It’s Only A Game

Prediction markets of all kinds pop up all over the world. As betting animals, we human beings get a kick out of knowing better. The illusion of having known things before others did is even more seductive. Yes, in retrospect, we are all right. But contracts that yield payments based on figuring out outcomes of an event —How will a new product do? Will a new movie be a hit? Who will win the election? When will the “big one” hit California?— are not about perceptions after the fact. Their object is the future. For me, they are anticipations, realizations in the space of possibilities.

Informed by aggregated knowledge (or by an illusion of knowledge expressed as a more-or-less informed guess), prediction markets, unlike a crystal ball readings, were adopted by many organizations. Some of them were simply desperate —here the US Defense Department qualifies for the trophy. Others, such IBM or Hewlett Packard, rich enough to try yet some other innovation —a can’t-lose proposition. Think about Google —how many will adopt Chrome, a browser that compares, in their views, only to sliced bread? But, surprise-surprise, at Google trading became yet another program, a trading robot that made its inventor rich in Gooblers (the currency for betting at Google). Do you feel like competing with such a program? Keep in mind that in the stock market performance over the last few weeks loss in value always came in swings. For people in the prediction market, this afforded higher profits than a steadily climbing market would have. You don’t need to short (especially when it’s not looked upon very kindly) to make money. Algorithms are good at this nerve-racking game —make money on melting markets, while everyone else loses their shirts (savings for retirement, for tougher days, for college education, etc.). Actually, on Intrade during the short McCain upswing (or was it Obama’s downswing?—depends how you look at it) more money could have been made than on the final outcome.

The low error rate in prediction markets made them popular not only among people dealing in Gooblers, but also in dollars or euros. More important is whatever we derive from the dynamics of aggregation. In the first place, information that can be used proactively (to minimize the impact of certain events, or to prepare for new opportunities). This is where things get really interesting. What does it take to make the “dangerous little knowledge” dispersed among many to aggregate? Ego will not do. In the aggregate prediction market we are only a nameless bit. The answers to the question of what does it take to become part of the new lotteries so far converge towards incentives. You can read “greed” here instead, if you don’t faint at the thought that making money is an instinct that unfailingly overrides haughty ethical or religious pronouncements (never mind political demagoguery in a day and age of massive corruption).

Having said this, I remember asking Cass Sunstein why he ignored prediction markets as games in his successful Infotopia. If I had advised either McCain or Obama —no, not my desire to advise candidates— my medium of choice for their success would have been games. Yes, prediction markets can make some people —actually, very few— richer. But embodied in games, driven by real-life data, prediction markets can engage many more people in having fun, while simultaneously giving up, in the aggregation of competition, the tidbits of information they somehow, knowingly or not, own. Playing games is not gambling. Therefore, the regulatory gauntlet that keeps Intrade in Ireland could disappear. Provided that Intrade reconstitutes itself into an MMPORP (Massively Multiplayer Online Role-Paying game, for those who don’t recognize the abbreviation). How would Intrade resist the pressure of sharing the game playing data to some homeland security outlet is a question above my payrate (if you allow me the innocuous quote).

Easier said than done. But I, for one, see the prediction markets make the transition (sooner or later). Anyone want to bet on this?

Mihai Nadin (University of Texas, Dallas)

No word on whether this will concern the X Universes. [I'm short of time, today. I will have more thoughts on the X Universes in the coming months.]

#1. BetFair‘s Mystery Bet is such an attempt to morph betting into gambling. They are working on the second version of this product.

#2. If the casinos and bookmakers make money proposing imaginary horse races to bet on, you can be sure that the prediction exchanges (betting exchanges) will propose the same kind of product, one day. It’s just a question of time. I don’t see any at BetFair Games (run from Malta, E.U.). You can bet on it.

Robin believes that betting markets should be used to rule many human affairs, including government policy. We should bet [*] on which policies will maximize national income, and governments should institute the policies what the betting markets show as most likely to succeed. [...]

[*] Speculate… not just “bet”.

—

As I wrote many times on Midas Oracle, whether Robin Hanson’s concept of decision markets (where conditional prediction markets are used to make decisions) will end up as applications is only half of the story. (Tyler Cowen is skeptical, just as I am.) The other half of the story is that Robin Hanson’s work is also about decision-aid markets (where conditional prediction markets are used to simulate the near future). In my view, it’s only a matter of years before Fortune-500 firms and think tanks pay close attention to this sub concept. They are just discovering the classic prediction markets, these days. Give them some time.

