2009 New Jersey Gubernatorial Race: Was Nate Silvers prediction more accurate than InTrades?

No Gravatar

Nate Silver&#8217-s prediction (November 2, 2009): &#8220-I&#8217-d make Christie about the 4:3 favorite.&#8221-

[ UPDATE: Nate Silver’s prediction post-mortem on the 2009 US elections.]

You can see that days before Elections 2009, InTrade was too heavy on Corzine:




Share This:

Statistician and econometrician Nate Silver is going to write up a book on… SUSPENSE, SUSPENSE… not on statistics, not on econometrics… on prediction markets.

No Gravatar

Via the unjustly suspected mister Cowgill, the LAT:

[Nate] Silver is now at a crossroads, trying to decide what to do with his newfound fame. He’s about to leave Chicago, where he now lives, and move to New York, and he’ll spend the next few months working on a book on prediction markets.

I bet that book will be more impartial that the writings of our economic canaries who are in bed with the exchanges and the vendors.

It wasnt about the predictions.

No Gravatar

Let&#8217-s not confuse media visibility with utility. Aside from the depressed Obama-to-win prices on one exchange, prediction market and polling aggregation results for the 2008 election were essentially the same using squared errors. Despite his insane schematics, Emile Servan-Schreiber has a good point about capturing the interest of the public, something that nerdy academic and libertarian-types aren&#8217-t necessarily good at. An Obama-backing baseball statistician out of Daily Kos nailed that part this year, a year where people were especially skeptical of markets, not to mention unregulated &#8220-offshore&#8221- ones. Likewise, if you put down the lens of considering markets as commission generators, you&#8217-ll see the value of contracts tied to social and cultural outcomes. Of one the biggest assets of prediction exchanges is media goodwill, which should be fostered by distilling information on subjects like global development and art prices.

Other things to keep in mind:

  • This year happened to have a lot of favorite-longshot States, which turned-out to be favorable to 538&#8217-s error relative to markets.
  • Prediction markets register information in real time. Since the difference in error is small, this is important.
  • Markets are more flexible, and useful in situations where you don&#8217-t have a rich data set and obvious statistical analyses. Elections are just one type of question. Even if you have data, it might be less expensive to set up a new contract than to undertake the analysis.
  • And of course, prediction markets have functions aside from forecasting, and provide incentives for uncovering new information.

Is InTrade being manipulated? Why does InTrade give a discounted probability for Barack Obama as US president?

No Gravatar

A quick link panorama.

#1. Is InTrade being manipulated?

– Nate Silver shows that there are abrupt downward pressures on the Barack Obama event derivative, while we also see some abrupt upward pressures on the Hillary Clinton event derivative.

However, you can see by yourself that InTrade is resilient enough and does a great job of going back to normal [*], after just a few hours of trading:

– At Portfolio, blogger Zubin Jelveh blows the incidents out of proportion.

– Professor Lance Fortnow has a more careful analysis and notes that the price of the Barack Obama bounces back quickly enough.

– Quick thought: Maybe the media should use an average of event derivate prices for the last 5 work days&#8230- so that the abrupt perturbations would be eliminated.


Professor Eric Zitzewitz:

I’m not sure you can conclude from Silver’s graphs that the market goes “back to normal.” You can conclude that it moves back in the opposite direction of the impact those large trades. Back when the Hillary for President market looked like it was being manipulated, it appeared that the manipulator was both placing a large purchase and then placing limit orders to provide price support and slow down the reversion of the price.

UPDATE: Are we witnessing manipulation attempts on the &#8220-Florida to vote Republican&#8221- prediction market at InTrade?

#2. Why does InTrade give a discounted probability for Barack Obama as US president?

– As you remember, Emile Servan-Schreiber of NewsFutures believes that it&#8217-s a Republican conspiracy all over.

– Professor Justin Wolfers puts up an hypothesis: it&#8217-s legally impossible for US traders to arbitrage on BetFair.

– InTrade put up a crappy excuse: the industry is still too &#8220-young&#8221-. How lame. How stupid. The industry was younger in the previous elections, where arbitrage opportunities didn&#8217-t exist according to professors Justin Wolfers and Eric Zitzewitz (see their 2004 paper and their other publications).

– Blogger Zubin Jelveh swallows the InTrade P.R. line, and adds another crappy InTrade P.R. line: More arbitrage opportunities are being exposed in open air because much more observers are hunting down arbitrage opportunities in 2008 than in previous elections. That&#8217-s a second blatant cretinery, uncorrected by the Portfolio blogger. Re-read Justin Wolfers&#8217- blog post. Professor Justin Wolfers states that:

The current variation in price is larger than I have ever seen in my years of studying prediction markets. The forces of arbitrage that would typically eliminate these differences have been handicapped by the legal restrictions preventing U.S.-based traders from using overseas markets.

– Finally, professor Lance Fortnow says nothing about the arbitrage opportunities between InTrade and BetFair, but does offer some technical points about the issue of polls versus the prediction markets, centered around the question of state correlations. Read on.

UPDATE: Eric Crampton (a Canadian exiled in New Zealand) says he has managed to turn a buck by arbitraging between InTrade and iPredict New Zealand. He also makes 2 theoretical points. Go read it.

UPDATE: Greg Mankiw just linked to Nate Silver.