I got it wrong. I got it spectacularly wrong. I even went so far as to trumpet how wrong I was in the opening paragraph of my final prediction of the result. I mentioned – not in passing, but bold as brass – that I’d never forecast the Tories to win the most number of seats. Not once. And I went on to mention that I still wasn’t forecasting the Tories to win the most number of seats. I didn’t even hedge my bets and suggest it could be a tie.
I got it all wrong. I spent the entire year getting it wrong. Why?
Part of the answer is I thought the polls were more of a guide than a snapshot. I didn’t listen to my own advice, repeated many times on this blog, to never try and extrapolate anything from the polls about the election result. I concluded that because so many people were predicting the same sort of thing so many times, there must be something in it. There wasn’t. There was nothing in it.
In particular, I thought Lord Ashcroft’s constituency polls were more of a guide than a snapshot – thereby disregarding the man’s very own catchphrase, and tending to treat his findings not as a reflection of temporary opinion but of permanent intention. His polls in Scotland were mirrored at the ballot box; his polls in England were not, sometimes spectacularly so.
I also got it wrong because my instincts were wrong. I couldn’t believe somewhere like Hendon, with a Tory majority of just 106, would not swing cleanly back to Labour. I live in the seat right next door. I thought sensed what would happen. Worse, I thought I had a conviction about what would happen. I even believed my own seat of Finchley & Golders Green was about to change hands, almost solely from what I had seen and heard of the campaign around me.
All of that was wrong. Something was missing from my calculations. Something that was not evident anecdotally, or in the polls, or arithmetically, or perhaps even rationally. It was the unexpected. I didn’t expect it. And so I got it wrong.
14 responses to ‘Some reflections’
I wouldn’t beat yourself up over it, Ian – you are in good company as no one got it right and at least you (unlike Ashcroft) didn’t squander millions of your own money of pointless polls.
I’m sure that it was something rational, and ultimately predictable in hindsight, that caused the difference. The question is – what?
Consider how many polls there were, using so many different methodologies, phone (with or without mobiles being contacted), online, naming and not naming candidates, using actual ballot papers or not. Any trends, late surges, online vs phone differences etc. of a magnitude sufficient to explain the final result should be evident somewhere in the data. But there is nothing convincing (despite what Survation say about their own final unpublished poll). Consider that the constituency polls seemed correct, even to those most familiar with the constituencies – there was one in the city where I live that seemed correct to those most familiar with its politics (on all sides) but was well out., even though it was conducted quite close to the election.
Consider also that the exit poll was broadly correct, but that polls taken on the day (YouGov etc) were still incorrect. I wonder: maybe there was an on-the-day swing that explains a part of it,, but what’s the one key difference with the exit poll? Surely it is that 100% of the people in it have really voted. Perhaps the model of the voting portion of the electorate is wrong. Now, what has happened recently to change that? And what are the effects of the changes to voter registration?
Just echoing what the other Alex said, everyone got it wrong. Even the Conservatives themselves were not expecting to do so well – just look at the surprise on the Tory candidate’s face at Ed Balls’s count.
Personally I thought the Conservative campaign was as dull as dishwater, and moreover quite nasty at points. David Cameron may yet find himself paying a heavy price for his victory – the break up of the union via an English nationalism he helped to create, and a party split down the middle by a referendum on EU membership. Or maybe that’s just sour grapes?
From my experience on the ground, there was a big swing to the Conservatives. Especially here in a ex-LD held seat, the Ashcroft poll showed the LDs borrowed lots of votes from natural Conservative supporters, and the Conservatives ran a very effective campaign to get these people to vote Tory instead of LD in the last 2 weeks of the campaign.
We all got it wrong, but it made for exciting viewing through Friday morning. I believe the one thing the polls didn’t show us was the number of ‘don’t knows’ which was considerable,and it seems many were undecided right up to when they had the pen hovering over the ballot paper. The analysis will now begin. Labour took their supporters for granted (especially in Scotland). They should have spent some of their time replaying the Kinnock ‘Alright !! speech in 1992 to show what complacency does. Lib/Dems paid for doing nothing in particular during the coalition. As for UKIP and the Greens they have a solid base to build on. Don’t expect another voting referendum. If anything the Conservatives will now be studying constituency borders to strengthen their position for next time. A parliamentary majority of just 8 over a five year fixed term (factoring in the Sinn Fein seats that will go unused). Will it be enough?. The fun and games are not over by a long way. Thank you for the blog its been a daily must and a bloody good read. see you in five years (or maybe before).
