How Kim Kardashian Killed Candy Crush

When King Digital Entertainment, the company behind the massively successful mobile app Candy Crush, filed for an IPO in February this year, I knew somebody was going to come along and drink its milkshake. I just didn't think that somebody was going to be a Kardashian.

But that's exactly what's happened in the last few weeks. As Kim Kardashian: Hollywood-a game of dubious aesthetic value and indubitable appeal- has climbed the app charts and nibbled away at the Candy empire, King's revenue has slipped commensurately.

The stock is down a vertiginous 20 percent this afternoon, after a Deutsche Bank analyst laid the blame squarely at Kim's feet. 'Competition within casual gaming is intense, with Kim Kardashian: Hollywood and 2048 [another addictive game] going after the same demographics,' Business Insider's Linette Lopez quoted.

Six months ago, I listed five reasons why King Digital seemed like a terrible investment, but today we only need one: It's the Kim problem. Or, if you'd prefer to sound more sophisticated, a mobile-distribution problem.

Making hit mobile games could be the most random business in pop culture. There are hundreds of games based on baubles and grids, but every once in a while, one of them-Candy Crush, or 2048 -will fly up the App Store charts and steal an enormous share of downloads. Why? Nobody knows, really. It is your classic 'winner-take-all,' fat-head/long-tail distribution. Many people determine their next download by looking at the most-downloaded list on app stores. In these kind of markets, where the rich get richer and small initial advantages in popularity accumulate over time, two things are true: (1) a few games will get fabulously unequal returns, and (2) it's very hard to guess which ones.

There is an element of randomness to making hits in any entertainment industry. 'Nobody knows anything,' William Goldman famously said of Hollywood's hit-prediction powers. 'Not one person in the entire motion picture field knows for a certainty what's going to work.' But the movie industry is quite lucky compared to mobile games. When a superhero movie debuts, $40 million of commercials and billboards blanket the country, and then the film suddenly and simultaneously debuts in thousands of theaters. Big-studio films aren't as quite predictable the afternoon's weather, but they have a clear marketing pattern and distribution platform that greatly mitigates their risk.

Mobile games don't control their own distribution. That means means they don't control their own fate. They can buy ads on Facebook, and clamor for celebrity endorsements, and design sequel after sequel. But that's about it. Most people discover apps on most-downloaded lists, or because they become viral sensations that the media obsesses about (Candy Crush), sequels to those viral sensations (Candy Crush: Saga), or novelties ('Kim Kardashian came out with a mobile game??'). But since virality is typically unpredictable in markets without distribution hubs, mobile game developers are ping-pong balls bouncing inside a bingo machine. It's hardly an entertainment business. It's a entertainment lottery.

King Digital's central promise-we can make the viral magic happen, over and over, even though we don't control the networks that make our games actually go viral- is really strange and almost certainly wrong. Look at Zynga, Farmville, or Angry Birds. Mobile games are mayflies, as I've said, bursting into experience and flaming out in the blink of an eye. It's pretty random that Kim Kardashian is destroying King Digital's stock. But chasing randomness is the business that King Digital chose.