be such that only the Conservatives could in practice form a government (Labour is too internally divided to make a government from second place conceivable, even in the highly unlikely event that the Northern Irish unionist contingent, whose shuddering loathing for Jeremy Corbyn could be measured on the Richter scale, and both the SNP and the Lib Dems could be squared off).  That government would be very unstable.

The Lib Dems have already said that they will not go into a coalition after the election, so the Conservatives would be running a minority government, relying upon the kindness of strangers to get Parliament’s backing for the most demanding set of negotiations ever undertaken by this country.  Either the government would seek to roll the dice for another election in fairly short order – and given how this election campaign has panned out, that seems unlikely – or the Conservatives would be forced to compromise extensively on their vision for Brexit in order to secure a strong and stable negotiating position.

What of Theresa May?  She would have sought a mandate to crush the saboteurs, only to find herself sabotaged.  Could she go on?  Unfortunately for her, there is another question: could she step down?  In the confusion of a hung Parliament, with a two year countdown on Brexit ticking, the Conservatives would lack time and space to replace her.  She could be speedily replaced only by a Conservative unity candidate who could work comfortably with other parties, possibilities for which are conspicuously lacking within the Conservative Parliamentary party at present. 

The ideal candidate would be David Cameron, but that option is now unavailable.  The Conservatives might very well find themselves stuck with a leader levitating at the top of the party without any means of support.  If you want to back a next Prime Minister though, Philip Hammond might be worth a flutter on this basis.  I managed to back him at 500/1 on Betfair Exchange to be the Prime Minister to form the first government after the election.  It’s a long odds bet, but not that long odds.

In 2015 YouGov substantially overstated the Labour seat tally and understated the Conservatives’.  For the reasons I have previously given, I expect that they are doing so again.  However, I shall be keeping an open mind and continuing to look for bets that might represent value if YouGov are right.  I’ve been wrong often enough in the past.

Alastair Meeks


On the day ICM gave the Tories a 12% lead, YouGov analysis has the Tories losing their majority

Tuesday, May 30th, 2017

If you YouGov are accurate then Mrs May will be announcing her resignation in 10 days time

I’ve never known anything like it. Such divergent figures from two pollsters, this is the sort of stuff that destroys reputations. YouGov had the Tory lead in the 20s a few weeks, if Mrs May does lose her majority then she really will be the worst Tory leader in history. To lose David Cameron’s majority against Corbyn, that’s just shocking, but after GE2015, Brexit, and Trump, we should learn not to take anything for granted.

I have doubts about these findings from YouGov because it doesn’t match up with the supplementaries/leader ratings.

No one can accuse the pollsters of herding at this election.




Youth and experience. Turnout among 18-24 year olds and past non-voters

Tuesday, May 30th, 2017


In the summer of 1969, I was not yet two years old.  The world watched in awe as Apollo 11 landed on the moon.  My parents sat me in front of the television set so that I could say in years to come that I had seen it.  Of course, I don’t remember it at all.

I try to keep this piece of family nostalgia in mind when thinking about how the world looks to younger voters.  The youngest voters in this election were two years old when the September 11 attacks took place.  The oldest voters in the 18-24 year old band were not yet five years old when the Good Friday Agreement was signed.  They will remember nothing of these events.  The miners’ strike and the Brighton bombing will be as remote history for them as rationing, the death of Stalin and the Suez crisis are for me.  The world looks very different to them as a result.

It is only too easy for older voters to forget this.  Many of them are baffled how anyone could consider voting for someone so morally compromised by his past dealings with terrorists.  Many younger voters are baffled in turn why anyone could think this particularly relevant in 2017.  They seem enthused by the policies that Labour are putting forward, and branding those policies as reheated seventies socialism also cuts no ice with them for the same reason.

Labour have been rising in the polls.  The extent of that rise depends on which pollster’s methods you prefer.  Those largely relying on self-recorded certainty to vote, such as YouGov and Survation, are finding that the Conservatives’ lead has closed to about 6%.  Those who weight this by past experience of voting practices in different groups, such as ICM and ComRes, are finding a Conservative lead in double digits.

Which is right?  The answer to that is not obvious.  It’s often said that the four most expensive words in the English language are “this time it’s different” and anyone betting on the young and previous non-voters turning out in large numbers this time is betting that something different is going to happen this time around.  Yet with the SNP’s performance in the 2015 general election, the EU referendum result and Donald Trump’s victory in the US presidential election, we’ve seen three results that relied on different behaviours in the last two years.  Jeremy Corbyn’s supporters have been quite explicit about their targeting of previous non-voters. Might we therefore see a fourth?

