Reality check

This is not a tied ball game. Romney’s down a touchdown with five minutes left on the clock.

Update Andrew Gelman points out, reasonably, that a 2:1 proposition sometimes comes in on the short side. So if all “too close to call” means is that the outcome isn’t certain, then yes: the election is too close to call. But it isn’t a toss-up. That’s the point I way trying to make below.

All the chatter – especially from the right wing, but also from the Village Idiots and the Eeyores on the progressive side – suggests that the race is too close to call.


If you’re willing to bet on Romney, you can get slightly better than 2:1 at Betfair. That suggests a 33% chance of disaster. Nate Silver thinks it’s closer to 25%. On InTrade it’s a slightly more discouraging 36%. IEM‘s winner-take-all market is about who wins the popular vote, but its vote-share market gives Obama a 4-point edge, about twice the margin Silver’s model predicts.

Yes, something horrible could happen. But this is not a tied ball game. Romney’s a touchdown behind with five minutes to play.

And Romney just fumbled.

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact:

13 thoughts on “Reality check”

  1. What are the transaction costs at each? I seem to recall it was 5 points at InTrade when I looked into it four years ago. If that’s still the case (if it was ever the case), and if it’s also the case at BetFair, the 25%-to-36% spread is just big enough to give you a guaranteed return on a two-week investment. Granted, not a large return, but still a guaranteed one.

    1. InTrade switched to a $5 monthly subscription and eliminated all transaction fees, about 8 months ago.

  2. Re the ‘fumble’.

    The story is true if it works.

    Always be closing.
    Glengarry/Glen Ross 2012!

    1. Yup. The story Mitt’s pushing is a lie from beginning to end, but it shares enough nouns and verbs with the actual news story that it’ll be damn hard for most people to catch him on it.

  3. I don’t see how this is a fumble. It seems like a typical Republican ploy to mislead low information voters who probably wouldn’t learn the truth in time for the election because the media is preoccupied with other narratives.

  4. Perhaps this would be more difficult if the Republicans had nominated somebody I actually liked, (Liked as a candidate, I mean; Romney is personally likable enough.) such as Gary Johnson. But I’ll be frank: I have no idea who’s going to win, and I’m perfectly content with this. In a couple of weeks I’ll know, and that’s good enough. Has to be good enough, because you can want to know the future all you like, and that doesn’t make it possible to know it.

    I actually find the people, on both sides, who think they’ve got some good basis for predicting the election, kind of laughable. Surely you know enough about the nitty gritty of statistics to understand that the polls you’re seeing are virtually useless, right? Error bars based on sample size are the lower bound for uncertainty, if every factor in the poll comes off perfectly. Entirely random sample, 100% response rate, and so forth. But we know for a fact that the other factors are turning into utter bilge; Response rates are down in the single digits, no reason to suppose respondents are representative of the larger population, BSing polsters is a sport… The real error bars on these polls may be two, three times as wide, nobody knows.

    And prediction markets? Really? I suppose the theory there is that insiders will inject non-public information into the market by trying to make a killing. (Perhaps some truth to that: The polls you read in the paper may be worthless, but the actual campaigns have the budgets for genuinely rigorous polling, and probably do have some semi-reliable information that could leak to investors.) Mix in a little wisdom of the crowd, and out pops a prediction. But suppose the “killing” the big investors are trying to make, is swaying the election? And the number we’re seeing is just a result of competing efforts in that area, not an attempt to make money by accurately predicting the election?

    Embrace uncertainty. It genuinely IS the “reality based” thing to do.

    1. Gary Johnson is a likable dunce. He promotes his ideas as “common sense,” when in fact they would send the US economy into a death-spiral. Cutting spending to 15% of GDP? US on the verge of monetary apocalypse? No. This guy is in the thrall of ideas for teenagers. The President had it right when he said that Ayn Rand was simplistic and laughable. Yet too many hard-headed Americans somehow think that common sense always makes sense. It doesn’t.

