*Prediction Market Technology – Experience March Madness Like Never Before*

March 16th: Anticipation is running high for the 2009 NCAA Division I Men’s
Basketball Championship (popularly known as March Madness) with the first
round tipping off on March 19th.

On the web users can watch all the live action for free on CBSSports.com.
Additionally, there are no shortage of sites offering brackets on which you
can predict who will win all 63 games in the tournament.

Hubdub however has launched a prediction competition with a difference.
Unlike a bracket, Hubdub players can trade predictions through out the
tournament and in the later stages during the actual games themselves.
Additionally, not only are all the usual match up questions but players will
also receive daily questions like “Will any team hit a 3 pointer with 5
seconds or less to win?” and “Will President Obama attend any March Madness
game?”.

The Hubdub Predict-a-thon is free to enter and is offering $500 in cash
prizes to its top players. On top of that, if you introduce a friend to the
competition and they win, then you receive the same amount.

-

BetFair:

Improvements to Betfair’s bet matching logic today, Monday 14th July:

What’s changing?
We’ve improved the code that matches bets. As well as matching backs against lays as we’ve always done, we’ll also try to match your bet against bets on other selections in the market. We‘ll give you an improvement over the price you‘ve requested where possible, and we‘ll match you against whichever bets get you the best price.

For example in a tennis market:

Roger Federer is 1.9 to back, 2.1 to lay.
Rafael Nadal is 1.8 to back, 2.0 to lay.

If you try to back Federer at 1.9 or less, previously we would have matched your bet against the customer looking to lay Federer at 1.9. Both bets would have been matched at 1.9, even if you‘d asked for a shorter price. In theory we could do even better than that though: we could match you against the customer trying to back Nadal at 2.0 (backing one player at 2.0 is of course the same as laying the other player at 2.0). Our new bet matching process will see which match gets you the better price. In this case we would get you 2.0 by matching you against the Nadal backer (who is offering a better price than the layer of Federer).

When placing a new bet you will only ever be matched by the new process if doing so gives you a better price than you would otherwise have got. We will match your bet at the best price possible that’s a valid increment on Betfair’s odds ladder, as we explained in our update of 6th June.

Does this only work for 2-runner markets like tennis?
No. The new matching logic works for any number of runners in a market. An example with a 2-runner market is probably easiest to understand, but the principle is the same for markets with 3 runners or more. For example if a football market looked like this:

Spain 2.3 to back, 2.5 to lay
Germany 3.9 to back, 4.0 to lay
The Draw 2.9 to back, 3.0 to lay

Then if you want to back Spain we could match you with customers looking to back Germany and the Draw at 4.0 and 3.0 respectively, which would result in you being matched at 2.4, a better price than you would have got had we matched you against Spain layers (who are only offering 2.3).

Which markets will this affect?
We’ll introduce the new code on Monday 14th July, but initially matching will be done exactly as before. As explained earlier in the year, introducing best execution across selections wasn’t possible without significant change to the existing code that matches backs and lays, so we will need verify that performance is as expected for the existing matching process before enabling the new functionality. All being well we’ll enable the new code for a small number of markets to ensure that everything is as it should be later on Monday. We’ll announce which markets on Monday. Again if all is well we’ll roll out to a wider range of markets on Tuesday.

We’d expect to match across selections on the same range of markets as we currently do:

Match Odds in Basketball, Boxing, Cricket, Ice Hockey, Rugby League, Rugby Union, Snooker, Tennis and Volleyball, Greyhounds win markets, Darts match odds, correct score and handicaps and Soccer match odds, HT/FT, correct score and unders/overs.

Horse racing will not be covered for now, due to the possibility of non-runners, and the new process isn’t applicable to markets where runners can be added (for example “Next manager” markets), where runners listed might not take part (e.g. First Goalscorer) or where the runners in a particular “market” are treated independently (e.g. Accumulators).

