Archive for the 'Pollsters/polling' Category


In the only 2018 polls to be tested against real results LAB shares were overstated by 7%+

Wednesday, September 19th, 2018

GE17 LAB polling understatement doesn’t mean that the same will happen next time

One of the things that true believer Corbynistas keep telling me on Twitter is that last year’s general election was a turning point in British politics and that the rules have changed. Thus anything that doesn’t fit into this narrative has to be swept away and dismissed.

A key point here is current polling both voting intention and leader ratings which don’t support the contention that their man is heading for an enormous victory next time. Given how understated Corbyn’s was in most polls at GE2017 it is little wonder that many Corbyn backers raise it in response to less than good current numbers. But is the undoubted polling failure of last year a good pointer to future elections?

Certainly LAB did far better in June last year than most predicted and their leader got most of the plaudits. But for Labour, it is sometimes forgotten, it was a third consecutive General Election defeat and the Tories remained in power.

One of the challenges when trying to assess polling accuracy is that there are very few occasions when election outcomes can be compared with actual pre-election polls. Since GE2017 there has been just when one set of elections when published surveys were put to the test – May’s London Borough elections.

YouGov in partnership with QMUL and the top GE2017 pollster Survation did produce surveys ahead of May’s borough elections in the capital and the party shares together with the GE2017 and the London local election party vote aggregates can be compared.

Those are all featured in the chart above and as can be seen LAB did much worse than any of the polling. The gap on the final polls was more than seven points and suggests that London Labour was being overstated.

Now there were special factors in May. The elections took place a few weeks after the party’s antisemitism crisis broke and the demonstration outside the Palace of Westminster. A feature of the London results was the very high turnout in areas with large Jewish populations and this had some impact on the overall numbers.

In recent years polling of local elections in London has proved to be pretty good. The 2016 Mayoral race was a case in point in 2012.

So I’d argue that the failure of some pollsters at GE17 cannot be taken as a reliable guide to the future in the same way that CON understatement at GE2015 was no guide to what happened two years later.

Mike Smithson


With just over six months to go until Brexit day YouGov looks where the public stands

Sunday, September 16th, 2018

Some good news for Theresa May?

Anthony Wells writes

There is an overwhelming perception that the Brexit negotiations are not turning out well. In our most recent tracker 73% of the public thought the negations were going badly, including majorities of both Remainers and Leavers, and both Tory and Labour supporters. Only 22% of people now think that it is likely that a deal will be struck in time for Britain to leave the EU in March 2019.

A majority (55%) think that the EU has had the upper hand in negotiations, around a quarter (24%) think there has been give-and-take on both sides and just 2% think the UK seems to have the advantage. Only 8% of people expect the government to get the sort of Brexit they have said they want – while approaching three in ten (28%) expect them to end up agreeing a deal for a softer Brexit than they want, and around the same proportion (27%) expect no deal at all.

However, these deeply negative judgements are not placed wholly at Theresa May’s door. Four in ten (42%) respondents think that any other leader would have done just as badly as May, compared to just over a quarter (27%) who believe someone else could have done better.

Neither is the perceived poor progress of negotiations necessarily seen as a reason to replace May. Just over one in five (22%) people now think an alternative leader would get a better Brexit deal, while over half (54%) think that a different leader would not be able to do any better in the time available.

My take from this poll is that if Mrs May gets a sub-optimal deal or no deal she personally won’t take a hit, low expectations may help her in the long term.

For those wishing for the UK to Remain in the UK that Bregret lead isn’t large enough to demand let alone win another referendum. I suspect the Bregret lead will increase a lot in the event of a ‘No Deal’ Brexit. Britons will need to experience the reality of a Hard Brexit if there’s to be any chance of overturning Brexit.



Some pretty grim polling in London for the Tories, Labour, and Sadiq Khan

Monday, September 10th, 2018

Brexit, anti-Semitism, and crime all seem to be having an impact

I think the collapse in the Tory and Lab vote shares is down to a mixture of Brexit not appealing to London and the anti-Semitism issues swirling around Labour.

For me the most interesting aspect has been the collapse in Sadiq Khan’s ratings, albeit he just maintains a net positive rating. The Standard report

Sadiq Khan’s ratings have plunged to their lowest yet after a long summer of violent crime.

The Mayor has fallen behind among key target groups including the over-Fifties, working-class voters, white people and the outer London “doughnut”.

Overall, Mr Khan’s ratings have slipped from +22 in May this year down to +4 today, according to the YouGov study commissioned by Queen Mary University of London.