Young voters haven’t always abstained in high numbers.  In 2015 turnout was something like 44% among 18-24 year olds, but it was north of 50% in 2010.  60% of registered voters in this age group voted in the EU referendum.  By definition, this is a group whose voting practices have not yet been set in stone.  They may have the capacity to surprise with their civic-mindedness, especially if they have been enthused by Labour’s offering. 

There are three indications that something more like normal service will continue on 8 June so far as the youngest section of the electorate and previous non-voters are concerned.  First, this enthusiasm has sprung up in the last month.  It is quite possible that part of Labour’s apparent rise in the polls is simply that Labour supporters are more willing to get involved in conversations about polling than previously, given that Labour’s campaign has been more sprightly than just about anyone expected at the outset.

Secondly, any real increase in enthusiasm seems not to be deep-seated, given it was not present a month ago.  It is reasonable to suspect that a relatively recent passion may not motivate non-voters in particular out of their past practices.

Thirdly, the levels of self-reported certainty to vote are implausible.  The recent Survation poll showing Labour on 37% found that 82% of 18-24 year olds said they were intending to vote.  Indeed, Survation found 10 out of 10 certainty to vote at 81% across the whole population.  Either would be unprecedented in the UK.  We can reasonably be politely sceptical about both of these figures, which may reflect either a tendency for those replying to opinion polls to be  unusually interested in politics or a human nature to overstate one’s sense of responsibility.  Some fairly hefty downweighting of both seems required.

We may see some improvement in voter turnout with these groups – probably counterbalanced to a considerable degree by voters unimpressed by any of the parties or their prospectuses.  The overall result, however, is likely to be closer to the picture presented by ComRes and ICM to that presented by YouGov and Survation.  So at present it seems likely that the Conservatives remain on course for a very big victory.

Alastair Meeks


Theresa May’s firewall – the oldies who appear to be staying loyal

Tuesday, May 30th, 2017

Mike Smithson


Author & ex-political journalist, Robert Harris, suggests TMay might be making Heath’s 1974 mistake

Tuesday, May 30th, 2017

During the big C4/Sky News event last night the best-selling author and former political journalist, Robert Harris, posted the above tweet pointing to the similarities between TMay going to the country three years early and Heath’s calling of the February 1974 election with a possible year and a half still to go.

At the time in 1974 the country was going through the oil crisis which was about to be exacerbated by a miners’ strike. The campaign slogan was “Who Governs Britain. This, from Wikipedia, is what Heath told the country in his broadcast after calling the election:

“Do you want a strong Government which has clear authority for the future to take decisions which will be needed? Do you want Parliament and the elected Government to continue to fight strenuously against inflation? Or do you want them to abandon the struggle against rising prices under pressure from one particularly powerful group of workers …

This time of strife has got to stop. Only you can stop it. It’s time for you to speak — with your vote. It’s time for your voice to be heard — the voice of the moderate and reasonable people of Britain: the voice of the majority. It’s time for you to say to the extremists, the militants, and the plain and simply misguided: we’ve had enough. There’s a lot to be done. For heaven’s sake, let’s get on with it.”

The Conservative manifesto, with echoes of today, claimed the Labour opposition had been taken over by “a small group of power-hungry trade union leaders“, who were “committed to a left-wing programme more dangerous and more extreme than ever before in its history”.

All but one poll during the campaign had the Tories ahead and, indeed, the party finished with most votes. Wilson’s LAB, however, had most seats and was able to form a minority government.

By far the weakest part of TMay’s current position is why we are having an election at all. She told Paxman last night it was because the LDs (a party then of just 9 MPs) wanted a second referendum. Eh?

Still I think that TMay is going to win but then in February 1974, my first election while working for BBC News, I believed that Heath would be returned with a large majority.

Mike Smithson


On Betfair the chances of a CON majority edges to lowest level since election was called

Monday, May 29th, 2017

Was a 95% chance – now 84%

During tonight’s Channel 4/Sky News Corbyn/May event I monitored the Betfair overall majority market to see if there was any movement. Half a million pounds is being traded on it every day and the liquidity is there.

The answer was that there was a bit of movement but it is hard to attribute this to the programme. The question now is whether the event and the coverage of it will have any impact on voting intentions.

I thought that both Corbyn and May did OK and I was surprised that the PM was not tempted to attack the Labour leader in anyway whatsoever.

Paxman was appalling with Corbyn and his absence from regular political coverage since leaving Newsnight certainly showed. His whole line of questioning seemed to provide the peg for the LAB leader to demonstrate that he wasn’t quite as left-wing as he’s portrayed. He was much better with May.

TMay overall gave an accomplished performance and was at her weakest when trying to explain why we are having an election at all blaming everything on the Lib Dems. I am sure this might become an issue in the closing phase.

Mike Smithson