“Lib/Dems paid for doing nothing in particular during the coalition.” Pardon? Among the Lib Dem successes in government that you seem to have missed were their responsibility for the triple lock on pensions, the pupil premium, the rise each year in the income tax threshold (£800 over the five years), the Green Investment Bank as well, of course, as quite a lot of deficit reduction. Perhaps you also missed some of the things they stopped from happening, e.g. a Greek-style financial crash or Portuguese/Spanish-scale austerity, the introduction of ID cards & a national databank for everyone’s DNA, employers being able to fire employees at will, like-for-like replacement of trident. That is not “nothing in particular” from the junior coalition partners (with only 57 MPs). Don’t rewrite History.
They did a lot, but they didn’t persuade the voters that they deserved much credit, and they were punished at the polls. The Tories stole their policy, for example, on the income tax personal allowance.
Don’t worry Ian as I and so many others including many Con candidates also got it wrong.
Let’s see what British Polling Council analysis comes up with. I am inclined to think for future elections we may need not only constituency polling but also regional polling as I think National Uniform Swing is over perhaps forever.
Firstly, I’d like to thank you for producing your entertaining and enlightening blog. I’m sure that many, like me, have enjoyed and benefited from reading it. You may have joined all the others in getting the final conclusion wrong but you gave us all food for thought and an insight in how to look at developments.
It has been a long time since my university politics classes – you helped to bring me up to date and rekindled my interest.
We were all astonished at how the night developed and the questions about that will continue into history. As Alex E says, Ashcroft spent millions & got it as wrong as you.
However, we are now left with a fascinating situation which could cause immense social, economic and political damage to the UK. Cameron has vilified Scotland and created a huge pool of UKIP voters who feel disenfranchised. Cameron may yet regret his victory as it has left the UK divided as never before. He has allowed the bankers to keep their bonuses and given the poor food banks. In Scotland he is now viewed as an alien invader.
On top of all that he will now have increasing difficulty in keeping the different wings of his own party together. And then we come to the European referendum…………..
The election is over, and it would be entirely understandable if you wished to stop the blog. However, I, for one, would be very pleased if you could keep something going, even with less frequent postings.
Your blog has always been fascinating and informative. Thank you for creating it and sustaining it on a daily basis, and if you feel like even occasional posts hereafter (e.g. on EU referendum and constituency boundary changes) then I for one will read them with great interest. I certainly hope you’ll revive it for 2020.
Ian. As a fellow election obsessive, I would like to thank you for writing this blog. You did get it wrong, but so did everyone. Even I, a committed conservative, thought a majority was absolutely beyond us. Thank you so much for all your political insights, consistently fascinating, delivered to me daily.
See you in 5 years time 😉
Thanks for all the comments, folks. I’m going to do a few updates over the next fortnight or so, looking at various aspects of the results and what they mean for the next election. Beyond that, I’m not sure at the moment.
Thanks, Ian. Many of us have greatly enjoyed your incisive and thought-provoking blog. Well done.
I think that the most important thing that the pollsters have got wrong is to leave out the “Don’t know” party from their results , and at times compound the felony by “allocating” in various ways these voters to whatever other party in order to come up with a !00% “who will vote for whom” result. Bring back the Don’t knows !! Also the thing about this overall result is that very few seem to have thought that what really happened is that voters re-allocated their votes in many ways no doubt , the actual Conservative vote was well under 1% more than 2010 but I sure was made up from votes coming from various directions (Liberal Democrat , Labour , in particular) that compensated for the defectors to UKIP , and beneath the simple result was a great deal more than the normal of “toing and froing. The main point of all is rather (to use a football analogy) like the Premier league this season , since Christmas 2014 it has been Chelsea’s to lose (they didn’t) , in the case of this election it was the Labour Party’s to lose …and THEY DID.