« Fear of Failure | Main | The Accidental Investment Banker »
Latest Stories
- Bubbles are Good For You Because ... They Are
- Quote of the Day: Commodities Markets
- San Diego Homes Bottoming?
- The Myth of the Angel Investor
- Pete Peterson on Subprime, Bailouts, etc.
July 2, 2006
Information Cascades, Randomness, and the Movie Business
There is a wonderfully interesting piece in the L.A. Times today about the random core of the movie business. The article synthesizes useful material from information cascades, Kahneman & Tversky, Arthur De Vany's work, and Hollywood lore to show that most of what is called success in Hollywood is more like dumb luck.But if picking films is like randomly tossing darts, why do some people hit the bull's-eye more often than others? For the same reason that in a group of apes tossing darts, some apes will do better than others. The answer has nothing to do with skill. Even random events occur in clusters and streaks.How long until people come to same realization about the venture business? Pace my "no-one knows anything" comment about venture capital practices in an PE Week Wire column earlier this week, I'm increasingly convinced that most of what we call success in venture capital is a combination of a few very rare people with a good gut, and a bunch of others who got lucky once or twice and are now deemed to have had skill.
Imagine this game: We line up 20,000 moviegoers who, one by one, flip a coin. If the coin lands heads, they see "X-Men"; if the coin lands tails, it's "The Da Vinci Code." Since the coin has an equal chance of coming up either way, you might think that in this experimental box-office war each film should be in the lead about 10,000 times. But the mathematics of randomness says otherwise: The most probable number of lead changes is zero, and it is 88 times more probable that one of the two films will lead through all 20,000 customers than that each film leads 10,000 times.
Sphere It
| 
Digg it
| 
Bookmark it
| 
Stumble it
Paul, I read the PE Week article and thought your line of questioning and analysis was spot on. And you conclusion nails it !
"There are no best practices in venture capital. For my part I think that's the real insight, however nihilistic it might first seem. And it has implications, like that venture remains a craft, a gut business where a few rare practitioners have a good instinct for the stuff. But no combination of apprenticeship, or thumb-sucking spreadsheets, is going to turn you into Mike Moritz or John Doerr. You've either got it, or you don't".
It's very refreshing to see someone who tells it like it really is !
BR....KM4
PS - great blog with excellent subject matter. I'm a daily reader.
It's a powerful argument, but there is something amiss.
The hypothesis that box office success is random may be mostly true, but it is likely NOT true in some very important ways. Scripts, directors and actors all vary in quality, just as software ideas and management teams do. If box office success and venture success are truly random, then a movie with Brittany Spears as the lead should have the same shot as a Julia Roberts flick, and the iToaster should have been as compelling an idea as the iPod. That doesn't make intuitive sense.
The postdiction criticism is another example of something mostly true but not always. In some cases you really can look back and see a meaningful pattern. Kevin Costner, for example, is the Al Gore of actors. He has zero ability to play anyone other than himself, and he happens to be an incredibly dull person. Or if you look at Jennifer Aniston's lack of success vs Julia Roberts, you can see that Roberts has a lot more emotional range and nuances in personality, where Aniston has consistently played the frazzled it-girl who is put upon by life until everything works out in the end.
Point being that, while it's probably wise to recognize the level of randomness in movie success and venture success, to just throw everything into the randomness box seems a bit nihilistic. There are learning curves and fruitful actions to be taken--they just happen to be floating in a sea of complexity and serendipity.
One last interesting thought, if randomness is all-dominant then perhaps that raises the importance of execution, visibility and marketing. If everyone is throwing random dice then the long term winners will be the ones who can hype their projects the most effectively and get the most bang for their marketing buck on the hits. This would favor the name players in the same way that the liquidity effect favors the major exchanges (them that has gets).
Good point. It would need a bit more finessing to properly explain persistence (Franklin Stubbs addresses that, as does your "a few very rare people").
As to success in the movie business, I can't let that go without (yet again) pointing to Roger Corman.
I agree. If you were to give some group of entrepreneurs $10 million, they would, with very few exceptions, be as likely to fail or succeed no matter if one group was helmed by bigwigs and the other by novices.