      I do, however, agree with your larger point that we have to embrace uncertainty and black swan thinking. But it’s easier when you dislike both of the major candidates for office–you can resign yourself to whatever happens.

      1. I honestly wonder to what extent people end up ‘liking’ one of the major candidates only because they don’t want to admit neither of them is genuinely likable. I know for a fact that many Republicans who ‘like’ Romney on a policy level are just blowing off/excusing a lot of policy differences because Obama is understood to be much worse than Romney on those issues.

        I’m pretty sure the same is going on to a great extent on the Democratic side, too.

        No, I think it’s more a matter of resigning yourself to ‘liking’ one of the choices.

        1. i hear you brett. if i thought romney would be better on drone strikes or global warming i might lean away from obama but i know romney will be worse, much worse, on both issues.

          i have to go with the one who represents the larger segment of my preferences which generally means the democrat which, in this election, means obama.

          1. It’s not only that Romney will be worse. It’s that he’s a black box, a political Schrodinger’s cat, inside which we have no idea what is happening. My belief is that there is a strong wind cycling around between his ears, and not much else.

          2. I believe that Obama isn’t much less a political chameleon; Liberals just like the background he’s trying to blend into. Of course I loathe him, I’m not his target audience, you are.

    2. Embrace uncertainty. It genuinely IS the “reality based” thing to do.

      Brett is so right on this point. It’s a difficult thing for a lot of people, dealing with uncertainties, but it’s unavoidable if one wishes not only an improved understanding of reality, but also an improved cognizance of the bounds limiting that understanding of reality, crucial to judging how well that reality is understood.

      There is no such thing as certainty. It doesn’t exist in the universe. There are only varying degrees of uncertainty. You can’t even be certain about how uncertain things are, you can only estimate that, and the accuracy of that estimation is crucial to the accuracy of the understanding of reality. These polling margins of error have often proven to have been underestimated, and no wonder — coming to an uncertainty estimate on the order of a few percentage points in the extrapolation of the polling data of a few hundred “likely voters” to represent a hundred million actual future votes requires complete ignorance of most, if not all, of the various sources of uncertainty involved in the data gathering and analysis. An understanding of reality based on inaccurate estimation of the uncertainty of that understanding is useless.

      As a metrologist (no, I don’t predict the weather, I practice the science of precision measurement) for nearly three decades now, I have extensive training and experience in the statistical methodologies used in the evaluation of uncertainties involved in precision measurement. Accurate uncertainty analysis is absolutely critical to precision measurement, and the more precise the measurement, the more vigorous the uncertainty analysis must be in order to maintain that precision. I can give you the Volt within a few parts per million of NIST’s representation, or the Hertz within a few parts in 10^-12. I regularly convincingly prove these uncertainties to auditors evaluating my uncertainty analyses and in ongoing round-robin comparisons of artifactual measurements. But nobody, not even NIST, can represent these things absolutely. There is always uncertainty involved, and you quickly reach a point where you cannot reduce that uncertainty without a very reality-based estimation of it and it’s significant sources.

      Uncertainty must be embraced, properly evaluated, and well-understood in order to estimate the quality of any perception of reality, but a lot of people talk in absolutes as if uncertainty didn’t exist at all (while certainty remains the myth), or speak of uncertainty as something to be avoided rather than embraced and understood. This is why I laugh when Kevin Sabet says we shouldn’t legalize marijuana because of the uncertainties of how society would be effected, or Romney says four more years of Obama would be bad for business because of economic uncertainty. It’s like the old Three Stooges routine — Moe: “Are you sure?” Curly: “I’m positive!” Moe: “Only a fool thinks he’s positive!” Curly: “Are you sure?” Moe: “I’m positive!” When you think about it, uncertainty is necessary. How could stock markets function without it? Anyone who truly understands uncertainty knows to embrace and evaluate it in order to use it to advantage, rather than hopelessly wishing to avoid the unavoidable.

Comments are closed.