What about bets placed in error?
We’re aware of a concern that this change might make it more likely that customers would match bets placed in error, for example asking for 1.2 when you really wanted to back at 2.2. One consequence of the change we’re making is that any bet you place is more likely to get matched – making it easier to get a match is the whole idea. Being realistic though, if you had placed a bet in error like that in the past, in the vast majority of cases you would have been matched (against lays on that selection). There’ll now be far, far more circumstances where you would have been matched anyway , but instead you’ll now get a better price, than situations where your bet would have been unmatched and you might have had the chance to cancel. On average we would expect customers who place bets in error to be better off as a result.

On a related point, we’d also expect this change to make it more difficult for people who place “trap bets” to get matched (a trap bet is an offer that is only likely to be matched if another customer places a bet in error). While putting up “trap bets” is against Betfair’s terms and conditions and we close the accounts of persistent offenders, on an exchange where any customer can ask for any price it’s difficult to eradicate this practice. In most instances where a trap bet is the best price available on a selection, customers will in future be matched at better prices against bets on other selections rather than matching the trap bet.

How will the change affect liquidity?
We would expect the change to be beneficial to liquidity. Obviously if we have opposing customer bets in the system that could be matched, whether on the same selection or across different selections, the best thing for liquidity is to match them.

-

Further to the above, we’ll be enabling the improved matching on the following markets later today.

Football:

Czech Republic U19 vs. England U19
FC Inter vs. MyPa

Tennis:
Andujar vs. Hanescu
Minar vs. Rochus

Greyhounds:
11:28 Sheffield
11:48 Oxford

If the biggest US-oriented prediction exchange doesn’t get that simple thing, then why would the journalists and bloggers get it?

John Delaney is skilled at sending e-mail insults, but as a web publisher, he is a zero.

The executives at the helm of the big prediction exchanges are imbéciles who can’t think straight about the future. There’s nothing more important than to put dynamic charts under the very nose of people interested in baseball, football, soccer, basketball, politics, movies, etc.

I can’t see something more important than that.

-

… asks ABC News (4 pages):

[...] NewsFutures creator Emile Servan-Schreiber believes that prediction markets like his are the “future of journalism.” [*] “It gives people the news of today and lets them give you the news of tomorrow in a probabilistic fashion by asking them to take bets on what is going to happen,” Servan-Schreiber said. [...]

While the money earned on NewsFutures isn’t real, Servan-Schreiber argues that prediction markets are an incredibly valuable tool. “I think of them as a brain,” he said. “You take a bunch of individually stupid neurons and put them together. Through massive interaction emerges intelligence. It’s the same thing with the market.” “The market really is smarter than the average player in there, and sometimes smarter than anyone in there, even the best player,” he said. This aggregation of many people’s opinions, Servan-Schreiber argued, produces an “incredible probability that the market prices really corresponds to the probability [of the event occurring].” [...]

“If you know nothing, you don’t participate,” he said. “It’s self-selection all the way. If you think you know something, and you don’t, you get eliminated very quickly.” [...]

-

[*] Prediction markets are one of the many futures of journalism, I would say. Like Emile, I’m bullish on prediction market journalism. But, as I implied in my previous post (where I cited extensively the meager but smiling Mike Linksvayer), PMJ is not for everyone —it should be the response to specific needs.

An other little remark. If I agree that prediction market journalism has a future, I would say that it won’t pass by the so-called “mainstream media”. Indeed, look at their first page, where they cited prediction markets and their prices/probabilities: ABC News didn’t link to those prediction markets.

-

Here’s my own short selection of NewsFutures prediction markets…

-

Some delegates from FL or MI will be seated at the Democratic Convention.

© NewsFutures

-

CERN will find the Higgs particle first.

© NewsFutures

-

Microsoft will buy Yahoo! in 2008.

© NewsFutures

-

Plenty of sports prediction markets here: baseball, basketball, hockey, auto-racing, American football, golf, tennis and soccer.

