Here’s a piece by Liam Schmidt, who attended #CollegeSeries Baseball Analytics Night on April 10 at Citizens Bank Park. He gives his thoughts on the evening while shedding some light on the Phillies’ data-gathering process.
With so much discussion on the Phillies’ increased reliance on analytics to make managerial decisions, I thought it would be interesting to share a few observations from attending this year’s #CollegeSeries Baseball Analytics Night on April 10 at Citizens Bank Park. The program featured an informal networking session and panel discussion for more than 200 college students and associates from schools across the Delaware Valley.
Gregg Murphy moderated the panel discussion, which began with Gary Matthews. Sarge professed an old-school view of analyzing baseball. He brought up the “eye-test” when asked about judging hitters, iterating that you can just tell when some guys have “it.” Before giving way to the panel, Sarge gave some final points on hitting; notably, how the scoreboard always dictates strategy. The panel included members of the Phillies Baseball Analytics & Operations team. After introducing themselves and their backgrounds, they took questions from those of us in the audience.
Leading off was Alex Nakahara, a senior quantitative analyst with the Phillies. He talked about his background as an air traffic controller, and said that he never imagined himself working in sports. For the Phillies, he creates analyses and tools for understanding and visualizing data. He fielded questions about the more technical aspects of baseball analytics. He said the team uses many data visualization tools, such as R, Python, JavaScript, ggplot2, and matplotlib, among others, to prepare applications, reports, and slides for players, coaches and fellow analysts. R and Python are programming languages used by data scientists to organize and analyze unstructured data. ggplot2 and matplotlib are data visualization tools for each of these respective languages, used to create easy to understands graphs, scatterplots and other types of data displays. He emphasized that he creates specialized graphics, customized for particular needs and audiences, hence the need for such a variety of tools. One student asked him about how he factors intangibles, such as team chemistry, into his models. He responded that he can add certain variables into the models that were not previously included, but a major part of his role is recognizing what is included and what is not. In these cases, the analytics staff will report to the coaching staff that, “we’re not considering this, but you should.”
Seated on the other side of the desk and talent evaluation spectrum was Johnny Almaraz. As the director of amateur scouting, Almaraz has nearly three decades of experience in baseball operations. He explained his role as attempting to marry tradition with analytics. At the amateur level, much scouting and analysis is done in person with small sample sizes. As an example of his scouting acumen he brought up Scott Kingery, who he scouted at the University of Arizona. When he locates players of note, he must test to see if this sample size is appropriate, and consider other factors other than what he sees in person. He said that the most dangerous phrase an organization can use is “that’s the way it’s always been done.” He has witnessed the growth of technology and analytics first-hand throughout his many years in the business, and emphasized that teams and front offices must be willing to adapt to maintain a competitive advantage.
In the middle of the Q&A session, Murphy surprised the crowd by introducing starting pitcher Ben Lively. As a major leaguer, he has fully embraced baseball analytics, since “winning is fun, [and] losing sucks.” He said that the entire roster, the youngest in the league, feels the same way. He references heat maps and hitters’ tendencies on a tablet in the morning before starts and in between innings. When he has a bad game, he wants to know why. He has been able to learn from Jake Arrieta, who brought over techniques that he had been using for years in Chicago.
Seated in between Nakahara and Almaraz were Andy Galdi and Sam Fuld. Galdi is director of baseball research and development, and Fuld has quite a unique role as a player information coordinator. Essentially, Galdi is in charge of the application of analytical research, while Fuld gathers relevant information and implements it; working closely with the front office, coaching staff and players. These two fielded a question about how Gabe Kapler balances instincts and analytics to make daily lineup decisions. They said he uses a combination of both and spends a lot of time analyzing objective reports and considering his own subjective opinions. Fuld talked us through his daily routine, which involves preparing information in the mornings and then interacting with players throughout the afternoon in preparation for a game, acting as a bridge between the analytics staff and players. A former outfielder himself, he started this role in November and believes it to be truly unique for major league baseball.
Some hot topics, such as the strategy behind defensive positioning, the Rays’ four-man rotation, the Astros’ four-man outfield, exit velocity and the use of other more obscure metrics also came up. All in all, the panel discussion was very interesting and informative from my perspective as a college student studying statistics. The $12 food voucher also proved useful during a Hatfield Dollar Dog Night. And on the chilly sub 40-degree evening, the Phils were able to pull out a 6-1 win against a struggling Cincinnati squad.
Latest Comments