This thread will end well. Buy at 61.7; sell at 60.1.

May 16, 2008 8:54 AM

The Promise of Prediction Markets (full text link; .pdf here). A group of distinguished economists and other scholars has published a call to exempt prediction markets (previously on MeFi: 1, 2) from American laws that restrict internet gambling.

The group was supported by the American Enterprise Institute's Reg-Markets Center, which includes several of the authors on its advisory board. They suggest that the Commodity Futures Trading Commission - which recently sought public comment on the issue - establish a "safe harbor" for small-stakes markets. PM boosters are ambivalent about the idea.

posted by googly (24 comments total)

So, gambling for the educated only then?
posted by phrontist at 9:10 AM on May 16 [3 favorites]

The presidential nomination market at Iowa Electronic Markets has been less a predictor of the future and more a daily update on what CNN's talking heads are suggesting today. Observe the huge Obama leap after Iowa and the huge Obama collapse after New Hampshire.

Still, I'm sure some people made some money.
posted by East Manitoba Regional Junior Kabaddi Champion '94 at 9:12 AM on May 16

I predict the Bears will beat the Vikings by 7 points. Any takers?
posted by squashotheclown at 9:16 AM on May 16

Is it just me who can't read the PDF in the post? If not, you can grab is from Brookings.
posted by quiet at 9:21 AM on May 16

less a predictor of the future and more a daily update on what CNN's talking heads are suggesting today.

I agree with the first half of this sentence. As for the rest: Until last week, all the talking heads were calling it "close" and "a horse race" while the IEM had Obama up around 90%.
posted by DU at 9:27 AM on May 16

Intrade is what I use. Sure call it gambling, all I know is that it is like having the stupidest bookie in the world. I'm sure it'll start to attract more professional players, but right now there's a lot of easy money. Did anyone see the swings on Obama and Hillary? It is like watching cable news, except cable news has no grounding in political reality. I mean sure Obama could lose, it is a possibility, but considering the payoff and the option price, there's no way you can say it was a bad buy during some of those down periods.
posted by geoff. at 9:34 AM on May 16 [1 favorite]

I'm skeptical of the usefulness of these even under the most favorable conditions. A perfectly efficient market will perfectly aggregate the information available to its participants. Okay, fine. I don't think the incentives are such that actual campaign insiders are going to be among those participants in large numbers, so you're just getting an assessment of lay opinion of the meaning of the polls, results, and general trends, expressed as a probability. Is that kind of neat? Yes. A big enough deal that I actually care? Meh.
posted by shadow vector at 9:41 AM on May 16

I don't believe in imaginary profits ( except when they are monetized and appropriated by me)
posted by elpapacito at 9:42 AM on May 16

Wow that's interesting - I had no idea these were deemed illegal back in The United States!

What's the rationale for banning besides online gambling = online sinning?

On a personal note, I'm pretty active on Tradefair's Binaries site, but they only match in financials - Commodities / Currencies / Indices and Interest Rates.

Like geoff's experience on Intrade, there are lots of ill-advised trades being placed on Tradefair. You sometimes wonder if these folks really know what it is they're trading, some of the positions are that far out of touch with normal market movements. Still, it is a pretty fast and easy way to pick up a few hundred quid here and there, mostly when news is expected that's likely to move the market (or not if you take the other side of the trade). But it doesn't seem as though these places have a lot of liquidity; at Tradefair I rarely see more than a few hundred pounds sitting on the table.

In any case, I only trade when I've got a view and money is piling up on the other side - its no fun running with the crowd!
posted by Mutant at 9:54 AM on May 16

"I don't think [...] that actual campaign insiders are going to be among those participants in large numbers"

Looking around various campaigns over the last four years I'm wondering why you would put much stock in what campaign insiders think. Terry McAuliffe? Patti Doyle? These are the people you'd look to for gambling tips?
posted by Ragma at 10:01 AM on May 16

Intrade is what I use. Sure call it gambling, all I know is that it is like having the stupidest bookie in the world. I'm sure it'll start to attract more professional players, but right now there's a lot of easy money.

A friend of mine from the poker world noticed that too and turned a very nice profit during the primaries. It was written up in the New York Times and documented in detail in his two+two thread. I think you can be sure a lot more skilled poker players will be trying to beat this market in the future.

Frankly, I'd much rather see the whole of UIGEA scrapped than establish an exemption, but I'd take what I can get.
posted by Lame_username at 10:04 AM on May 16

Hmm, prediction markets ... you mean like figuring out through workouts, past performances, conditions and the like which horse in a race is going to come in first, and deciding whether or not too many people also like that horse to make the odds of a parimutuel wager to win worth the risk?

Is online paramutuel gambling legal in the US? I see little difference.
posted by mrgrimm at 10:07 AM on May 16

Is it possible to imagine a scenario where a political candidate hedges their campaign spending by betting on futures predicting their opponents' victory?

