Reason Writers Around Town

November 7, 2006

Over at Time.com, David Weigel takes stock of the bettors who, unlike pundits, actually have something to lose if they blow an election prediction: their money.

Comments

|11.7.06 @ 11:46AM|#

Money talks and Bullshit walks.
Everyone has an opinion, but the people willing to bet on it are the one's to listen to.

|11.7.06 @ 1:41PM|#

Yeah...money money talks STUPID, every day, in Lost Wages NV.

I have no reason to believe that the on-line bettors at tradesports are any more "rational" than the devotees of sports-betting, video poker or stock market day-trading.
Most of these guys know a lot more about sports, poker or the market than I do...and most of them regularly lose huge gobs of their own money. They are addicted plungers...not the smart money.

|11.7.06 @ 4:15PM|#

I agree with Andrew.

The guys betting college football know a lot more about college football than the guys betting on politics - and they still lose money.

|11.7.06 @ 5:03PM|#

It's not that any individual knows more about politics or football or anything else - it's the aggregation of all those estimates of probability, made with a fairly strong incentive to try to get it right, that give us some insight into the overall probability (or expected value) of an event. The fact that lots of people lose money is actually evidence of the market's overall accuracy. The way to make money is to bet against the current market because you believe it either under or over values a particular concern. The more the market price approximates the true odds, the harder it is to beat and the more losses you would expect from those who try (this is akin to why most mutual funds lose compared to the broad market averges like the SP 500).

Of course it is important to remember that the price is estimating the odds of the event - so it is to be expected that it won't get it every binary prediction (win/lose) right - it would be rather remarkable if it did since there isn't much difference between a 49% chance of winning and a 51% chance. But then the claim isn't that it is perfect, but that it represents the best estimate available of the true odds - better than any individual pundit's prediction.

As a purely anecdotal aside, as of the night before the 2004 presidential election the individual state markets for president ended up exactly coinciding with the final state-by-state outcome.

|11.7.06 @ 8:08PM|#

Andrew & Joe:

What I mean is, if you're sure about something, bet on it. If you don't bet, you probably don't know what you're talking about. But betting your opinion requires that you use your smarts & balls, that's why I think a better's opinion counts more than a talkers.
Proposition betting is different than casino gambling.