It revealed that both Labour and the Conservatives have slumped in popularity in London over the summer, with the Tories falling to a dismal vote share of barely one voter in four. And former Ukip leader Nigel Farage, who has hinted at running in the 2020 mayoral contest, is overwhelmingly rejected by Londoners.

Professor Philip Cowley, of Queen Mary’s Mile End Institute, said: “Sadiq Khan has suffered a noticeable hit in his ratings, down from being the most popular politician in Britain just 18 months ago, to being ahead but not by much.

“The 2020 contest is looking a lot more interesting now than it did even before this summer. Justine Greening and other big Tory beasts might now regret their decision not to have a tilt at him.” Key details of the survey include:

Londoners are divided over Mr Khan’s record. Overall, 44 per cent say he is doing well, while 40 per cent say he is doing badly. That is a major change since May when the figures were 52 per cent well, 30 per cent badly.

It is even more pronounced compared with March 2017 when his ratings were 58 per cent positive and 23 per cent negative.

Several key voter groups have turned against Mr Khan. He remains hugely popular among younger voters but now people in the 50-64 age group give him a negative rating of -5.

While Mr Khan was previously ahead among all ethnic groups, now white Londoners rate him at -5, while among black and minority ethnic Londoners he is ahead by +21.

Mr Khan has slipped among the C2DE social class, where he is seen as doing badly by 46 per cent and well by 39, a net -7. He is backed in the wealthier ABC1 social class, with 48 per cent rating him as doing well and 35 badly.

Voters in the big outer London ring now say Mr Khan is doing badly by 44 to 39 per cent, while inner Londoners say he is doing well by 49-35.

So whoever is the Tory candidate there’s potential for them to do well, although after the year Khan’s had that he still has positive ratings might indicate a certain Teflon like qualities that Khan possesses.

I suspect demographics, Brexit, and the fact the next Mayoral election will take place after a decade of Tory or Tory led governments, which will make it sub-optimal for any Tory candidate.

But Sadiq Khan winning in 2020 isn’t the slam dunk many think it is, especially if the crime rates deteriorate further.



Sweden sours? Will the far right make further inroads in Scandinavian social democracy?

Sunday, September 9th, 2018

For the last 100 years, the Social Democrats have dominated Swedish politics.  They have been in government for all bar 22 of those years.  It used its hegemonic status to establish a social democratic culture that worked with the country’s Lutheran ethos and with business, and for a long time managed to preside over a successful and distinctive blend of high taxation supporting a strong social safety net and a dynamic economy.

Their grip has been loosening for a generation.  For over fifty years the Social Democrats tallied more than 40% at successive general elections (a mighty achievement under a system of proportional representation), but they have not hit that mark since 1994 and they have barely scraped 30% at the last two elections.

Sweden goes to the polls again today and that decline looks set to continue.  No poll has shown the Social Democrats getting anything like 30%: they are averaging just under the 25% mark.  The Europe-wide crisis on the centre left looks set to intensify.

This is not the tale that most outside observers are telling of the Swedish election.  All the journalism has focussed on the rise in the polls of the Sweden Democrats, a socially conservative party whose USP is an opposition to multiculturalism and immigration.  Certainly the Sweden Democrats are performing considerably better than at the last election in 2014, scoring in the high teens with most pollsters (and higher with YouGov and Sentio).  This seems, however, to be part of a wider fragmentation of the Swedish vote. 

The Left party looks set to almost double its share from 5.7%.  The Centre Party, the Liberals and the Christian Democrats all look set to see an uptick in their support too.  At the last election the Social Democrats and the Moderates took over 54% of the vote between them (in 1994 they got a combined share of over two thirds of the vote).  This time they look set to get between 40% and 45% combined.  The Social Democrats got a larger vote share on their own in 1994.

This fragmentation, of course, is not unique to Sweden.  Germany, Spain, Ireland and the Netherlands have all seen the same phenomenon.  This has led to a string of fragile governments, unable to take strong action because they lack the stable majorities to do so.  Sweden looks set to join them.

Sweden’s electoral system is essentially one of proportional representation.  There are 349 seats.  Voting takes place in 29 constituencies and 310 seats are allocated internally within those constituencies on a proportionate basis (the constituencies do not have equal numbers of MPs). 

The threshold within a constituency for eligibility for a seat is a 12% vote share.  The remaining 39 seats are allocated to correct deviations from the national vote share that have arisen: the threshold for eligibility for a seat on this basis is a 4% national vote share.  The end result has historically been highly proportional. 