-

-

They would do anything to sell politics, these days.

-

Foreign Policy:

It’s a clever marketing ploy by team McCain, but why stop there? Why not have the candidate take positions in predictions markets such as Intrade? Wouldn’t we rather know how prescient Senator McCain is about the odds of bird flu striking the United States by the end of September, the chances Pervez Musharraf will step down as Pakistan’s president anytime soon, or the likelihood of a peace treaty between Israel and the Palestinians by January 2009? Put your money where your mouth is, senator.

InTrade-TradeSports and BetFair-TradeFair should let traders opt to make public their positions, and they should create dynamic prediction registries and leagues.

Inkling’s markets come in a couple of flavors, so far as I know all using an automated market maker based on a logarithmic market scoring rule (LMSR). In the multi-outcome case – for example, a market to pick the winner of a 65-team single elimination tournament – the market ensures that all prices sum to exactly 100. If a purchase of team A shares causes its share price to increase by 5, then the prices of all 64 other team shares will decrease by a total of 5.

The logic of the LMSR doesn’t tell you exactly how to redistribute the counter-balancing price decreases. In Inkling’s case they appear to redistribute the counter-balancing price movements in proportion to each team’s previous share price (so, for example, a team with an initial price of 10 would decrease twice as much as a team with a previous price of 5). While for generic multi-outcome prediction markets this approach seems reasonable, it doesn’t seem right for a tournament structure. (I raised this point in a comment posted here at Midas Oracle last September, and responses in that comment thread by David Pennock and Chris Hibbert were helpful.)

The problem arises for pricing tournament markets because the tournament structure imposes certain relationships between teams that the generic pricing rule ignores. Incorporating the structure into the price rule in principle seems like the way to go. Robin Hanson, in his original articles on the LMSR, suggests a Bayes net could be used in such cases. Now three scientists at Yahoo Research have shown this approach works.

In “Pricing Combinatorial Markets For Tournaments,” Yiling Chen, Sharad Goel and David Pennock demonstrate that the pricing problem involved in running a LMSR-based combinatorial market for tournaments is computationally tractable so long as the shares are defined in a particular manner. In the abstract the authors report, “This is the first example of a tractable market-maker driven combinatorial market.”

An introduction to the broader research effort at Yahoo describes the “Bracketology” project in a less technical manner:

Fantasy stock market games are all the rage with Internet users…. Though many types of exchanges abound, they all operate in a similar fashion.

For the most part, each bet is managed independently, even when the bets are logically related. For example, picking Duke to win the final game of the NCAA college basketball tournament in your online office pool will not change the odds of Duke winning any of its earlier round games, even though that pick implies that Duke will have had to win all of those games to get to the finals.

This approach struck the Yahoo! Research team of Yiling Chen, Sharad Goel, George Levchenko, David Pennock and Daniel Reeves as fundamentally flawed. In a research project called “Bracketology,” they set about to create a “combinatorial market” that spreads information appropriately across logically related bets.…

In a standard market design, there are only about 400 possible betting options for the 63-game [sic] NCAA basketball tournament. But in a combinatorial market, where many more combinations are possible, the number of potential combinations is billions of billions. “That’s why you’ll never see anyone get every game right,” says Goel.…

At its core, the Bracketology project is about using a combinatorial approach to aggregate opinions in a more efficient manner. “I view it as collaborative problem solving,” Goel explains. “This kind of market collects lots of opinions from lots of people who have lots of information sources, in order to accurately determine the perceived likelihood of an event.”

Now that they know they can manage a 65-team single elimination tournament, I wonder about more complicated tournament structures. For example, how about a prediction market asking which Major League Baseball teams will reach the playoffs? Eight teams total advance, three division leaders and a wild-card team from the National League and the same from the American League. The wild-card team is the team with the best overall record in the league excepting the three division winners.

