« Does downward nominal wage rigidity matter? | Main | Where is Marginal Revolution? »
For the curious
At about 8:30 a.m., Tradesports.com is giving the Republicans a 13 to 15 percent chance of Senate control. The Webb-Allen race, described by newspapers as "too close to call," is being called for Webb; Allen's chance is about five percent.
Addendum: Chris Masse reports: if Virginia and Montana go Democratic, as of morning yesterday, the prediction markets called every race correctly.
Posted by Tyler Cowen on November 8, 2006 at 08:39 AM in Current Affairs | Permalink
Comments
a win for libertarians! of course they are not taken actual leadership positons, but as much as could reasonably be hoped. The international community has also been sent a message that the US no longer whole heartedly backs torture. Perhaps not as lofty an issue as we may have wished, but we have to start somewhere.
Posted by: theCoach at Nov 8, 2006 9:14:25 AM
Big swing in the value of the Senate contract, it was pricing a ~70% chance that the Republican's would retain the senate yesterday.
Posted by: nelsonal at Nov 8, 2006 9:17:10 AM
How is a victory for lockstep Democrats a win for libertarians, theCoach?
Posted by: Neal at Nov 8, 2006 9:29:50 AM
Man, then tradesports stunk it bad predicting, since two days ago, it was 75% Reps would retain the senate. Is this normal for contracts to become volatile before closing?
Posted by: Jason Voorhees at Nov 8, 2006 9:34:11 AM
Tyler, the prediction markets were way off. They only started giving the Senate to the Democrats after the Dems had basically won. It's like saying I was correct in calling the game for the Redskins but doing so only after that crazy play with no time left that setup a second game winning field goal.
Posted by: eriks at Nov 8, 2006 9:56:51 AM
Nonsense,the market was predicting > 80% at one time that the Republicans would keep Senate control,they wont if Montana and Virgianis go Dems.
Posted by: jck at Nov 8, 2006 10:02:59 AM
The volatility before closing is a feature, not a bug; it reflects the prediction markets assimilating new information as it comes in from exit polls, partial returns, etc. IMHO, the prediction markets were much better and faster than the major media at calling various races last night.
But yes, calling this accurate prediction is highly misleading. it's like saying that a rain gauge on your roof "predicts" weather better than a meteorologist. You need to judge predictions on a clear time frame shorter than "after polls close".
Posted by: DK at Nov 8, 2006 10:24:12 AM
To Masse's point that if Virginia and Montana go Democratic, then the prediction markets called every race correctly. That is strange, still, because the contract on GOP keeping the Senate was calling it a Rep win at high probability, but Masse's saying the individual contracts on the races were saying something different. Why didn't someone arbitrage that away - if the states were in aggregate moving Dems, then I don't see why the larger contract on the entire outcome wouldn't have been traded away. Lots of money left on the table sounds like.
Posted by: Jason Voorhees at Nov 8, 2006 10:29:30 AM
The relevant question is whether the prediction markets got it right before people voted, not after the exit polls, partial returns, etc. started coming out.
Posted by: Andy at Nov 8, 2006 10:42:03 AM
"The relevant question is whether the prediction markets got it right before people voted"
Exactly,and they didn't...the Senate was a coin flip and that's not worth 70%
Posted by: jck at Nov 8, 2006 10:49:52 AM
I'm surprised Chris stopped masterbating about BetFair long enough to type out a post for you. (Speaking of arbitrage, Betfair still gives nearly even odds. With a rake of 5% and a good 5k on the table, that's a hefty profit.)
Posted by: Chi at Nov 8, 2006 11:02:21 AM
I think I'll add to the voices of the other commentators. I watched the prediction markets closely last night, and the "GOP Senate" contract was trading at ~86 when I went to bed. The market failed badly on the Senate. Why?
Posted by: Giant Step at Nov 8, 2006 11:03:48 AM
The prediction markets weren't crazy. Remember that for the Democrats to take the Senate, they had to win a whole bunch of individual races. Even if the Dems were favored by, say, 70% in each of the five races, the odds that they'd win all five was under 17%, assuming that the outcome of each race was an independent event.
Posted by: Anonymous at Nov 8, 2006 11:16:19 AM
Neal,
I have no idea what you mean by lockstep, but a Dem win means that the Congress will not be in lockstep with the President, a.k.a. divided govt. PLus, it is hard to imagine a more anti-libertarian group than the President and the GOP congress - Secret trials, torture, warrantless phone taps, coupled with pandering to religious right social restrictions. Some people seem to think that lowering current tax rates while balooning spending is cutting taxes, but I do not believe that. In my opinion, a better metric for taxes is actual spending, and by that measure Bush has most certainly been worse than a Democratic administration would have been.
I would have thought that would be obvious to a libertarian (perhaps not a glibertarian or a conservatarian ;)). However, I am most decidedly not a libertarian. Perhaps a better take would be the guys at Reason.
Posted by: theCoach at Nov 8, 2006 11:30:11 AM
Interesting observation Anonymous. Did I mention I got a C in stats/probability? Now I remember why.