Granted the authors in this case are proposing that the prediction markets be limited to "small-stakes" contracts, in which the capital deposited by an individual would not exceed $2000. But who knows? Perhaps someone could find clever ways of having individual campaign donors contribute these "small" futures contracts...
posted by Kabanos at 10:10 AM on May 16 [1 favorite]

Is it possible to imagine a scenario where a political candidate hedges their campaign spending by betting on futures predicting their opponents' victory?

Even better, where a candidate sabotages his own campaign in order to capitalize on a long-odds prediction.

Imagine Barack Obama preparing some reverse October Surprise (admitted drug addict?) just because he bet big on McCain to win. It would have to be something big enough to cost him the election but small enough to keep his nomination. Homosexual affair?

By the time the truth comes out, Obama's sitting on millions in Argentina. I smell Showtime Original Movie.
posted by mrgrimm at 11:22 AM on May 16

so you're just getting an assessment of lay opinion of the meaning of the polls, results, and general trends, expressed as a probability

Part of the reason why these markets have such great predictive value is because they are not an aggregate of "lay opinions" or a reflection of the general population; the vast majority of IEM's traders in the 2004 presidential election market were "well educated, affluent, white, male Republicans."^
posted by youarenothere at 11:42 AM on May 16

Am I the only one getting the subscription wall for the article, then?
posted by Shepherd at 12:44 PM on May 16

Because Republicans are just great at making predictions.

Erm, when it comes to mainstream U.S. politics in the context of prediction markets, it seems that the well-educated, affluent, white, male ones are, yes.
posted by youarenothere at 12:47 PM on May 16

Part of the reason why these markets have such great predictive value is because they are not an aggregate of "lay opinions" or a reflection of the general population; the vast majority of IEM's traders in the 2004 presidential election market were "well educated, affluent, white, male Republicans."

It doesn't matter who they are, it matters what information they have access to. Do these players have qualitatively different access to information than what's available to any reasonably diligent person with an internet connection?

Note that it doesn't matter at all that they may have superior information processing skills compared to the average person. The market, by virtue of being a market, already takes care of sorting out which people are more efficient players than others.

If the answer to my question is no, and I think it is no (was John Kerry playing this market in 2004?), then I reiterate my reaction of "who cares." Why should I be impressed that the market has better predictive power than polls? It's based on the information contained in polls, plus additional information. It would have to be a completely nonfunctional market to fail to win in that comparison.

I really don't understand why people think this stuff is revolutionary.
posted by shadow vector at 12:53 PM on May 16

mrgrimm writes "Imagine Barack Obama preparing some reverse October Surprise (admitted drug addict?) just because he bet big on McCain to win. It would have to be something big enough to cost him the election but small enough to keep his nomination. Homosexual affair?

"By the time the truth comes out, Obama's sitting on millions in Argentina. I smell Showtime Original Movie."

Well, I know you aren't serious, but if we're going to bat this around ... It would almost certainly be more lucrative to win the presidency. How much did Bill Clinton get for his book deals again?
posted by krinklyfig at 1:12 PM on May 16 [1 favorite]

It doesn't matter who they are, it matters what information they have access to. Do these players have qualitatively different access to information than what's available to any reasonably diligent person with an internet connection?

This is exactly my reaction to these kinds of "markets". The reason a real market is so effective is because it pools lots of information that no one person has. But these kinds of markets are really just pooling a large number of people's analysis of basically the same information.

They aren't pooling information, they're pooling opinion. Which might be slightly more accurate that any given expert's opinion, but it's not really what people usually mean when they talk about "harnessing market forces." It's really more like the Delphi method than a market.
posted by straight at 3:20 PM on May 16

It would almost certainly be more lucrative to win the presidency

That, in a nutshell, explains the Bush Administration.
posted by aeschenkarnos at 7:16 PM on May 16

hmmm, the real flaw in this is that some single person can easily change the value of a political candidate and make their bet before announcing the news (Genifer Flowers). Also, as far as the "real" markets are concerned, what is stopping the Nigerian rebels from placing a huge bet on oil futures right before they are about to attack a pipeline considering the price of oil seems to jump depending on their activity.
posted by any major dude at 10:19 PM on May 18

Many equity and commodity derivatives are just as much like gambling as predictions markets, and pose just as much of a risk of insider misconduct (like intentional sabotage or trading on insider information). Professional traders routinely "bet" indirectly on market occurrences using these instruments.

Yet these instruments are legal and well-regulated. Why do we let people gamble like this? There is a social good that comes from speculation: commodities futures allow buyers and sellers to lower their risk by locking in prices ahead of time; speculation on equities provides market liquidity, allowing capital to be allocated efficiently.

Similarly, prediction markets have a socially positive component, but also raise a possibility misconduct. Why should we view them differently than financial markets?
posted by goingonit at 9:32 AM on May 22

And I'm going to side with the people who think that high-stakes prediction markets should also be free. I can buy and sell snowfall futures; why not presidential-election futures? What if I own a bumper-sticker-making business, and the election exposes me to a big risk?
posted by goingonit at 9:36 AM on May 22