We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Has the day come when election polls are nearly always right?

Famously, in the last US presidential election, Nate Silver correctly predicted the winner of all 50 states and the District of Columbia. His prediction for the election before that was correct for 49 out of 50 states.

Both times, I had hoped it would turn out otherwise. My hopes had been a little higher than they should have been because of the residual glow from the Shy Tory factor, first exhibited to a dramatic extent in the 1992 UK general election and still apparent, though in lesser degree, for several elections after that. I had known about that factor in my guts before that election, from listening to people on the tube, and had correctly guessed the final result would be more Conservative than the polls claimed. As the results came in I did not rejoice that the Government would be Conservative, but I did rejoice that the Chattering Classes had been confounded, their bubble burst, their conversational hegemony broken open and their flary-nostrilled noses put out of joint. Yeah.

Unfortunately not-yeah since then. I haven’t eaten a hearty post-election breakfast with schadenfreude sauce about the polls for many a year now. George Bush winning in 2004 was splendid fun, of course, but it was no great surprise to anyone who had been paying attention. The polls had given him a consistent small lead for months before the election. In the same year there was an unexpected result in the Spanish general election, but that could be attributed to the the Madrid train bombings three days earlier and the cowardice shown by the Spanish people in their reaction to the attack.

In the years since then I have had the impression that polls have been getting ever more accurate. But my attention has wandered from politics so my impression might be wrong. In recent months the approaching Scottish independence referendum has rekindled the old flame and I have begun to follow the polls. If you want to know, I am of the Unionist persuasion, but it is one of those questions where my libertarianism isn’t telling me which way to steer; and in this post I am not arguing either way. I am just observing that the polls diverge and wobble much more widely than they seem to for either British or US general elections. Is that because it is a referendum rather than an election? I would expect the simplicity of a yes-or-no referendum to make prediction easier, but polling for the voting system referendum in 2011, while correct about the result, did significantly understate the vote to continue with the First Past the Post system, causing my heart to beat faintly once more to the happy rhythm of 1992.

Here is an interesting article by John Curtice, Professor of Politics at Strathclyde University, entitled Scotland’s referendum: can we trust the polls? Mind you, despite being a professor of politics and running a referendum polling blog he does not actually say whether we can or cannot. Quite like old times.

11 comments to Has the day come when election polls are nearly always right?

  • Tedd

    Notwithstanding that the word pollster was originally coined because of it’s similarity to the word huckster, I don’t see any reason in principle that polls couldn’t be quite accurate. (Barring some unexpected event occurring between the poll and the election, such as the Madrid train bombing). It seems like a fairly straightforward sampling and analysis problem. Even factors such as the “shy Tory” effect or self-reporting error ought to be quantifiable.

    I”m sure technology helps a little — computers and data analysis software, online surveying, and so on. But I suspect that inaccuracy in polls in times past was more a function of low competition, low expectations on the part of pollster’s customers, and possibly even deliberately biased results, to some degree. If polls have gotten more accurate it’s probably mostly a product of greater competition (which, admittedly, is mostly a by-product of technology).

  • Mr Ed

    The margin of error in polling is capable of being 100%, since what a poll measures is what the pollsters are told, not what the voter shall do or has done.

    Add the long history of socialist violence into the mix, and it is hardly a surprise if polls overstate support for socialism, plus the ‘political’ bias of polling itself, assuming an interest in politics, the rewarding of failure and the punishment of sucess, which is, of itself, socialist in nature.

    And the dead are notoriously shy about their voting intentions.

  • Libertarian

    As a Florida resident, I feel the same way about hurricane tracking/predictions. No more excitement, sense of suspense, etc.

  • Alsadius

    I suspect a big part of it is that data aggregation has gotten much faster. You can dump big poll results into a hopper and do demographic adjustments to make them more accurate within seconds today, when it would have taken days in 1992 when such things were often still done by hand. As such, the best polls available on E-day were a few days stale then, and could miss a last-minute swing.

