When the 2013 Pittsburgh Pirates broke not only a 20-year streak without making the playoffs but also a 20-year streak of losing records, it wasn’t just a one-year fluke. Pittsburgh would be back in the postseason a year later, and as teams report to spring training this week, the Pirates are once again aiming for a berth from the increasingly competitive National League Central.
The story of how the Pirates turned their whole profile around is about more than just improving the talent on the team and using analytic principles, it’s also about aligning an entire organization around a new approach.
Travis Sawchik covers the Pirates for the Pittsburgh Tribune-Review and is the author of the book “Big Data Baseball: Math, Miracles and the End of a 20-Year Losing Streak.” In the latest in our Q&A series, we talked to Sawchik about how the Pirates set themselves apart in analytics, the role of MIT alum and rare traveling analyst Mike Fitzgerald and also about the broader lessons of his book. You can follow him on Twitter at @Sawchik_Trib.
Questions and answers have been slightly edited for length and clarity.
Q: How would you describe the timing of the Pirates’ analytics push? Were they one of the early adopters and we just didn’t hear about it much? Or was it only when (general manager) Neal Huntington took over?
A: They were not an early adopter. When Neal Huntington came to Pittsburgh at the end of 2007, there was no proprietary database, there was no analytics department. When he got here, they were basically building it from scratch. It was a big project. (Huntington) was influenced by his time in Cleveland. The Indians had developed DiamondView (a proprietary database) and he was exposed to that. He saw the value there. They hired bright people out of Ivy League schools or MIT-type institutions in the late ’90s and early 2000s. Billy Beane hired Paul DePodesta from the Indians.
Huntington’s from that school, so one of his first orders of business was, with limited resources, to find not only someone who could build the architecture of their proprietary database but also an analyst. They found both of those originally in Dan Fox, who was with Baseball Prospectus and had a computer science background.
So to answer your question, they were not early adopters, but when Huntington got here, they tried to accelerate that as much as they could, and they saw the urgency in building that department.
Q: How did Mike Fitzgerald, the MIT alum, come to the Pirates and how did he become such a central character?
A: That’s a pretty good story. When he was at MIT – he’s a big Boston Celtics fan from the suburbs of Boston – he and a friend traveled to Detroit to watch a couple games of the Celtics-Pistons playoff series. They were staying in a hotel near the team hotel, so on an off day between games, they stumbled upon Glen “Big Baby” Davis throwing a football around, and then Jeff Van Gundy walked over to this park by the hotels. Van Gundy and Fitzgerald started a conversation that turned to MIT and analytics, and Van Gundy had worked for Daryl Morey with the Rockets, who were the pioneers in the NBA. That conversation led to an internship with the Celtics, which in turn led to an internship with TrackMan, the pitch-tracking company.
The Pirates were installing TrackMan at PNC Park in 2011, and Dan Fox was asking Trackman if they had any bright young people who could help the Pirates understand this data. In 2012, the Pirates met with Fitzgerald at the Sloan Conference, and that led to him getting hired. It was kind of serendipitous.
Q: One unique thing about the Pirates is that Fitzgerald travels with the team and serves as an analyst on the road from the clubhouse. How did that come about and how quickly did manager Clint Hurdle and the players come around to it?
A: This is where I think the Pirates’ competitive edge comes in with analytics. This really starts with Clint Hurdle and Dan Fox and their relationship. When Hurdle was first hired in 2011, he sort of kept Fox at a distance. There was information available, but he was still sort of skeptical – Hurdle’s kind of an old-school guy. But in 2012, they had developed more of a relationship and they were meeting more often, and Hurdle was developing a trust level. So more of the data started to trickle onto the field in the second half of 2012 – we saw defensive shifts increase a little bit by the end of that season.
So in 2013, they started to ramp up the frequency of these meetings even more. In 2012, Hurdle was meeting with Fox at the start of every series. But by 2013, they were meeting every day and on the road having a conference call. And by the second half of 2013, he wanted his coaches to have even more interaction with a quantitative analyst, so they started to have Fitzgerald do some traveling, and Fitzgerald made most of the trips in 2014.