In principle the MLB case seems doable, though it would be a lot more complicated that a mere 65-team tournament that has only billions of billions of possible outcomes.

[NOTE: A longer version of this post appeared at Knowledge Problem as “At the intersection of prediction markets and basketball tournaments.”]

“When people make these predictions, they have this little sliver of doubt – ‘What if I’m wrong?,’” [Stephen] Nowlis says. … The researchers dubbed this emotion “anticipated regret.”

A pair Arizona State University professors, Stephen Nowlis and Naomi Mandel, have research forthcoming in the Journal of Consumer Research in which they claim that predicting an outcome takes the fun out of watching the event.

My personal experience is pretty much the opposite. So while I find the research interesting, I don’t believe they have fully understood the phenomena. Of course, they’ve done the research and I am just reporting introspection and anecdote, so weigh my commentary appropriately.

I couldn’t find the actual research article online, but here are selected points from a story about the research appearing on the ASU Carey School of Business website:

[Nowlis and Mandel] found that when people make a prediction about an event – whether it’s the basketball games of March Madness, the singing showdown on “American Idol,” or the all-or-nothing Super Bowl – their enjoyment of that experience is inevitably lessened.

In other words, the researchers say, the very action that would seem to give viewers an actual stake in these experiences – that is, having something to win (or lose) – is also the very thing that makes the viewing experience uncomfortable and, even, unpleasant.

On its face, Nowlis admits the finding seems counterintuitive. And, sports betting is skyrocketing in popularity.…

In my experience – mostly small-time office pools, various on-line contests, and play-money prediction markets – making a prediction enhances my interest and enjoyment of an event. For example, I watched more American Idol, was more interested, and enjoyed it more because of a friendly bet among family members and positions taken on the Inkling “American Idol” market. Were it not for these outside interests, I would have stopped watching after the show finished with the regional qualifiers. I also cared more about the NCAA men’s basketball tournament because of office pools, ESPN’s online bracket contest, and, again, a couple of Inkling markets.

The article concludes with some speculation about why the prediction business is booming if people find it so unpleasant. It makes some sense to me that the immediate experience of watching the event may have been more stressful, perhaps because the predicting participants felt like they had something at stake, but I suspect that later the more involved participants remembered the experience more favorably. I would guess that the predicting participants would remember more details about the event a few days afterwards, too.

More from the article:

With the “prediction” business booming, Nowlis says, it would be natural to assume prediction-based marketing schemes would be a good thing for business, but the results of the study just don’t bear that out.

Well, based on the description in the article, it is probably safe to say that you shouldn’t corral a bunch of your customers, randomly required half of them to make predictions about events that they may not care about, give them too little information to predict with, and then make them all sit and watch a recorded event.

My inclination is to believe that the “prediction” business is booming for good reason, and the study doesn’t help explain why.

As mentioned previously at the Knowledge Problem, Inkling offers a public play-money prediction market. I stumbled across them a year or so ago, and because I’m interested in market design and prediction markets, I decided to try them out. Partly because playing the Inkling markets amuses me and partly because I started doing well, I’ve continue to play on Inkling. Eventually I wormed my way onto their Top Traders list, where I’ve remained for several months.

In the process of amassing my play-money riches, I’ve learned a few useful things about Inkling’s markets. Here I am “giving back” to the Inkling user community by sharing one of my secrets.

This is it: when a market offers you free, riskless profits, take them.

Obvious, right? The trick is in spotting the riskless profits. I recently was able to take some free, riskless profits when Inkling allowed two markets to be set up for the same event: the UEFA Champions League. It isn’t necessary for there to be two markets on the same event for arbitrage to work – I did something similar in the NCAA men’s basketball final four market – but the two market case makes the arbitrage process easier to understand.

By the way, as I write the markets are still live, and I have existing risky holdings in these markets. These stakes are standard bet-I-know-better-than-the-market plays which may or may not pay off. None of what I am about the explain involves taking risks, just taking profits.