Posted by: Jason Voorhees at Nov 8, 2006 11:38:37 AM
Right, consider two races that are 70 percent likely to go to Dems and 30 percent likely to go to Pubs. Here's the binomial probabilities of the aggregate results:
Dems win 0 - .09
Dems win 1 - .42
Dems win 2 - .49
Even with the Dems heavily favored to win each race individually, they would only win both 49% of the time.
BTW, the republicans definitely deserved yesterday. Although the media has trumpeted it as some rebuke for the Iraq war, the democratic gains mainly happened because of the Republican's corruption. Tradesports was actually showing that Republicans would keep the house before the Foley scandal.
But this isn't a win for libertarianism by any stretch of the imagination. Maybe it will help a bit on immigration, but the out of control pork and government spending will only continue or increase. Also, many of the democrats won on a protectionist, xenophobic platform. There's no doubt that this is a huge set back for free trade as America takes on this kind of European thinking.
Posted by: Matthew at Nov 8, 2006 11:58:39 AM
Its easy to be right when you pick both sides. Yesterday, tradesports showed the GOP keeping control of the senate, but the democrats winning the individual races necessary to take away control. http://volokh.com/posts/1162881852.shtml
Posted by: guy in the veal calf office at Nov 8, 2006 12:02:49 PM
Think Barron's will issue a mea culpa?
Huh?
Posted by: Bill Stepp at Nov 8, 2006 12:21:38 PM
if Virginia and Montana go Democratic, as of morning yesterday, the prediction markets called every race correctly -- which is to say, the lower-value contracts were probably overpriced and the higher-value ones underpriced. If predication markets are right all of the time, then they are underpricing variance.
Posted by: Grant Gould at Nov 8, 2006 12:26:42 PM
veal calf office,
Thanks for the link. Let me expand on my point earlier with the data you gave me.
Here's the probability of a democratic win in each Senate race:
VA: 57%
MO: 60%
RI: 66%
MD: 73%
MT: 77%
Remember that the democrats have to win each and every one of those races in order to gain control of the Senate. This means the probability of the democrats taking over the senate is:
.57*.60*.66*.73*.77 = .127
Now, this analysis only includes the closest races. Other states, such as Tenn., were not close according to Tradesports, with only a 15% probability that Ford would win. It's possible for a computer program to take all 33 (34?) senate markets and calculate the probability of a democratic takeover, but I don't have the time for that and that number shouldn't be too far off from .127.
So, tradesports really picked republicans in both the individual senate markets and the senate control market. I'm guessing the traders underestimated how much support the democrats got.
Posted by: Matthew at Nov 8, 2006 12:43:36 PM
All the explanations of how the Dems can be favored in every race but not to take the overall Senate ignore the fact that there is a very high correlation in the results. In some sense, the big unknown before the election was the degree to which people would vote locally (uncorrelated) or vote nationally (a few national issues on which opinion swings are highly correlated across districts). Tradesports et al clearly got the correlation structure wrong.
Posted by: DK at Nov 8, 2006 12:44:20 PM
Your probabilities are assuming that each Senate race is an independent event. I don't think that is a safe assumption at all.
Posted by: Sebastian Holsclaw at Nov 8, 2006 1:20:04 PM
I wonder if the prediction markets factor in possible voter fraud into their models. I think they must. It would also be interesting to see how much of a difference it makes. I think it would make a pretty big difference since we have elections that are being run so tight. The race in Montana had a difference of how many votes between the democrats and the republics? I think it was about 1,200. So just goes to show that small amounts of voter fraud can swing elections. On a side note, The state of our election process is pretty dire as I see it. The mechanisms that we use to vote can't really be trusted. This problem needs to be solved. Don't get me wrong though, I'm happy with control being distributed amongst the parties. However, for future elections, it would be nice if failsafe voting systems were installed so that there won't be a question regarding accuracy and count.
Posted by: Government Gopher at Nov 8, 2006 1:40:38 PM
First, you must specify the forecast horizon. Clearly the forecasts needed to be made before the voting started, but should we take the probabilities from a year ago? Probably not.
And second what does prediction even mean in this context? Remeber the markets give us a probability of some discrete outcome, they do not predict(or forecast) that outcome. Rather they give us a frequentist notion of how often the outcome would occur, could we run the election a large number of times.
The reality is we have a sample of one. We can not tell if the probabity was correct, in any sense from that.
As for those that have calculated the joint probabilities, I believe you are all mistaken as treating each election as independent is not appropriate in an environment where people are said to be voting as though this is a referendum on Bush and Iraq.
Posted by: Taggert J. Brooks at Nov 8, 2006 1:57:44 PM
Hello Professor Tyler Cowen and all the commenters,
#1. Professor Lance Fortnow made a specific point: the TradeSports's prediction markets for the individual 2006 US Senate races were accurate (provided that Virginia and Montana go democratic).
#2. Professor Lance Fortnow DID NOT SAY that the TradeSports's prediction market for the control of the US Senate was accurate. Please, don't put words in his mouth.
#3. Analysis reports from economists and statisticians are coming, but, please give them time to digest the data... once the dust has settled.
Thanks for your attention,
Chris Masse
Posted by: Midas Oracle at Nov 8, 2006 2:02:14 PM
