pageok
pageok
pageok
According to Tradesports Betting, President Bush Will Pick John Roberts:
Although Tradesports.com gave John Roberts less than a 1% likelihood of getting the nomination as of about two hours ago, right now it is giving Roberts a 99.5% chance of getting the nomination. Looks like Tradesports was right after all -- at least after the announcement was made.
Mark:
Wow from 1% to 99% in a few hours -- that's an outstanding return for a brave few. From what I've garnered in the last hour or so myself, I surely hope it's correct.
7.19.2005 9:42pm
Anon, Who Will Stand Up to the Bully? Or Not:
Actually, Orin, exactly 2 hours before your post, Roberts was trading at 10% and 1 hour and 50 minutes before your post, Roberts was the favorite at 30%, as my post above shows.

Jim Lindgren
7.19.2005 9:57pm
guest:
How exactly did that provide any useful information? It was correct an hour before the news reports? OK, but was there any way to tell at the time that the upward move of Roberts was any more reliable than the surge in Clement earlier in the day? I don't see how that move, at the time it happened, could have been discerned from any other incorrect move in the market. If we are only able to know ex post that it was right at 6:47, but we didn't know at 6:47 that it was right, then so what?
7.19.2005 10:17pm
OrinKerr:
Jim, if Tradesports was completely and utterly wrong about Roberts until 2 hours before the announcement, what does that say about Tradesports? That it is useless for the several months leading up to the moment the choice it is announced, but that there is a one-hour window in which it might have some potential valuable?
7.19.2005 10:18pm
anonymous coward:
I agree with Orin that Tradesports doesn't impress.

(From Lindgren's next post) "The market anticipated by about an hour the NBC announcement at about 7:47 that AP was reporting Roberts as the nominee." But his own chart shows Roberts trading a hair above 10 around 7:40, and he never traded as high as either Edith at her high until around the time of the AP report.

I don't think the spike to 30 and back to 10 has any particular significance, either--probably just bid-ask bounce. How wide was the bid-ask spread at that point? (Anyone save the quotes?)
7.19.2005 10:34pm
frank cross (mail):
Some of you are expecting a little too much from markets. Even financial markets don't measure the true value of the company, they only measure the perceived value of the company. We know on average that perceptions are pretty accurate, but they're never perfect (Enron) and their accuracy varies with circumstanced. Likewise this market only measured perceptions about Bush's choice.

As Orin noted, this is a circumstance where one might think markets are relatively weak (save for leaks). They certainly are weak if the White House used misdirection, as it appears may have happened. And this is a thinly traded market.

Tradesports is certainly not clairvoyant. But attacking the market is something of a strawman. It provides a measure of the contemporaneous state of perceptions, and that's all it purports to provide. The criticism makes sense only as applied to those who overclaim the significance of the results.
7.19.2005 11:03pm
guest:

criticism makes sense only as applied to those who overclaim the significance of the results.


See, e.g., Lindgren, James.
7.19.2005 11:20pm
42USC1983 (mail):
It's interesting that Prof. Lindgren will use Prof. Kerr's comments section to defend his position, but will not enable comments so that his position can be analyzed. As a famous law prof would say, Heh.
7.19.2005 11:35pm
OrinKerr:
Jim,

One reason that Tradesports may have seen an uptick on Roberts about two before the annoucement may have been this post at The Corner at 5:24 pm:

********************************
I SAY DERB'S WRONG, WRONG, WRONG [K. J. Lopez]
Prediction: Bush will nominate a white male tonight. Just because they said he couldn't.

John Roberts. Bill Pryor! John Bolton!! Karl Rove.

I'm serious about one of those guys, by the way (and, no, I'm not being a right-wing lunatic this time, so this one's an easy guess).
Posted at 05:24 PM
********************************

Twenty minutes after that, Lopez explained that she didn't mean Pryor, leaving clear that she meant that she was predicting Roberts would get the nod. I assume that other similar posts appeared around the blogosphere at around the same time, and someone read those posts and placed some bets that raised Roberts from 1% to 10%.
7.20.2005 12:32am
KenB:
And according to a post by Erick at redstate.org at 6:42 PM, "Newrooms in Washington and Senate staffers have begun pulling data on John Roberts." This same Erick had earlier in the day pointed first to Clement, then to Jones.

So AFAICT, TradeSports merely followed the same rumor mill that many in the blogosphere did. Let's put aside the question of whether TS did a "good" or "poor" job -- my question is, where's the added value? What could we learn from TradeSports that we couldn't have picked up from a bit of blogsurfing? And with a blogger, you have some chance of gauging how reliable the opinion might be, but with the market, there's no way to know what information is driving the movement of the prices.
7.20.2005 12:56am
Jim Lindgren (mail):
Guest commented:



It was correct an hour before the news reports? OK, but was there any way to tell at the time that the upward move of Roberts was any more reliable than the surge in Clement earlier in the day? I don't see how that move, at the time it happened, could have been discerned from any other incorrect move in the market. If we are only able to know ex post that it was right at 6:47, but we didn't know at 6:47 that it was right, then so what?



This is exactly the point. A well functioning market should provide the best estimate of who the nominee will be at the time of the trade. It does not indicate that this is the final move. If a move were known to be final (and a prediction locked in), then the trading should be approaching 100. I never said that Clement or anyone else would be the nominee. I was reporting what a trading market said was the best guess at the time. Very few people this morning thought that Roberts would be the nominee. I shared the view of most experts that Bush would choose a woman or a minority. And I would expect that at 7:40pm (about 6 minutes before the first seemingly definitive press story) Tradesports was one of the few sources suggesting that Roberts was slightly more likely to be the nominee than anyone else. It is a way of aggregating the opinions of those who are willing to risk money on their expectations.

[I posted this earlier, but it came through under some goofy anonymous name, so I am reposting it.]
Jim Lindgren
7.20.2005 12:58am
chris:
One way to think about TradeSports is how it, well, trades sports. I recall watching a world series game and seeing how every ball and strike in a key inning was moving the price. Now, of course, since one of the teams actually won and the price was never 0 or 100 during the game, tradesports "got it wrong." What tradesports does is give you a quantitative handle (through real people betting real money) on how bad, in terms of how much the price (or probability), was the fact that the pitcher just put the winning run on first, or, for some other event, some other piece of information that has predictive value.
7.20.2005 1:07am
Zaoem (mail):
The Roberts jump is undoubtedly a consequence of "insider trading," which is probably allowed on Tradesports. All that the market does here is to give insiders incentives to reveal valuable hidden information to outsiders really fast. (i.e., some outsiders are foolish enough to bet with insiders, which gives the latter group the incentive to come forward rapidly and cash in).
7.20.2005 1:48am