Chew. Spit. Repeat.

But Morita's underlings, showing a loyalty unknown in Hollywood itself, didn't want to see their elderly patriarch disappointed, and the next day the group reconvened and decided to buy the studio. They ultimately had to pay an exorbitant $3.2 billion and assume an additional $1.6 billion in debt. Sony was further seduced into hiring two Hollywood insiders, Peter Guber and Jon Peters, to run the place. After a string of flops in the early 1990s, it cost Sony $250 million in production deals and money to ease them out. (Sony persevered, and in recent years, with megahits such as "Spider-Man," which brought in an estimated $500-million profit, the studio has been considerably more successful.)

"They don't do the same sort of traditional business analysis that they would if they were entering, say, the machine-tool business," says Dartmouth College business school professor Sydney Finkelstein, author of the book "Why Smart Executives Fail." "Then again, when you're making machine tools, you're not seduced by the idea of sitting in the audience with a bunch of movie stars at the Academy Awards. Instead, you get seduced by the glamour and it screws you up"--even someone as astute in business as Morita.

Not that conventional business analysis would do much good. Arthur De Vany, a former UC Irvine professor, industry consultant and author of the book "Hollywood Economics," points out that outsider-newcomers have three times the failure rate as veterans in any business, so it's not a total shock that they usually bomb in the movie biz. From an empirical point of view, though, an outsider would have to be totally insane to try the movie industry in particular, because the economic model is bizarrely different from just about any other business. The movie industry actually earns only a 3% to 4% return on investment, which is lousy when compared with steel-making or book publishing.

To make it worse, the statistical curve for movie profits isn't much of a curve at all. If the movie industry followed a bell curve, the typical movie would make money, and it would be extremely rare for a movie to take in more than three times the standard deviation--the average amount that the films differ from the middle. Instead, it's shaped more like a playground slide --6% of the product earns 90% of the money, and 70% to 80% of the product sinks into oblivion. This results in what economist De Vany calls an industry of "extreme uncertainty." That is, successes are aberrantly rare and outlandishly enormous. The movie "Titanic," for example, grossed $600 million domestically in 1997, in a year when the average film grossed $23 million. Results like that are impossible to predict.