Moneyball: Improve Performance Using Predictive Analytics

Baseball is often thought of as a sport dominated by teams with the highest payrolls. Certainly, there is a strong correlation between high payrolls and successful teams. However, there are many examples of teams that enjoy great success that don’t even reach the league average in total payroll. Two current teams that are examples of that are the Houston Astros and the Cleveland Indians. In fact, the Cleveland Indians are sitting in first place atop the American League Central, and their historic twenty-two game winning streak ended just six days ago. TWENTY-TWO. WINS. STRAIGHT. That’s almost as many wins as the Los Angeles Lakers had the entire 2016-2017 NBA season. Many factors go into achieving success like that; without a doubt, one factor is an effective predictive analytics team.

Predictive analytics are now as intertwined with sports as stories of cheating and heartbreak are with Taylor Swift. Predictive analytics weren’t always this commonplace in sports though. General Manager of the Oakland Athletics, Billy Beane, was the first known to use prioritization of statistics and data to make personnel decisions in professional sports. Beane, alongside Paul DePodesta, introduced sabermetrics to baseball clear back in 2000. Their decision to implement predictive analytics in baseball has changed not only baseball, but also the entire world of sports, for good. It’s made such a significant impact that it was made into a book (Moneyball) by Michael Lewis and then a movie starring Brad Pitt.

Misvaluing people

However, the story is about much more than baseball. In the words of author Michael Lewis, “Moneyball was about how people get misvalued and then, in turn, warped value systems encourage warped behavior in people. But primarily about people being misvalued and how a perfectly free market—what market could be more free than a market in professional athletes—can completely screw up peoples valuation. And there are all these biases that infect the human mind when it’s making intuitive judments—value judgments—especially about other people. We overvalue handsome people. We overvalue tall people…Moneyball wasn’t so much a prescription for how baseball players should succeed as it was for how someone who is evaluating other people should go about doing that.” So how was Billy Beane evaluating who to pay and who to pass on? The answer is, unsurprisingly, predictive analytics.

Billy Beane had the guts to analyze whether or not the current method of evaluating players was effective. Turns out it wasn’t. Not even a little bit. Boldly, he adopted a new and untested strategy of evaluating players. He employed a strategy that nobody else was using at the time, and many people wanted him fired for his unique tactics. His strategy was to compile players that scored poorly on traditional baseball metrics but very well on predictive statistical analytics. The Oakland Athletics had success beyond what most people outside the organization would have ever imagined. Unfortunately, they collided with the colossal New York Yankees. To put it in perspective, in 2001 the Yankees had a $125 million payroll while the Oakland Athletics had a $41 million payroll. The A’s lost, but they took the Yankees to a full five game series with an $84 million difference in payroll.

Predictive analytics explained

Now that a real world example of predictive analytics has been demonstrated, what exactly is predictive analytics? Predictive data analytics encompasses a variety of statistical techniques for modeling, machine learning, and data mining that analyze current and historical facts to make predictions about future and unknown values. Predictive analytics are used in scientific and research industries, marketing, financial services, manufacturing, sports, and many other fields. There are a variety of predictions that can be made using analytics: examples include evaluating baseball personnel, forecasting if stocks will go up or down, predicting inventory requirements in warehouses, or anticipating if factory machinery is starting to fail. In summary, predictive analytics is about employing advanced analytical techniques that enable you to extract patterns in behavioral data.

Benefits of predictive analytics

There are many ways to utilize predictive analytics beyond Billy Beane’s usage in evaluating baseball players. Predictive analytics are great for prediction, classification, and even approximation of unknown or complex functions. The example of Billy Beane evaluating baseball players using sabermetrics is just one example of prediction. Classification of data is essential to organize data for its most effective and efficient use. Furthermore, predictive analytics takes complex functions and simplifies them so sets of data can be utilized. The quality of predictive analytics is dependent on the quality of the data that is being gathered. As long as the data that is being input is quality data, then the predictive analytics that are being output will be high quality as well.

Another benefit is that predictive analytics allow businesses to target customers in the moments that really influence their decision-making. Consumers now expect to be targeted with relevant information that fit their wants and needs, and utilizing predictive analytics is one of the ways that happens. Research has shown that organizations that are competent in using predictive analytics are twice as likely to see a significant increase in sales as organizations that aren’t competent in predictive analytics. Obviously, being able to influence consumers in this type of way improves sales and the bottom line.

So whether you’re evaluating personnel for a baseball franchise with a significant payroll disadvantage or just wanting to reach your customers at the right moments, predictive analytics is a tool you should be utilizing to help you achieve your goals.

Picture Source: Pexel