The Moneyball Approach to Data & Analytics
After months of enduring shorter days and colder nights, spring has finally sprung. And while I am certainly excited by the promise of warmer weather, I’m really stoked for the start of baseball season. Being a diehard New York Mets fan, my hope usually turns to despair within a few weeks. But, seeing as the Mets are reigning National League Champions, my confidence is slightly higher this year. Not to mention, statistically speaking, they are the best team in baseball. So my money is on them to win big.
It may sound crazy to be motivated by a bunch of data and statistical analysis, but don’t we all do this everyday? Whether it’s on the field or in the office, we're constantly looking at data to deliver insights that will help us succeed – be it winning a ballgame or driving business growth. And while the level of analytics maturity varies, it has become an important part of sports and business. Unfortunately, not everyone agrees.
"The game is becoming a freaking joke because of the nerds who are running it. I'll tell you what has happened, these guys played Rotisserie baseball at Harvard or wherever the f--- they went and they thought they figured the f---ing game out. They don't know s---."
This blunt view on the state of baseball comes last month from former New York Yankees star, Goose Gossage. Clearly, not a fan of math, the former Hall of Fame pitcher believes the sport has become too reliant on data and analytics. What Mr. Gossage doesn’t realize is that analytics have been part of the game for a very long time – even used before the Cubs won their last World Series more than a century ago. During the years, baseball has seen the use of statistics evolve from pure observations on what happened in the past to complex predictions on what will happen in the future.
If you saw the movie Moneyball, or read the book, you know that the Oakland A's built a championship-caliber team by signing ancillary players based on looking at an array of oft-ignored statistics. In a time when most scouts and general managers were focused on numbers like batting average or ERA (earned-run average) to determine the value or potential success of a player, A's general manager Billy Beane bucked the norm and relied on lesser-applied measurements to evaluate players. Spoiler alert: It worked. The A's won their division when nobody gave them a chance.
And while baseball has upped its analytics game – despite the scorn from a select few like the Goose – the business world has yet to step up to the plate and leverage analytics for all its capable of delivering. Instead, most organizations today rely heavily on descriptive analytics (what happened) and diagnostic analytics (why did it happen) to inform decisions. And while it can be valuable for uncovering basic patterns, it may not always help in telling the whole story. It’s akin to merely looking at a baseball team’s win/loss record to make a decision.
For instance, one of the most misleading statistics in baseball is a pitcher's win/loss record. Unless a pitcher is in the midst of throwing a perfect game, they aren't the sole factor in whether the team wins or loses; the entire team plays a significant role. Despite this, pitchers are often rewarded with huge contracts and showered with praise based on wins alone. Being a fan of the New York Mets, I can attest to the fact that performance alone does not always translate to a win. I can't count the number of times a Mets starting pitcher actually exceeded expectations, allowing only one run, and was still burdened with a loss because the rest of the team couldn't hit. Just ask Matt Harvey, who last season had been the unlucky recipient of almost no run support – meaning he had to be next to perfect to win. On the opposite end of the spectrum, how many times has a pitcher on the New York Yankees earned a win despite pitching poorly, just because the insanely overpaid team scored double-digit runs? The numbers don’t always paint the clearest picture.
During a recent Mets game, broadcaster Ron Darling brought up an interesting point of view on the potentially one-sided view statistics can have on how a team approaches the game. He was discussing all the data on a young Phillies pitcher and his inability to pick off base runners throughout his career and how that information may have the Mets looking to steal more bases when he is on the mound. All well and good, explained Darling, except for the fact that the data doesn't show if perhaps the Phillies pitcher is working on this flaw and may be getting better at picking off runners. Basically, the data can only tell us what happened in the past, but not what may happen in the future. This is where predictive analytics comes in. By looking at the data in a number of ways, not just concentrating on prior behaviors, you will be able to identify the likelihood of future outcomes. It's something we are starting to see happen in baseball and business.
“Any business looking to gain a true competitive advantage should look into using advanced, predictive analytics,” explains Nipa Basu, Dun & Bradstreet’s Chief Analytics Officer. “All too often we get too fixated on taking the data at face value, but numbers without proper analytics to draw insight from can be misleading. Diagnostics of what has happened is only half the story, and is not very effective. Drilling down to predict what will happen can help companies identify new revenue opportunities and circumvent risk previously unseen.”
The use of any type of analytics can have enormous benefits for your team – be it baseball or B2B. But if you’re hoping to achieve All-Star business growth, it’s going to be important to look beyond what the basic data is telling you and employ predictive analytics to get at the heart of the story.