The Moneyball Strategy: Making Big Moves With Little Leverage

News sources dubbed the 2012 election “the Moneyball election.” Using aggregative statistical models, newfangled forecasters like Nate Silver upstaged traditional electoral analysts, who, like Billy Beane’s original team of scouts, rely more heavily on experience and intuition. Fact is, in our information age almost nothing remains entirely unquantifiable, and more and more sectors are taking note of the Moneyball strategy: employing data to identify key factors others overlook. Financial HR groups might want to take note.

For those with an aversion to America’s favorite pastime, Moneyball is a 2003 book by Michael Lewis (adapted into a 2011 movie), which tells the story of the Oakland Athletics’ 2002 season. With a $39 million budget (versus the Yankees’ $114 million one for the same year), general manager Billy Beane (Brad Pitt) is stuck on one question: How do you compete for talent with a lower budget? Peter Brand (Jonah Hill), a young Yale economics major, provides the solution: Fill the roster with players who get on base, and never mind if they have an ugly girlfriend or less-than-perfect training background.

The older scouts do pay attention to those types of factors—they think having an ugly girlfriend shows a lack in confidence. But in reality, such considerations don’t predict performance on the field—Brand’s data does.

The problem of misjudging up-and-coming talent is hardly limited to baseball. Hiring groups for Wall Street firms often overvalue factors like where candidates went to school or whether they wear the proper type of suit, because they have no set of statistics that effectively predicts how a certain candidate will perform. And with ever-lower hiring quotas and budgets, HR groups are as much between a bat and hard place as Beane was: They have few resources but big expectations to deliver on.

Lewis’ subtitle for his book is The Art of Winning an Unfair Game, and from the talent’s point of view, nowhere is getting onto a roster less of a fair game than in finance. If you didn’t go to that right school or don’t have a well-appointed uncle, you might as well be Chad Bradford, the pitcher with a funny throw that no team wants to take a chance on.

But using Brand’s predictive statistical models, the Oakland Athletics signed Bradford, along with a bunch of other misfits. At first, it didn’t go so well. By the end of the summer, though, the Oakland Athletics were rounding out a 20-game winning streak, setting the American League record that still stands today. The message? Gamble big, have patience with your hand, and come out a winner. Don’t just shake up the hiring model—reinvent it.

Business Dictionary guru Kevin Mulligan suggests that recruiters “identify the factors that matter most to your organization, and pursue those candidates no matter the industry norm.” The hardest part for HR groups in finance right now is successfully determining which candidates have those factors that matter. They need a way to aggregate the right data on applicants, then to be able to model it to see how a certain candidate—or certain team of candidates—will perform.

At the end of the film version, the owner of the Boston Red Sox invites Beane to Fenway Park and slides a $12.5 million figure across the table to him, calling anyone who wouldn’t want to employ Beane’s new methods a “dinosaur.” Banks cannot afford to be ambling brontosauruses—they need to find their “Brand” of statistical modeling to improve talent-sourcing now, rather than later.