Ryan Avent says that if I hate political prediction markets so much, why don't I bet on them and make a bunch of money. Well, I don't do that because it seems to me that it would call my professional work into question. I don't want it to be the case that if I say stuff about the candidates I get accused of trying to juice my contracts.
« Are Hedge Funds a Giant Scam? | Main | Liquid Ban »
Money Meet Mouth
29 Dec 2007 05:20 pm
Comments (18)
How about wagering on NBA basketball?
I bet if you always wagered on a home team with 2+ days of rest versus a road team on the second game of a back-to-back you'd make some real cabbage. Fresh legs = $$$.
Look at this chart:
http://128.255.244.60/graphs/graph_DConv04.cfm
From just looking at that chart, tell me the day the Iowa caucuses were held in '04. Then tell me if prediction markets are leading indicators or lagging indicators.
I don't want it to be the case if I say stuff about the candidates I get accused of trying to juice my contracts.
I mean this in the nicest way, but anyone who made that accusation would be seriously overexaggerating your significance as a pundit.
Matt's original point is correct the markets don't predict anything. They merely reflect conventional wisdom. He doesn't have to be a fortune teller to prove that the gypsy with the crystal ball is full of it.
If you are so great at political predictions then why don't you post your 'bets' on this site.
The correct rebuttal would be to cite any instances in which the political markets predicted the outcome of an election or a change in a candidates polling numbers before the polls themselves did.
Maybe the better response to why don;t you try it is because you do not believe you have a special insight as to who will win.
That does not in any way change your point that these markets are merely lagging indicators of CW.
OF course if someone is good at predicting who will win they can make money. Just because you are not that person does not in any way take away from your point that these markets provide no special insights.
"Maybe the better response to why don;t you try it is because you do not believe you have a special insight as to who will win."
He doesn't believe that. He implied that he thinks Hillary is overpiced.
Not many people claim the markets are magical. I would claim that most people don't know any better than the market prices and can't make money at them.
From a quick look at the real money political prediction markets, Matt is definitely right. There are major price discrepancies between intrade and iem prices on most of the potential nominees. Edwards, for example costs over 60% more on iew than intrade. This pretty conclusively tells me that 1) there's not enough trading volume in the markets for it to be worth someone's time to game these discrepancies, and 2) there's not enough trading volume in these markets for them to be accurate predictors.
There is a lot of misunderstanding about what the prediction markets are supposed to do.
They don't create new information so much as they aggregate it. It is not that easy to take the welter of polling data and other information and consolidate it into a single number. Thats what the prediction markets do for you.
The markets allow anyone to look at a single summary number that gives them a very good estimate of what the current outlook is. That's without looking at any polls, knowing anything about the primary calendar, etc. And it's not easy to get a better prediction any other way, otherwise people would put lots of money down and push the prices back towards the better estimate.
Great point Ed. It's true--the markets just summarize current sentiment (weighted by the amount of confidence people have as measured in dollars). Depending on how loose you wanna be with your terms, you could call this conventional wisdom. I will just add that it is easier to call the market prices conventional wisdom after the fact than before the fact. There wouldn't be any conflict of interest for Matt Yglesias to put up a list of contracts on this site now that he believes are over or undervalued. You could even do this as a regular feature. He does not need to put any of his own money on the line. Then us readers can look back after the election and see how those predictions fared. If they were successful, then reading this blog gives you insight into the outcome of these races beyond so called conventional wisdom. If they were not (or statistically equivalent to guessing), then I'd think Matt would have to eat his words. You can test your predictions without putting money on the line.
There is a lot of misunderstanding about what the prediction markets are supposed to do.They don't create new information so much as they aggregate it.
No, they do, in theory, a bit more than that: they screen out random noise and errors with high enough volume trading. Of course, "random" is a key word there. Market actors have systematic biases, especially in political futures markets.
"High enough" is also key, even for reasons beyond decreasing the margin of error attributable to random noise and errors. When biases can be directly seen to move markets because of the size of even small individual trades as compared to the market as a whole, there's a lot of incentive to act on bias, especially if people consider the markets as serious predictive tools. Cheap advertising is always sought in political campaigns.
R Johnston's points about volume and systematic biases are the key here. I've noted wide divergences between the iew and intrade markets before. If this was occurring in high volume markets someone would be arbitraging those differences until they disappeared. Since it isn't then all the "wisdom of crowds" theories about the predictive value of markets are largely inapplicable.
Matt, your argument is circular. Presumably, the reason you would be betting on any contract in the first place would be because you honestly believed it was under or over priced. And so then anything you said in your blog that could possibly aid your bet would be something you would be saying anyway because you sincerely do believe it.
The only possible breech of ethics could be if you decided to say something you actually did not believe, in an event where for some reason you thought you could move the market further in the direction of a false conclusion than in (what you believed to be) the correct conclusion. But the very nature of the market is to make that situation much more rare than the normal situation where you'll make the most money by both betting and blogging in the direction you really believe in.
(Unless you're in the NRO crowd or something, where history suggests what you really believe will happen tends to be the opposite of what happens.)
Matt, your argument is circular
I think Matt's argument is that the political markets are of no value to those of us interested in knowing who's really leading in popular support. You win in these markets by predicting who other bidders consider the frontrunner, not by actually knowing who's gaining with the public. You can win in these markets by picking a losing candidate (say a Thompson last August or a Dean in December 2003) and getting out before the rest of the pack realize it.
The markets gives no fundamental insight into the race itself.
http://online.wsj.com/article/SB119902559340658043.html?mod=hps_us_editors_picks
“More recently, in the 2004 primaries, prediction markets pointed to the disintegration of Howard Dean's candidacy in advance of the fateful Iowa caucuses. In the 2004 presidential election, the market favorite won the Electoral College in all fifty states; in 2006 the markets also picked every Senate race.“
Does anyone else have this good a record?
Matt: do you want to reconsider your comments?
@ neil wilson
"Does anyone else have this good a record?"
That is key IMO. These markets work. I think Wolfers was referring to the InTrade market.
Even if you dont like (fear, mean, moral whatever) to put your money where your mouth is, as seems to be the case with Matt, then logically you cant dis the informaiton from the markets.
There is an easy solution to this conflict: pre-announce all your trades before you do them.
Comments closed January 12, 2008.
| 
Heads in the Sand
what a crock of shit.
If you are so great at political predictions then why don't you post your 'bets' on this site.
Put down the trades that you would make on paper and see how well you actually do.
It is a whole lot harder in real life.
Posted by neil wilson | December 29, 2007 6:10 PM