Robin Hanson is an inventor, here. As I wrote many times on Midas Oracle, many times, an innovator comes up, captures the essence of the invention, and applies it in a way that was not envisioned by the inventor. Example: Could be that the best application of the decision markets would be, not in the world of reality (as Robin Hanson hopes), but in the world of imagination. Let’s say that the Libertarians (capital L) creates a libertarian government within a virtual game like Second Life. They could adopt the concept of decision markets to run their public, virtual government. Other ideas are possible; this is just one example, to make the point that the innovators should think outside of the box… framed by the inventors like Robin Hanson.

—

Addendum: Robin Hanson’s response to Tyler Cowen. + Tyler Cowen’s response to Robin Hanson + Tyler Cowen on Robin Hanson + Robin Hanson’s answer to Tyler Cowen

- Generate revenues and profits for Midas Oracle. 4 / 23
- Make Jason Ruspini quit whining about the fact that he blogs on Midas Oracle for free. 1 / 23
- Incite the free-market thinkers into *practicing* prediction markets. 4 / 23
- DIY prediction markets 2 / 23
- The “X Groups” (the relationships between media/blogs and the prediction markets) 1 / 23
- No Response Recorded 11 / 23

—

CHRIS MASSE’S “REMARKS” (as Bo Cowgill says)

#1. Surveying the Midas Oracle readers is a cheap way to induce some kind of participation. I’ll do that more often. (I also said that I should do more e-mail interviews. I will do. I have a long to-do list, please be patient.)

#2. The survey above was half serious and half tongue-in-cheek (as that wet blanket of Bo Cowgill quickly “remarked”). I will do more on the same topic, or on related topics. My readers are invited to submit survey suggestions, too.

#3. “No Response Recorded.”: A high number of respondents either encountered a problem (maybe they answered the survey within their feed reader, and the responses were messed up, dunno) or they wanted to vote for “Dunno, Have No Answer, Your Survey Stinks”. (I should have minded a “No Opinion” answer option, maybe.) That’s OK. That’s internet democracy at works.

#4. “Jason Ruspini whining”: Actually, he “remarked” that finance experts who blog on prediction markets teach for free the corporate users of this forecasting tool. Jason Ruspini‘s point, if I understood well, is that organizations could write up their own software for prediction markets and avoid paying any external consultants. I think I disagree. Every Bo Cowgill needs a Hal Varian.

#5. “Imagination-based prediction markets (the ‘X Universes’)”: NOBODY VOTED FOR THIS. Well, it may be not a “priority” (important + urgent), but it’s important, long term. Look at the industry right now. 100% of the prediction markets are reality-based prediction markets, that is, whose expiry are based on news events (winner of a game, an election, etc.). The prediction exchanges (betting exchanges) are missing the other half of the market: event derivatives whose expiry is based on… fiction events (whether Harry Potter will die at the end, whether the Star Wars kid will fall to the dark side of the Force, etc.). You will tell me that, since the outcome is known by the universe creator, secrecy is needed. Yeah, that’s what the “entertainment industry” is based on, see: make-believe stuff + plot outcome secrecy. What we need for the X Universes to take off are prediction exchanges and universe creators working in symbiosis. More on this in the coming weeks, months, years and decades on Midas Oracle. Stay tuned. (Psstt… Don’t expect our good friends the scholars to foresee the future of the prediction market industry during their little surprise party at San Diego, next June. They are only interested in the past.)

#6. “DIY prediction markets“: I was first bearish on this, and now I’m bullish. If you think of it, the idea is to ease (via web services) the outsourcing of the creation and management of some event derivatives to either the Joe Six-Packs or the media/blog organizations.

#7. “The ‘X Groups’ (the relationship between media/blogs and the prediction markets)“: I’m not really convinced by the associations I’ve seen so far. But that’s not the point. If many more people propose event derivatives, then of course they will market their own event derivatives, and thus popularize the whole field of prediction markets. The long-term hope is that we see many more people trading socially relevant event derivatives. That’s the marketing part of the X Groups. But there’s a much, much more interesting aspect. Imagine a group of bloggers managing together an event derivative whose expiry will be based on… their collective judgment on an issue (e.g., their opinion on whether George W. Bush was true to the conservative movement). During the course of this prediction market, the traders reading the different blogs will be able to see how the bloggers are going to vote at the end, and whether their early opinions vary. See, DIY + X Groups = a brand-new kind of prediction market, whose expiry can be based on the opinions of the sub-exchange managers, as opposed to external facts. (Well, the aggregation of all the bloggers’ opinions becomes a de facto fact. ) One idea among thousands.

