on optimism

(note: I am a Democrat, and I support Obama.)

Is the Obama optimism justified?
In a word, yes.

http://www.intrade.com/ does basically legalized betting on the outcomes of political processes. The idea is that the market as a whole is more accurate than any individual person, (and more accurate than polling as well,) because money is on the line. So this is a pretty good place to see the consensus view of the race.

I was looking at the numbers and here's where it stands now:

Democratic Nomination:
Obama 72
Clinton 28

General Election:
Obama 48
McCain 38
Clinton 18

(Note, the numbers don't have to add to 100, since these are share prices, but they should be close. (inaccuracies mean inefficiencies))

So, the market seems to see a greater than 60% chance of a Democrat taking the white House, and a roughly 70% chance of Obama taking the Democratic nomination. For comparison, during the 2004 race, Kerry was never trading above Bush for any significant amount of time.

Now, the futures markets are not perfect predictors, but they are a good guide to what the informed conventional wisdom is. For someone like me who gets a lot of his information from secondary or tertiary sources (on purpose! it's filtered!), having a tap on this kind of brutally non-partisan information is pretty valuable. It helps keep me grounded.

Specifically, for this election, there are a lot of factors out there that make me optimistic, but as a scientist (more on this subject later) it's important to me that I have some kind of numerical evidence to back up my anecdote-derived hypotheses. These markets are providing that evidence.

So yes, despite having been burned in the past, I am optimistic this year.


  1. Here's the thing that bothers me about intrade (and probably any sort of stock market): it very much exists in the present, with little future sight. So, yeah, it may reflect current wisdom. However, the trading number for the candidates six months ago are very diffent than they are now. Back then, it was totally justified to say that McCain was dead in the water.

    Essentially, I'm saying intrade acts as a snapshot of the current thinking, but doesn't really take into account the actual momentum of a candidate (at least, not in a way that I think is meaningful). Maybe its average over time is more statistically accurate. Or maybe not.

  2. Well, in theory it represents the best prediction of the future that the market is capable of making, but yes, in practice the markets can be very swingy, and can be caught by surprise by all sorts of things. (Look at the Iowa and New Hampshire swings in the Democratic Primary market for a god example.)

    Anyway, it's predictive data that isn't purely based on current polls, it's mostly unbiased, and it's interesting. ;-)

  3. How are the predictions affected by self-selecting demographics which are typically higher income and more educated? (I don't see a lot of lower-income, hard working yet uneducated Hillary supporters with curiosity/time/literacy/interest going online and spending (even small amounts of) money on prediction markets. They'll just watch the evening news for entertainment and infotainment, right?)

    Whether or not they are always accurate (though in time they will always correct, as the event horizon approaches) they can say a lot (current aggregate of wisdom and meta-opinions of many people) by saying very little (a few numbers). Who wouldn't want more unbiased (or more accurately, completely biased in possible all ways) one dimensional news to compete with the usual Large News Chumpany spin?

  4. Haha, I never thought of that. Though on balance, I always figured the markets would have a Republican bias, what with their love of the free market and all.