It wasn’t just top-down communication from the analytics department to the field staff. The coaches developed more of a confidence in Fox and Fitzgerald and started to want more information from them. It really became a very good spirit of collaboration and trust within the clubhouse, which is perhaps unique to the Pirates. I think that’s really where they did their damage – relationships and communicating ideas.
Q: Do you have any specific examples of what the coaching staff wanted to know from the stat guys?
A: To enhance the effectiveness of the shifts, the Pirates wanted to spike their pitching staff’s groundball rate. Several members of the coaching staff asked Fox and Fitzgerald to give them the pitch types, and locations of the strike zone, that every MLB batter was most likely to hit into the ground. They also wanted to see the psychological impact of pitching inside in an at bat and then going outside later in the at bat as it related to inducing groundballs. It’s an example that not every data-based theory or idea was driven from stat crunchers.
Q: What sort of background did Fitzgerald have in a clubhouse? What was the fit like there?
A: One aspect that Dan Fox liked about Fitzgerald was he was a true outsider that had never played at the professional level or been in a major league baseball front office. He thought there was value in adding a completely different perspective to the analytics staff. And it was Fitzgerald, more than anyone, who pounded the table for Russell Martin in the 2012-13 free agency period, in large part due to his pitch framing. Fitzgerald was free of any baseball bias entering the Pirates’ organization – as basketball had been his primary sport of interest.
Q: How open were the Pirates to talking for the book? I know there are some organizations like the Rays that don’t talk publicly about any of this and others that you see in the media much more. Where on that spectrum do the Pirates fall?
A: Somewhere in the middle. They weren’t as closed off as the Rays were. They were willing to let me interview most people I requested. They weren’t going to give me all access – every team’s become more guarded about protecting their secret sauce or their proprietary stuff. But they were willing to cooperate in the spirit of the book and the story I was trying to tell, so I have to be thankful to the Pirates for giving me that cooperation and giving me that openness.
I think it helped being around the team. I didn’t just parachute in – I was covering the team for the last two seasons and being there and seeing that also helped.
Q: At what point in those two years of covering the team did you know that there was a book here?
A: Toward the second half of the 2013 season (the Pirates’ first playoff campaign since 1992). It wasn’t just a fluky first two months of the season; it was apparent that they were going to win 90 games. Something had really changed here, but a lot of the roster had remained from the previous season; outside of Russell Martin and (Francisco) Liriano, a lot of the key players had remained from the 2012 season. And while the farm system had improved, it still wasn’t graduating a lot of talent yet to the major league team.
So the question was: How are they doing this? Because there weren’t a lot of external additions or even a lot of young players improving. The shifting had become obvious. They were starting to ramp that up. Then that led into if they’re shifting, what else are they doing? Every pitcher on the staff’s groundball rate spiked up. It became apparent to me that something special was going on. It had been a dramatic pivot too.
Those questions led to some early exploration of the data changes we could see just looking at FanGraphs or Baseball Prospectus. Those questions of why and how led to some initial answers and digging deeper. I wrote a few newspaper articles for the Pittsburgh Tribune-Review, but I started thinking this was bigger than a series of newspaper articles. This could be a bigger project.
Q: What are the broader lessons of this book beyond baseball?
A: That’s a good question, and the publisher and I did want to have some broader lessons that come out of this book, and communication is a big one. In the whole community, there’s been this “scouts vs. stats” debate going on since Moneyball. I think some of it’s overstated, but I do think a lot of organizations didn’t see all the available data trickle to the field because of resistance or communication failures. Some of that had been going on in Pittsburgh in Hurdle’s first two years here too.
So when you see the trust level with Fox and Fitzgerald and Hurdle and the coaching staff increase in 2013 and see the impact of that, the communication and spirit of cooperation and collaboration is a big one. It’s not just communication; it’s communication leading to the collaborative spirit. I think that explains a lot of their success the last two years.
Even though baseball has tremendous amount of data, it’s still a business that’s about people and personalities, and you have to have good leaders and find ways for people to work together. It’s still a sport very much about people, and it’s still a book very much about people.