#8. “Incite the free-market thinkers into *practicing* prediction markets.“: The basic idea is simple. There are dozens of thousands of free-market, free-trade thinkers, out there, in institutions like universities, think tanks, consulting firms, and media/blog organizations. Here’s what to tell them: “You, guys/gals, will convey your free-market message better if you get your people into practicing socially relevant prediction markets, because that’s free markets at works.” Getting many free-market think tanks on this would solve two problems: number one, the socially relevant prediction markets are thinly traded (the think tanks would bring in traders), and, number two, the socially relevant prediction markets are unprofitable (the think tanks would subsidy the co-organization of these markets). Easy to say, difficult to do, of course.

#9. “Generate revenues and profits for Midas Oracle.“: I just found out about the Rolex Awards. Interesting. Any other prizes you can think of? Any other sources of cash you would think of, folks?

—

For your information, I’m in the process of installing Word Press on Midas Oracle .NET and Midas Oracle .COM. I don’t know exactly what I’ll do. I have one idea: setting up a constantly updated wiki-like resource on prediction markets. See, WordPress is actually a “content management system” and, besides “blog posts”, you can organize your content as “pages”. Does this fly in your book, folks? Another idea would be to set up a comparator of market-generated probabilities. If you have other crazy ideas, feel free to submit them. (They can’t be crazier than the ideas I have developed just above. )

Previous: Prediction Markets Timeline + DIY prediction markets — X Universes — X Groups + How prediction exchanges can best encourage participation

—

This NewsFutures header gives me the opportunity to renew my bold prediction: Will come a day, in the future, when imagination-based prediction markets will bring in more revenues than sports prediction markets.

Mark my word. And bookmark this blog post. Reality check in the next decade, or the following.

Now that the Neteller founders are being prosecuted on money laundering in conjunction with racketeering laws, the existence of The LindeX Currency Exchange will become increasingly controversial. (By the way, here is a casino property that was auctioned on EBay.)

When I first mentioned ways to gamble with real money in virtual worlds in July, reaction was muted. To this day at Terra Nova, there is almost no discussion on this issue although its authors pride themselves on inhabiting the real/virtual economic milieu. Nonetheless, more people are starting to ask questions.

One moral of this story is that we should eschew legal “hacks”. I hope to have something more constructive to say in a couple of days.

Today, I learn that Second Life goes GPL. The leftist Boing Boing guy writes: “This is HUGE.” —and lists like 100 related links (ouch!). Indeed, man. Here’s from the Secong Life blog:

A lot of the Second Life development work currently in progress is focused on building the Second Life Grid — a vision of a globally interconnected grid with clients and servers published and managed by different groups.

Would a prediction exchange have an interest in building such a “grid”. Would a TradeSports or BetFair Grid be useful? Hummm… There’s a strong centralization aspect in the business model of a prediction exchange. More than that, a successful real-money prediction exchange evolves towards a de facto monopoly —think BetFair (95% of the British betting exchange market, probably).

Could we have a private variation of a public prediction market on a private, corporate server? For instance, let’s say we would have a public prediction market on the business fate of Apple Computer (at TradeSports or BetFair) linked in some kind of way with an internal prediction market run on Apple Computer’s server (using an open-source software for prediction markets from the prediction exchange in question). No idea whether it would make sense. To render the problematic more complex, you could envision imaginary prediction markets… Would a prediction exchange grid come handy, then? No idea.

I’m probably thinking too much.

Addendum: The reason I file this in the “marketing” category is that I don’t buy the official rationale. I believe going GPL is a marketing tactic to expand, but maybe I dead wrong.

My Question: Do you see any kind of connection between natural language search and prediction markets?

Addendum (November 7): In a 2005 blog post, comes the explanation on how the prediction market virus disseminated into Barney Pell’s body and mind. One usual suspect (pun intended, see below) is involved:

I have been interested in this idea ever Robin Hanson (the founder of the field of prediction markets) and I were summer students in the AI Lab at NASA Ames Research Center, both working under Peter Cheeseman. This was right around the time when Robin wrote his first papers on the topic: “Idea Futures: An idea whose time has come?”, and “Could gambling save science?”. In order to match the work its NASA funding, Robin created a demonstration of how a set of Mars Rovers could place bets on the presence of scientific interest at different locations on Mars. The result would be the emergence of consensus beliefs that could be more accurate than the knowledge of any individual.

I also remember the time Robin hosted [a] Murder Mystery evening with a prediction market as an added twist. Like a normal such party, actors would read out each scene. But in Robin’s version, the audience members would then place bets on the identity of the murderer by buying options. Based on the market dynamics, a ticket marked “This ticket is worth $1 if Professor Plum was the murderer” might start out having equal value as the other suspect tickets (so that you could be the whole set of them for $1), but then the price would fluctuate as [events] unfolded until the prices ultimately went to $0 for all but the actual murderer. It was really fun, an improvement on the original version.