The standard multiple result market at Inkling pays off at 100 units for each share of the winning outcome you hold at the closing of the market, and pays off at 0 for all other outcomes. In this case both markets will pay 100 in play-money units for picking the eventual winner in the 2006-07 UEFA Champions League, and 0 for shares in all other outcomes. Because two separate markets are running over the same event, when the prices of the markets diverge, there is a simple and obvious arbitrage opportunity. Of a pair of matching outcomes (Say “Chelsea wins”), buy the lower priced of the two and sell short the same number of shares in the higher priced market.

The final outcome doesn’t matter, because whatever you win on one you will lose on the other. Your entire profit is captured in the net proceeds of the buy low-sell high pair of transactions. Here is a simple example drawn from my trading report:

I sold into the first market at an average price of about $13.72 and bought in the second market at an average price of $10.91. The result: a $140.11 profit for a minute or two of effort. Not a lot of profit, but not too bad for riskless trading. (The short sale does tie up some credit, but since we’re dealing in a play-money exchange that doesn’t pay even play-money interest on your holdings of pretend cash, the only thing you sacrifice is other trading opportunities.)

Outside of the rare pair of auctions covering the same event, I’ve found similar riskless trading opportunities in the recent market to pick the final four teams in the NCAA men’s basketball tournament. The market would pay 100 for each of the four teams that reached the final four weekend. The market started even before the tournament seedings were announced, with a large number of possible teams, and shares traded on various market expectations. Curiously, early on the total value of all of the stocks summed to more than 400 even though owning one of each share was guaranteed to pay off exactly 400. (This is a clue that arbitrage opportunities are available.)

Once the tournament seedings were announced, it is clear that owning one of each share of all 16 teams in a single region would payoff exactly 100. If the share prices summed to less than 100, then buying one share of each team generates a payoff at the difference between the cost of the shares and 100. If the share prices for all teams in a region sum to more than 100, then selling one share each will generate profit to you at the difference between the proceeds of the sales and 100.

For simplicity, say that there are just two teams left in the “West Region.” Let’s call them, hypothetically, “UCLA” and “Kansas.” Buying one share of UCLA and one share of Kansas, will lock in a payoff of 100, so if the prices of the two teams sum to less than 100 then you can obtain riskless profits by making the purchase. (Or sell short if the prices sum to more than 100.)

While it may be less obvious, the underlying arbitrage is similar to the AC Milan example. At this stage of the contest, a share of “UCLA wins” is the logical equivalent of “Kansas loses.” Buying one share of UCLA and one share of Kansas is the logical equivalent of buying a stake in “Kansas loses” and a simultaneous stake in “Kansas wins” – with offsetting holdings your payoff doesn’t change based upon the outcome of the event, and your profit is risklessly captured from arbitraging the market at the time of the transactions.

Many Inkling markets are self-arbitraging in the sense that they automatically account for these interrelationships in pricing multiple outcome prediction markets. For example in the separate market to pick the eventual champion of the NCAA men’s basketball tournament, all prices automatically adjusted in response to any purchase or sale such that the sum of the prices always totaled exactly 100. (In fact, both of the two UEFA Champions League markets are self-abitraging within the markets, but I profitted by arbitraging between the two markets.

As an economist, my opinion is that such self-arbitraging markets likely exhibit superior efficiency properties that would make them desirable in real-money practice. (As a sometime experimentalist and aspiring prediction-market geek, I’d love to test that conjecture in an econ lab.) As an Inkling trader, however, I love to discover and exploit riskless trading opportunities in non-self-arbitraging markets.

[Okay, I’ll admit that I didn’t become an Inkling Top Trader via riskless arbitrage. I took many risky steps along the way, some of which paid off handsomely. But explaining that I often got lucky doesn’t appeal to my inner aspiring PM-geek.

