My friends over at Tradesports are catching some heat. Tradesports is an online prediction market that allows you to make bets on all sorts of unusual outcomes.
A controversy is brewing, though, over the outcome of a contract on whether North Korea would have a missile launch before July 31. Reading the newspapers, you might think that North Korea did a test missile launch, but the folks at Tradesports, based on the specific wording of the contract, decided that the launch that took place did not qualify. You can read about the controversy at Chris Maase’s website and in SmartMoney magazine. The SmartMoney article has a nice discussion of why these prediction markets might ultimately prove to be more important than just a way for gamblers like me to have some fun.
According to the SmartMoney article the controversy boils down to the following:
One of the contract’s rules is that “the source used to confirm a test missile being launched and leaving North Korean airspace will be the U.S. Department of Defense.”
The problem is that, according to Tradesports spokesman Matt Bonner, they made “numerous efforts to receive direct confirmation from the DoD” but were told “no statement involving the missile test and North Korean airspace would be forthcoming, as those specifics are considered a matter of national intelligence/security.”
Bonner emphasized that “a confirmation source is, by definition and necessity, an integral part of the proposition on which contracts trade” — and said that traders are “obligated to be familiar with the rules of a contract before they place an order.”
I have no strong opinion on the outcome — I think it is a tricky call. There are a couple of points that this controversy raises in my mind:
1) Tradesports has no direct stake in the outcome. Unlike a bookie, they don’t take a position. They just create contracts and provide a place for people to trade against other people. This is important, because Tradesports has no incentive to make the wrong choice for their own profits. You want the person deciding whether a contract won or lost to be indifferent to the outcome, and Tradesports is. The contract terms may have turned out to be the “wrong” ones ex post in the sense that the contract didn’t match public perception. But these markets are in their infancy and will improve with time.
2) There is nothing like creating a group of people with direct financial interests to create and stir controversy. I have been contacted by at least five different people who lost money on this Tradesports decision. I can’t think of another issue not directly related to Freakonomics or one of our columns that has spontaneously led so many people to independently write me.
3) I am unable to find out exactly how many contracts were outstanding on July 6, the day that North Korea fired the missiles. I don’t think the total dollars at risk were very large — maybe only $10,000 or less. If I were Tradesports, I would go back to that data and just go ahead and payoff both side to generate goodwill, good publicity, and make future traders confident that in the case of future controversy their interests will be handled fairly.
As for me, I’ll stick to betting on the really important current events, like who will win American Idol.
17 Comments
Hello professor Steve Levitt,
Thanks for airing this controversy. You’re brave.
I’d like to say to your readers that very few scholars have dared expressing their viewpoint publicly, because scholars need the data coming from TradeSports (and BetFair) to feed their research and publications. They fear retaliation from TradeSports (which has already retaliated agaisnt me for airing this dispute).
As you said perfectly, this issue is complex and there’s no clear cut. As I wrote, all opinions are respectable (expire to 0, expire to 100, unwind all trades and void all bets, or pay both sides).
That said, one thing bothers me a lot: How TradeSports treated its customers.
The intent of the contract was whether NK would launch missiles. It became whether the US DOD would confirm the NK missile launches after July the 4th, but that was not wanted originally by TradeSports.
Which is why TradeSports apologized to its customers and promised to carry some reforms.
http://www.tradesports.com/aav2/news/news_58.html
The problem is that TradeSports apologized and promised to reform, but they did nothing to compensate the victims.
What’s problematic with TradeSports is that their customers can not complain anywhere. As you know, TradeSports is illegal in the U.S. And contrary to Betfair, TS is not a member of an independent arbitration service.
http://www.ibas-uk.com/
And TradeSports is not regulated in Ireland, contrary to BetFair, which is regulated in the U.K.
http://www.gamblingcommission.gov.uk/
So if you’re a disgruntled traders (mostly U.S. residents, where TradeSports is illegal), where do you complain? Not to the US courts. Not to trade associations. The only people who will listen are some sportsbook guide, where TradeSports has just been downgraded to C+.
http://www.sportsbookreview.com/Search/default.aspx?Search=tradesports&Area=Site&Submit=Home
This situation is very sad, and I hope that TEN CEO John Delaney will come forward to make the necessary reforms to TradeSports / InTrade.
Thanks.
— Chris. F. Masse .COMTo me it seems that they did the right thing.
By paying off both sides would they not increase the ‘moral risk’ of kicking up a storm on a controversy. If the contract is clear and their ruling is fair based on that contract they would set an awful precedent by paying both sides.
It seems to me that paying off both sides would not only cost them money this time, but also involve them in litigation on every decision that could be challanged. In the end this would eat away at their goodwill.
By sticking to the terms of their contract they’re saying that they will do this in all situations thus making people take proper notice of the contracts and in the long term engender a feeling of trust that what people sign up for is what they get.
This to me seems a better long term proposition.
— KiritIt seems to me that they should have refunded the money to all bettors. Their ‘race’ did not take place since the arbitor/judge did not show up (so to speak). That is, if you bet on something and the outcome is not judged by the sanctioning or judging body, then it did not happen at all. So, they should just refund the money and make more of an effort to find someone who will authoritatively determine the outcome.
— pkimelmaKirit:
I respect your thoughtful views, but in following your proposition, we win the battle of contract enforcement, and lose the war of credibility of predictive information markets.