  • AngryTory

    Well we certainly can in NZ. In spite of what some leftist commentators say, the polls have barely moved since 2007.
    As a result anything other than a National government in NZ will be illegitimate

    Of course -many people believe the whole idea of a Labour or Green government in what is supposed to be a free-marked capitalist country is by definition illegitimate – but that’s not what I mean here. Simply that the “will of the voters” has been very clear since 2006 – and of course the “will of the taxpayers” and especially the “will of the nett taxpayers” has been even clearer.

    Which is why, for the foreseeable future, but certainly for the next 6 if not 9 years, anything other than a National government in NZ would be illegitimate.

  • Michael Jennings

    As for that 2004 Madrid train bombing, the Conservative Spanish government spent three days lying about the likely perpetrators between the bombings and the election: they claimed it was domestic Basque terrorists ETA when they knew full well it was al Qaeda related. For these lies they manifestly deserved to be voted out. As to whether their left wing opposition deserved to be voted in, that is another question. I’m not sure that it fair to simply accuse Spanish voters of “cowardice” for voting as they did, however.

  • bloke in spain

    ” the Madrid train bombings three days earlier and the cowardice shown by the Spanish people in their reaction to the attack. ”
    As a Spanish resident I strongly deplore that remark. As Michael Jennings says above, the government of the time lied. Everyone knew they were lying. They paid the price.
    Madrid’s nervous of its people. Not without reason as events during the crisis have shown. You want to look for a truly craven people, try that big island off the Atlantic coast of France. They can be pushed around any way you like & barely give a murmur.

  • Natalie Solent (Essex)

    I wondered if someone would call me out on that line, Michael, and carried on typing anyway in a spirit of “Bring it on”.

    Let me say at once that all judgements about the actions of “the Spanish people” or “the British people” or “the Muslims” or even “humanity” are not to be taken literally as applying to all individuals, or even a majority of individuals, and, further narrowing the scope, a judgement about the actions of a people on one particular occasion or at one point in history does not imply racial essentialism. It would be almost impossible to discuss history if instead of just saying, for example, that “the Japanese treated prisoners badly in WW2″ one had to add every time that most Japanese people never even saw an Allied prisoner, and those that did often were coerced into committing crimes by brutal military discipline.

    By that standard of everyday non-literalism I think my reference “the cowardice of the Spanish people” on 14th March 2004 is fair. I think the real reason that a substantial proportion of Spanish voters changed their vote was in order to beg Muslim fanatics to bomb someone else next time. I followed the story very closely in 2004 and my impression was then and still is now that the mis-identification of the bombers as ETA was (a) more tentative and (b) less the conspiracy of the “government” as a whole rather than of certain bodies and individual spokesmen than later accounts made it seem. I remember at the time there were plenty of Spanish official voices saying that it was likely to be Al-Qaeda.

    People would rather think of themselves as having righteously punished a government for lying than of having handed Al-Qaeda the huge victory of having changed the result of an election by mass murder, and hence of having given Al-Qaeda a big incentive to bomb again.

  • Natalie Solent (Essex)

    bloke in Spain,

    ” You want to look for a truly craven people, try that big island off the Atlantic coast of France. They can be pushed around any way you like & barely give a murmur.”

    Did I deny it?

    A minute ago I googled “the cowardice of the British people” for purposes of comparison and found this eloquent denunciation. I don’t agree with all the author, evidently a left winger, says, but I agree with a lot of it.

  • bloke in spain

    Just a small word on the Spanish people’s attitude to Islam. For the Brits, it either happens in countries a long way away or in areas of their major cities. Islam, for me sitting here typing this, is a grey smudge on my southern horizon. With the wind in the right direction, I can smell it. It tends to concentrate the mind.

  • James Waterton

    Silver’s achievement was impressive, but not quite as impressive as it sounds. Any chump could have correctly “guessed” 42 of those 50 states plus DC. So basically Silver guessed 8 correctly.