The Science of Success.(The Talk of the Town)(media prediction)

Last month, the publisher Simon & Schuster announced a partnership with a Web site called MediaPredict, which would use the collective judgment of readers to evaluate book proposals. The deal drew scorn from many, who saw it as evidence that publishers, in an era of stagnant sales, had so lost confidence in their own judgment that they were reduced to the methods of "American Idol." Asking readers to weigh in on a book's commercial prospects was a recipe for mediocrity, and the experiment was "doomed to fail." Yet even the idea's critics recognized that it was a response to a real problem: most books today are not economically successful, which means that much of the time and money that publishers invest in projects is wasted.

For years, publishers have accepted this as unavoidable. After all, in nearly every media business most products flop, and most of the profits come from a small number of huge successes. In the music industry, only ten per cent or so of recordings actually turn a profit. On network television each season, most new shows fail. And in Hollywood, according to the economist Art De Vany, six per cent of films account for the bulk of the industry's profits. Media companies have learned, accordingly, to diversify their bets and, increasingly, to share their risk with others. (Hollywood studios, for instance, now often bring in outside investors on big projects, limiting their potential downside.) But the process of predicting whether a product will be a hit remains remarkably haphazard and erratic.

Many people argue that it's foolish to expect otherwise, and that no science of success is possible. In the famous words of the screenwriter William Goldman, "Nobody knows anything." The fate of a book or a movie, the argument goes, is determined by too many factors to be predictable--advertising, reviews, word of mouth, luck, and, in the case of big hits, a simple desire to see what all the fuss is about. De Vany, for instance, says that the box-office performance of Hollywood films is "chaotic" in the mathematical sense of the term. Three Columbia sociologists recently found something similar in a series of online experiments in which people were divided into eight groups, asked to listen to songs by unknown artists, rate them, and then decide which ones they'd like to download--after being told how often others had downloaded the songs. The highest-rated songs, it turned out, were not always the most frequently downloaded. And in each group a different song ended up topping the charts. In the laboratory, at least, success appeared to be essentially random.

MediaPredict, however, is wagering that in the real world success is, at least in part, predictable, and it follows a model that, over the past decade, has proved surprisingly effective in forecasting a wide range of events: the prediction market. Prediction markets function like futures markets, except that, instead of betting on the future performance of a company or a ...