BTW, in addition to pulling riskless profits out of the UEFA Champions League markets I was also carrying very substantial risky holdings in both Chelsea and Man U. Surely one or the other would win it all, right?

Ouch.]

Nate Kontny:

[...] Obviously, it’s no fun to have broken ribs. They really kick your ass. It’s also been real trying on my wife who has to do so much because I can’t, like driving or carrying the heavy groceries home. [...] My original plan was to go back to the gym on May 1, but it’s only been about 5 weeks since the injury. Also, after a brief workout yesterday of climbing stairs and a few minutes of shadow boxing, I woke up with some sore ribs. So it sucks, but I’m going to have to take at least another week or 2 off from the gym so I don’t just make this worse. That delays getting promoted to another belt by another month, which really sucks because I was just a couple days away over a month ago. Oh well such is life. People have mentioned switching hobbies to something less “injury prone”? All I know is I’ve broken 2 bones playing basketball and been in the ER for a possible concussion playing baseball. I’ve even managed to break a finger in acting class. Play golf you say? A few years ago, my wrist was feeling odd from playing lots of golf. The doctor? He tells me to lay off golf for a few months. Almost anything I do I can and do get hurt doing, especially since I hate doing things just part way. [...]

Chris [Masse] speculated on whether the simExchange will become a more generalist exchange, like BetFair, which began as a horseracing market before transforming into a general prediction market.

The simExchange is indeed exploring the possibility of expanding beyond traditional console and PC games, such as the immediately close tangent of downloaded games, such as Xbox Live Arcade and Wii Virtual Console games, and new online games, such as Second Life type games. However, our focus continues to be developing a strong prediction market in which the video game industry and gamers can rely upon to gauge video games.

BetFair was at inception a British horse racing betting exchange (real-money prediction exchange), and it as been expanding into other sports (soccer, cricket, basketball, football, rugby, tennis, etc.) and a little bit into politics and financials. On the other hand, the Hollywood Stock Exchange can be described as a betting exchange that hasn’t expanded much beyond movies and movie stars. (Previous: US presidential election prediction markets: HedgeStreet or Hollywood Stock Exchange??)

Of course, the difference between the Sim Exchange and BetFair is that the latter provides its service to play-money speculators… for free. How would the Sim Exchange finances the managing of non-game event derivatives? Let’s look at the Inkling’s business model for a partial answer. Inkling is unique in the prediction market scene in that it is an unmanaged prediction exchange: only the users (e.g., ABC7 San Francisco) do create, run and expire the prediction markets —the exchange managers don’t create and expire anything. In a way, Inkling resembles Second Life, whose CEO made the determination early on not to be a game creator —Second Life is in the business of selling/renting lands (servers), only, and does not create itself any content. Look at it this way: Inkling does not pay for event derivative management —it does outsource that to its users. Just like TypePad.com or WordPress.com do not pay for blog content creation —they outsource that to volunteers (the bloggers who live as a sub-domain). The two Inkling managers, instead, focus on creating their great on-demand software, and on servicing a fistful of corporate customers, which puts food on the table.

The Sim Exchange could look into the eccentric Inkling’s business model for inspiration, if it decided to move beyond video games. In a Web 2.0 world, user-generated content (here, event derivative) is King. One idea (among many others) would be to have Brian Shiau et al. creating and managing the prediction markets on video games, while letting qualified users designing and managing prediction markets on the topic they have a strong interest in. In the case of Midas Oracle (as a group), we would love to partner with one or more betting exchange(s) so as to do prediction markets on the prediction market industry, among other themes. (Previous: Midas Oracle Opportunities For Your Business)

Just one idea among many… Do the Midas Oracle readers have any suggestions to Brian Shiau?

Previous: An invitation to join the simExchange beta

Previous: Are Nathan Kontny and Adam Siegel of Inkling Markets being EXPLOITED by Paul Graham’s Y Combinator?