All those of you who think Mitt Romney’s going to win by a landslide: why aren’t you down the bookies?

Get dahhn ‘ere lad and stick yer life savings on Romney

There is a certain breed of Internet Person who thinks that the Main Stream Media (or, occasionally, Lame Stream Media: I may need to revisit this at some stage) are either a) deliberately pretending that Barack Obama‘s going to win the US election or b) so deep in denial about the coming Zom-Romney Apocalypse that we can’t bring ourselves to face reality. Take for instance this quote from underneath Dan Hodges’s blog earlier:

Excellent tactic Mr Hodges – obviously Romney is going to win and you are setting yourself up for the ignominious next day admission you were wrong by getting in the excuses early.

I’d like to ask Mr UlyssesReturns how much money he’s put on this sure thing. Because at the moment, in a two-horse race, you can get 11/5 on Romney – better than tripling your stake. If I was as confident as so many people seem to be, I’d look at those odds and think something along the lines of “cher-ching!”

It’s not only commenters, of course. Michael Novak at the National Review has written a post explaining “Why Romney will win”, only a few hours ago. Frank Donatelli, showing that great minds do indeed think alike, has written an identically titled piece at Politico.

Both these guys, along with everyone else who has written a piece about why Romney will win but didn’t use the words “romney will win” in that order so it didn’t show up in my Google search, are doing something interesting. They’re looking into the same crystal ball as everyone else – all the poll-tracker sites, the bookmakers, the exchange markets – and seeing something different. Or are they?

The thing is, of course, that no one – except the “bold swashbuckler” Dan, who doesn’t muck around – is quite saying “Obama will definitely win”. If they were saying that, they’d stop taking bets, or they’d put “100 per cent Obama 0 per cent Romney” on their forecasts, or have Obama’s stocks going at infinite price. They’re saying “on the balance of probability, taking into account all the polls, it looks like Obama is the more likely victor”. Most of them are putting it around the two-to-one mark; some lean slightly more to Romney, some go as high as 90 per cent likely for an Obama win. But that’s the ballpark.

As researchers of prediction will tell you (read Dan Gardner’s excellent book Future Babble for more on the subject), the only way of telling whether someone’s predictions are reliable is to check whether they were right in the past. But when you’re talking about percentage-prediction, such as “It’s about 70 per cent likely that Candidate A will win”, you shouldn’t expect Candidate A to win every time. You should expect their “70 per cent” predictions to come in about 70 per cent of the time. As Nate Silver of FiveThirtyEight says:

Not only will the underdog — Mr. Romney — win some of the time, but he should win some of the time if we have estimated the odds correctly. If the set of candidates you have listed as 67 percent favorites in fact win 95 percent of the time, or 100 percent of the time, you’ve done something wrong. Over the long run, such candidates should win two out of three times — no less and no more.

(We love Nate Silver around here, by the way.)

So I have a theory about what Novak, Donatelli et al might be doing, whether consciously or not. If, as expected, Obama wins next week, no one who predicted it is going to look that prescient, because it’s what’s expected. But there is a significant possibility – a less-than-50-per-cent possibility, of course, but in a two-horse race it’s not nonexistent – that he won’t, that Mr Romney will win.

And then, if you were one of those long-shot bettors, the Novaks, Donatellis and so on, then you will look like quite the soothsayer. You won’t, of course, have had any more reason than anyone else beforehand to believe it to be the case (except for “gut instinct”, of course, but let’s face it if you think with your gut then you have no one to blame but yourself when you are wrong). But one time in three, you get to bask in the glory. So in answer to my own headline question: people who say they think Mitt Romney’s going to win have already made their bets. But their stake isn’t cash, it’s credibility, and the odds are excellent.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s