19. Moneyball, Zynga, and Other Unconventional Applications of Economics, 2012

What actually happened in Moneyball is analysed, and applied simply to game development.

The movie (and book) Moneyball centers on a real life person named Billy Beane who changed baseball by introducing new methods of applied analytics into a relatively ancient and archaic industry. In essence, by using analytical methods, he was able to do more with less, leading to amazing profits. The movie is absolutely a must see if you are in any position higher than a programmer in the IM industry.

As portrayed in Moneyball, Billy Beane was not extraordinarily gifted in the mathematical arts. The character Pete, who is extremely undervalued at the start of the movie, is. Beane is extraordinary in that he is able to identify immediately how undervalued Pete is, and hires him instead of the baseball player he originally had sought to hire. This took tremendous courage and insight, as this required the pair to challenge the pillars of convention in America’s most cherished sport. Beane’s detractors, including possibly his own family, are already digging his proverbial grave from the very start. Many openly question whether the man has lost his mind.

Beane goes on to be vindicated and at the end of the movie gets offered an eye-popping amount of money ($12.5M) to bring his techniques to a competing team. Those techniques were actually acquired from Pete, the pudgy 24 year old fresh economics grad, who was influenced by the writings of Bill James, a baseball journalist with a background in economics.

Now let’s jump to the interactive media industry, where a man named Mark Pincus co-founded a little company called Zynga a few years ago. Pincus has his undergraduate education in economics (see a theme here?) and an MBA from Harvard. Pincus had this idea that he could use analytical methods to generate more profitable games. His first year results were underwhelming but after four years his company is steamrolling most competition and our entire industry is scrambling to adapt.

The similarities between Moneyball and Zynga are not coincidence. Analytical methods, when applied appropriately, do absolutely increase profitability. In both cases, when these methods are first introduced there is an initial period of awkwardness, followed by huge success, and then a levelling off as the competition adapts by employing similar methods.

There is a big difference, however, between Pincus and James. James was a fanatic about baseball that made the application of analytics to baseball his life, not knowing how it would turn out. Pincus had a hunch that something similar would work for games, but he was not possessed of the historical, journalistic, or creative background that James had. This meant that Pincus had to identify the creative works of others to modify and improve. He did not really understand the mechanics of games so he had to start with very simplistic products.

Pincus was the right person at the right time to do what he did, but his accomplishments were, to some extent, a one trick pony. He was the first person in our industry to take A, B, and C and put them together to make ABC. This really is not hard so now a lot of people are making ABC so ABC is not doing so well and will do even worse as competition increases. He didn’t invent A, or B, or C. This makes him different from Bill James, who really was way ahead of his time. James understood A, B, and C, so he was free to rearrange the elements to find the best fit, or maybe even change one of the elements. Pincus is more like Beane in the Moneyball scenario, so his options for innovation in IM are as limited as Beane’s were for baseball.

I find the character of Pete in Moneyball fascinating. At the start of the movie I would estimate his real yearly value to be in excess of 10M. His perceived value is probably around 30k or 40k. Pete is super intelligent, and to some extent visionary, but not particularly ambitious. What he does is essentially unique in the industry at the start, but is not prohibitively difficult to copy. This means that as he opens his mouth, and results start to happen, his perceived value goes up rapidly. What may not be obvious is that as this allows his techniques to be copied, his real value is going down even as his perceived value is going up.

Now if Pete keeps his mouth shut, his perceived value will never rise, and his real value will never be utilized. Eventually someone else will come along, and have the same potential. I find it interesting that at the end of the movie, Beane gets offered all the money, and there is no mention of what happens to Pete. Presumably the sport gets overrun with analytics specialists (called sabereconomists in baseball), and their perceived value matches their real value at some point, which is probably in the order of $100k-$200k once the labor pool becomes saturated. Beane, having demonstrated his ability to use Pete effectively, is also going to see his value drop over time as others acquire the same skills.

Zynga games and baseball are very similar in that they are static. The rules for each are almost identical. Once you learn how to play baseball in one place, you probably know how to play it everywhere with little variation. The same is true of Zynga games, as each iteration is almost identical to the previous ones. While in both cases the application of economic methods revolutionized their success, their static nature made them easy to copy and this boost in success was short-lived.

Games in general, as opposed to Zynga games, are by nature a creative process, and the more creative and innovative a game is, the more customers will pay for it. Baseball owes its success, to some extent, by its lack of innovation. The game has changed little in the last 150 years. This makes it familiar and as such it is something an entire family can enjoy without much of a learning curve. The introduction of sabereconomics, as described in Moneyball, really just pointed out that the only thing that is not static in baseball is the players. So the game becomes more about who can arrange their players in the most effective configuration for the least amount of money.

For both Zynga and baseball, the introduction of economic methods to these games multiplied the profits, at least for a few years. Given the diversity of products in the interactive media industry, what if the same methods could be applied on a game by game basis, without forcing the game to conform to previous games? What if each game product could have its success multiplied without looking just like all the games that came before it? This would require the development of fundamental game economic principles that could be learned and then applied to any product without copying anything that had come before. This way every game studio could have a person trained this way to maximize the effectiveness of their product, or could at least share such a person with a few allied studios. Instead of “Pete” nodding or frowning every time you consider adding or cutting a player, your “GamerPete” could nod or frown every time you consider a design or feature, or at least advise you how not to screw it up. I realize that level of power would frighten most of you, as it did in Moneyball, but this could save/earn companies millions or even billions of dollars annually.

I’ve been calling this discipline applied virtual economics for the last few years but I think the field requires a new name. Something sexier. Baseball uses the term sabereconomics. What can we use?

[I’m still taking suggestions for a sexier name. Some months later I would get very serious about Moneyball, and I took the time to convert it to an economic model that could be applied to transform any industry. That paper is titled Moneyballification.]