I invite you to check this out at your convenience http://caveatbettor.blogspot.com/2006/08/north-korea-contract-update-department.html
— caveatBettorFor those of you with an interest in some economic theories on the value of predictive information markets, here are a couple of teasers:
1. On helping those in poverty http://www.aei-brookings.org/admin/authorpdfs/page.php?id=1311
2. On spurring scientific innovation http://www.tomwbell.com/writings/SPEx.pdf
From a collection of papers published by the AEI-Brookings Joint Center of Policy Markets http://www.aei-brookings.org/pages/index.php?id=37
— caveatBettorAn important issue in contract design is making sure that the right arbiter is selected. In this case, I’m curious if the Department of Defense would ever acknowledge a missile test. Does it have a policy about acknowledgements?
This comes to mind because I have been asked a number of times to provide information on inflation adjustment clauses that were written very poorly; the clauses would refer to versions of the CPI that do not exist, have never existed, and probably will never exist (such as “the CPI for Oregon”).
— Bill ConerlyThe call wasn’t as tricky as you suggest. Tradesports had ample evidence from sources it deemed suitable — the DOD and the White House — to expire the contract at 100 within the contract rules, so no “spirit of the contract” argument is necessary. Furthermore, Tradesports was in possession of emails from the top DOD spokesman which also confirmed the events. A more in-depth discussion can be read here:
http://www.poorandstupid.com/2006_08_06_chronArchive.asp#115509785796914301
— Trader ABill, I think TS used DoD to confirm Saddam capture and assumed they would be a stable source in the future. Oops.
As for your comment pkim, the contract is worded such that the default condition is “no missle.” And the burden of proof was on the long side.
This is a tough call for TS. Caught between honoring the strict wording of their contracts and using a “spirit of the rules” argument like in the pine tar incident. Strict wording won out.
Regarding incentives: TS has only 1 incentive, that is to maximize volume in the long run. This missile contract will liekly mean TS will just avoid offering current event contracts that may have an ambiguous result. That way they won’t piss of 50% of their customers.
— nphebelnphebel, as I just wrote above, there was a plethora of evidence to expire the contract at 100 within the strict wording of the contract, so it shouldn’t have been a tough call.
Also, the wording of the contract stated that, in the event of missile launches (no one disputes that there were seven launches) the contract would expire at 0 if “all launches were confined to North Korean air space.” So the burden of proof should be the same in proving that all seven missiles remained in NK air space, something Tradesports obviously can’t prove.
Tradesports’ incentive should be to repair its reputation and maintain market integrity.
— Trader AYeah, what Trader A said!
— caveatBettorSorry Trader A, your post came while I was writing and didn’t see it. I’m going to leave the debate of “DoD did or did not confirm” to the TS Forum, caveat’s blog, and god knows where else. The only point I’m making here is that there’s no real right answer and any result (short of Levitt’s precedent-busting free lunch option) would have alienated many traders. This means that predictive markets have limits in that expiration conditions can be murky. If this much uproar occured over $10K, imagine $10B! Use as a institutional hedging tool will only occur in events with irrefutable conclusions. (As a side note, if big $ does ever come in, binary contracts will drag with it the non-linear incentives for manipulation.)
By the way, I personally thought they would unwind the whole contract… I bought 80’s and sold 35’s on this so I feel your pain.
— nphebelnphebel, I’m not sure why you would say there’s no right answer, as the event the contract was about — missile launches that left NK air space– did happen no one can reasonably dispute it. Here are three relavant quotes from Bryan Whitman, the Deputy Director for Defense Information in the Office of the Assistant Secretary of Defense for Public Affairs:
“they did NOT fall in territorial waters.”
–July 11
“[they] landed outside the territorial of N. Korea in the Sea of Japan.”
–July 14
“the launches did go beyond the territorial water of N Korea.”
–July 30
TradeSports had all of these communications well before July 31st.
— Trader A> As for your comment pkim, the contract is worded such that the default condition is “no missle.” And the burden of proof was on the long side.
TraderA, I have to disagree. The two options were fired out of NK air space or not (not fired or fired and stayed within NK). Like any two outcome race/event/competition, there has to be a judge to determine valid outcome (we will ignore post-determination sub-rulings such as TourDeFrance or other disqual mechanisms). So, this means that you rely on an arbiter to determine fired-outside vs. not-fired-outside. The possible outcomes are fired-outside, not-fired-outside, not-possible-to-determine (tie). The latter would go to the not (unless specified otherwise). But, if the arbiter/judge cannot or will not rule, then your event never happened. Now, I agree that TS is within their rights technically, but I am suggesting that book-making on a non-occurring event (or never judged) is a kind of fraud and harms the industry as a whole. The bettors are not in a position to know if the arbiter/judge will be a no-show or the event is called on account of rain or whatever. So, the natural response is to treat it like a cancelled event and refund the bets.
— pkimelmaI play in a similar, but not for money, market called NewsFutures.com, and very similar issues sometimes come up there as well. In fact, we had a contract on essentially the same issue, and it was closed shortly after the media started reporting the missile launches. I had actually argued that it was closed too early since there had been no confirmation from any non-media source (namely, some government, any government).
— TaedI wonder if the Freakonomist helped foil today’s attack. Any bets on that?
— sophistryMr. Conerly, it would be especially difficult to even create a CPI for Oregon due to the basket of goods.
Consumer staples in Oregon include hydroponic marijuana, Birkenstocks adorned with miniature “Nader in 2000″ stickers, artisan patchouli, Toyota Prii, organic tofu smoothies, bead jewelry sold by hipster “homeless” teens, and homemade dye-free soaps.
While one could gather reliable data regarding the Prii and Birkenstocks, the other items would be problematic.
If you came up with a reliable method, however, you could likely sell the data to Vermont too.
— anahitPersonally, I’m glad the DoD has better things to do than to provide verification for a gambling site.
(Not that they are actually doing all the things I’d like them to be doing / not doing.)
— zbicyclist