A bizarre service went live this past Monday: Media Predict, a self-described online game that uses a prediction market setup to rate media content (in prediction markets, people buy and sell shares based on the likelihood of a particular outcome--for instance, whether a certain candidate will be elected--with share prices reflecting the likelihood, or not, of that outcome). According to Media Predict's FAQ, "Although media companies sometimes find hits, historically most media products lose money or fail to find an audience. At the same time media companies often overlook valuable content...Media Predict uses prediction markets to identify media products worthy of investment and development." Why prediction markets? Because they have "an astonishing record of success in forecasting economic indicators, election results, sales levels and more."
Media Predict has markets for music, television, and--most relevant for Writer Beware readers and what I'll be focusing on in this post--books. Book proposals are posted by authors or their agents (subject to screening--Media Predict claims to work with "established" literary agents to vet proposals, though it doesn't say who the agents are). Traders are given $5,000 in fantasy cash to speculate on things like "Will 'Yoo-Hoo, Bunnie McFoo!' Get a Book Deal?". Trading starts at $10 per share. High share prices, theoretically, suggest future success.
Strip away all the folderol, and here's what it boils down to: in its book aspect, Media Predict is basically a manuscript display site with a twist.
What can you say about a setup where random players attempt to forecast success or failure for literary properties based on a brief four-point description (such as this one for Kate Shindle's Crown Chasers) and a no-more-than-10-page proposal/excerpt? Not much, in my opinion. For one thing, the available information is extremely limited--most notably by the fact that you can't actually read the book, and so must bet on success based almost entirely on external factors, such as the author's celebrity (Dorothy Hamill has a proposal for a children's book), or the poignancy of the author's circumstances (self-pubbed author Jason Cooper wrote his fantasy story for his kids), or just the fact that share prices are high, suggesting that other people think the project will succeed. Traders in other prediction markets--such as the University of Iowa's Iowa Electronic Markets, which often predict the results of presidential elections more accurately than traditional polls--surely have far more objective information at their disposal.
Also, the value of the predictions presumably stems from the fact that they arise from the book's potential market. But unless you have huge participation, how likely is it that results on Media Predict will have genuine market relevance? And even with a large group of speculators, how much correlation will there be between traders and potential readers for any given book? Betting that Dorothy Hamill will get a deal for Skatey Katey doesn't mean you'll run out and buy it for your kids (assuming you remember it when, after a year or two, it finally gets published). Will a strong Media Predict showing translate into actual book sales? There is only ever one answer to this kind of question: No one knows.
The law of averages alone suggests that Media Predict will flag some successes. But all in all, it's hard to see how this process will be any less like rune-casting than the more conventional decision-making methods of agents and editors.
At least one publisher seems to be willing to bet otherwise: Simon & Schuster, whose contract negotiation policy changes grabbed publishing industry headlines last week, and which is involved in another popular-pick-style contest, the First Chapters Writing Competition at Gather.com. Media Predict and S&S's Touchstone imprint are co-sponsoring Project Publish, a contest that will select a book proposal from the Media Predict website for (possible) future publication. This coming September, S&S editors and Touchstone publisher Mark Gompertz will choose five finalists from Media Predict's 50 top-scoring book proposals. "Additional supporting content" (an extra 20 pages) will be posted for the finalists, and trading will continue until October, when one grand prize winner will be announced. The prize: a S&S book contract with an "approximate minimum value" of $2,000. (Note to entrants: this is not quite the same as a commitment to pay an advance.)
So S&S is promising to offer a contract to a book of unknown subject and genre on the basis of fantasy stock trading and 30 pages of content. (What was that about casting runes?) According to Touchstone publisher Mark Gompertz, quoted in the New York Times, this is a reasonable method of book selection because "many, many more eyeballs will have looked this over, and that’s what’s intriguing about this." (This is interestingly similar to what he had to say about the First Chapters contest, which also selects finalists by popular vote: "You know that many more eyes have read the thing than if it had been seen by a single agent.")
All those eyeballs notwithstanding, Media Predict and S&S are hedging their bets a bit. According to Clause 5.c. of the Project Publish Official Contest Rules, the contest can be canceled if "an insufficient number of qualified Book Proposals are received," and also "in the event the Media Predict Web Site does not attract a sufficient number of users." (The meaning of "insufficient" and "sufficient" is not specified.) And according to Clause 8.a., S&S has "the right but not the obligation" to negotiate a book contract with the 5 finalists; and can, if it wishes, determine that none of the top 50 are publishable (Clause 8.d.). Clause 8.c. acknowledges the possibility that S&S and the grand prize winner won't be able to come to terms, and also allows for nonpublication if the winning book doesn't pass legal review.
But that's not all. Careful readers of the fine print will surely pause at Clause 7.a., which states: "If a selected author does not have a literary agent, Sponsor will contact such authors to arrange for representation." Does that mean Media Predict is claiming the right to act as your agent if you're an unagented contest entrant? This is certainly suggested by Clause 7.b., which begins: "If Sponsor is engaged as an agent for a Book Proposal..." and by similar language in Clause 8.c.: "If the winner does not have an agent other than Sponsor..." And according to an article in the May 21 edition of Publishers Lunch:
When unagented writers have their submitted work accepted for the contest, they need to sign an agreement--not posted, and which [Media Predict founder Brent] Stinski declined to provide--based on "a standard literary agreement" that he says makes Media Predict the "agent on a temporary basis." Stinksi says the company's intention is not to actually engage in literary representation itself, though. "We want everybody to have access to literary representation.... If you want to sign on, we'll put you on the site, and if you score well we'll find an agent for you."
He says the initial term of the representation agreement is four months, calling it "sort of a placeholder while the book is tested on the site." If the book doesn't score well, he plans to release rights on a shorter time-frame. If it does score well, "we'll look for a real agent"--starting with the site's agent partners, and the initial agreement can be extended for another 12 months.
This raises a lot of questions--such as why why on earth it should be necessary for Media Predict to act in an agent's role for unagented authors, especially since it won't "actually engage in literary representation." The obvious answer is that Media Predict stands to benefit financially, since presumably it will take a cut of any book deals that result. That's 15% to Media Predict, and a double commission for the author--since if she does get a book deal, she'll probably want to hire someone who will "actually engage in literary representation."
It will be interesting to see how this shakes out over the next few months.