A month ago, when I last looked at this contest, the Sweden Democrats were odds-on favourites to take most seats.  This was surprising, given they had been ahead in few polls.  Since then, the Sweden Democrats have drifted in most polls and so has their price on Betfair.  However, they still hold a slender lead with the same two pollsters who had found them to be in the lead a month ago (though they are well adrift with other pollsters and sometimes in third behind the Moderates).  It is not impossible that they finish top.

Impossible is not the same as likely.  Chris Hanretty has nailed his colours to the mast and estimated their chance of finishing top as being between 0.1% and 12.5%, depending on how you approach the problem.  (I think he was unaware of YouGov finding the Sweden Democrats in the lead, which would up the second figure to 25%.) 

At the time of writing, they were last traded on Betfair for most seats at 3, giving in implied probability of 1/3 that they will finish top. There is therefore still substantial value in laying them, and, conversely, still value in backing the Social Democrats to finish top at the 1.4 mark where they currently sit.

Whatever happens, is almost certain that all the headlines afterwards will be about the Sweden Democrats. The bigger story of fragmentation and the consequent enfeebling of government is likely to be missed. Sweden looks set to join the lengthening list of European countries consumed by introspection.

Alastair Meeks


Your regular reminder on why you should ignore self selecting polls (aka voodoo polls)

Wednesday, September 5th, 2018

The above is screen grab from a ‘poll’ that The Press and Journal are conducting online about Scottish Independence.

Any poll that involves self selection, has no weightings, allows multiple voting, and doesn’t publish data tables can be safely ignored. Because as in the tweet below these polls can be ‘gamed’ to favour one side making the findings even more unreliable.

As an aside, as an Englishman living in God’s own country of Yorkshire, I’ve voted three times in this ‘poll’. I’m fairly certain I won’t be involved in the franchise for a future Scottish Independence referendum, if we have one. It just confirms that anyone can vote in these types of ‘polls’.



The latest Ipsos MORI finding should worry all politicians

Wednesday, September 5th, 2018

The increasing lack of faith in politicians & governments needs to be reversed if politicians want to be ultimately successful

One of the many reasons I like Ipsos MORI is that their polling goes back forty years plus you can help current figures into context and spot new trends.

So this new addition to the Issues Index is very noteworthy, what makes it very noteworthy is that this is based on prompted questions but what is unprompted spontaneously mentions by the respondents.

This finding isn’t surprising as we’ve seen across the world faith in politicians, governments, and national institutions fall, this has seen the rise of what has been characterised as populism

I’d like to see the precise split between a lack of faith in all politicians and the government to see if this driven by a general unpopularity of the government or not. I suspect whatever shade of Brexit Mrs May delivers this figure will increase in the short at least as she’ll end up disappointing a lot of voters.

You can access the full Ipsos MORI data tables by clicking here.



Why you should be wary of hypothetical polling

Monday, September 3rd, 2018

The tweet thread above by the HuffPo’s Polling Editor is very useful when we see this type of polling.

It is the second tweet of this thread that is particularly noteworthy, so the next time you see a poll about voters being more or less likely to vote for a candidate if the candidate did X thing remember this thread.

In the next few days I’ll do a fuller a thread on why for similar reasons polling asking what the voting intentions would be if parties were led by X,Y, or Z are similarly flawed.



A big August polling development has been the restoration of TMay’s double digit “best PM” lead

Tuesday, August 28th, 2018

I love tables like the one from YouGov featured above showing the trend in responses to a political tracker question. This one is on who would make the best prime minister and what seems quite striking is that Theresa May has held up pretty well apart from July when her numbers slipped following the critical response to the Chequers agreement.

The big mover, of course, has been the Labour leader Mr Corbyn who has seen a fair amount of slip back which if you look at the dates sort of links to the start of the anti-semitism row which began at the end of March.

The problem with best prime minister ratings is that the incumbent almost always gets a boost and I think the numbers have to be looked at in that context. The other highlight thing about recent polling on this question is that the don’t knows are now the biggest segment.

What appears to happen is that when Corbyn’s numbers slip it is the don’t knows that increase which could point to something worrying for the Labour leadership.

What might have helped the Prime Minister during this holiday month is that the has been much less focus in the media on the huge Brexit divide within a party. When that starts to be getting the attention of the media again then maybe we could see a slippage.

Mike Smithson