Bradford: How analytics helped the Red Sox win their World Series

Rob Bradford
November 20, 2018 - 10:06 am

USA Today Sports

Of all the regular season moments that helped pave the way for the Red Sox' World Series title, what happened in the sixth inning back on June 21 in Minneapolis has to be pushed to the top of the list.

With the Sox holding a 2-0 lead over the Twins, Rick Porcello delivered his 0-2 offering to Joe Mauer. The lefty hitter jumped on the pitch, sending into the right-center field gap, almost exactly where Mookie Betts was standing. In fact, the right fielder had to take exactly three steps back before casually hauling in the liner. After throwing the ball back in and signaling the first out, Betts couldn't help himself.

The right fielder who had been adamant that he and his Red Sox' outfield-mates didn't need those cards and analytics being thrown the infielders way since spring training -- going so far as telling WEEI.com back in April, "It doesn't seem like baseball sometimes with all of the extreme shifts," -- couldn't help himself. He took the card out of his back pocket and waved it toward the Red Sox' dugout with a huge grin.

The plan had officially taken root.

"We play the Twins quite a bit in spring training and we face Joe Mauer and Joe Mauer is probably the most extreme opposite field shift in baseball," remembered the head of the Red Sox' analytics department, Zack Scott. "We would be trying this out in spring training. We knew that was a tricky thing in terms of buy-in. That moment while we realize it was a lot of good luck that it was lined right at him, that moment stood out because that was a hurdle for us in spring training because they were like, ‘Whoa, what is this?’ That extreme shift made people nervous."

All of it made people nervous. You know why? It was new.

Yet it should be understood how important the Red Sox' use of advanced analytics (the most polarizing word in Major League Baseball these days) was in accomplishing their ultimate goal.

It is a story that is about much more than simply looking at how far Scott's crew has come -- growing to 10 people for the 2018 season -- or where it's going -- hiring five more analysts in the coming weeks). This was an evolution that started a revolution.

Proof? 

Examples are easy to find, and they aren't just limited to citing shifts. Sure, if you want to go to BaseballSavant.com and reference the fact the Red Sox shifted 43.2 percent of the time against lefty hitters compared to a 32.1 percent rate the season before, go ahead. This is much, much more.

Take a look back, for instance at Christian Vazquez catching the likes of Chris Sale, David Price and Rick Porcello in the postseason. This wasn't just a matter of throwing out the catcher and hoping for the best. What you saw with Vazquez -- whose game-calling skills throughout the regular season weren't perceived to be on par with the likes of Sandy Leon -- was a product of information and his willingness to implement it. It was a microcosm of what made this brave new world of increased analytics actually work.

"Vazquez started catching a lot more so my focus was kind of how he was calling the guy based on what conversations I knew had taken place or what information I knew he was getting," Scott said. "Watching that, seeing how he worked with pitchers he hadn’t typically worked with, it’s a pretty significant change especially for some of the guys who were just used to throwing to Sandy. It was important for Alex (Cora) to go through with more detail with the information we were providing because he was taking somewhat of a risk of changing that up. That was where I was primarily focused.

"(Vazquez) definitely stepped up. Defensively, the way he was calling the game there was more conviction with what he was doing. I thought his plan was really good. It wasn’t against Sandy, but it was a point of development for (Vazquez)."

KNOWING THINGS HAD TO CHANGE

Cora had come in with a bunch of ideas after spending his one season as the Astros' bench coach. When it came to manpower in analytics departments, there was the Dodgers, the Rays, the Astros, the Yankees and then everyone else. Heading into 2018, the Red Sox weren't in the conversation.

But just because Scott got the go-ahead to hire a few more people, and there was a manager who wanted to prioritize uncovering new kinds of information didn't mean this was going to be a turn-key process. The infielders, for instance, could be seen awkwardly trying to look at their new positional cards throughout the second half of spring training, often times dropping them on the ground just before a pitch. And the outfielders? It wasn't until early June they were finally convinced to even carry the small pieces of paper.

Some of the pitchers weren't convinced all this new shifting was the way to go. And the coaches found themselves trying to develop relationships and trust while opening eyes to a brave new world.

"It was a transitional phase going with a mostly new staff. So I didn’t really have a blueprint," Scott said. "It was more collaboratively we had ideas. Alex came in with ideas. Pretty firm ideas with what he wanted to do, which made the jobs of my analysts pretty easy because it was pretty clear. When I say what he wanted to do I mean the kinds of questions he wanted to be able to answer its the information we were able to provide. The kind of things he cared about when preparing. The kind of ideas he thought could push things forward, even beyond what he was exposed to in Houston. So he had a lot of good ideas and when we would talk they would grow from there. We would brainstorm on things. Me with him and me with my team. It wasn’t like, ‘OK this is exactly how it was going to go and we had this plan ready to go as soon as we hired Alex.’ It was more him driving it and us contributing new things he might not be aware of and we were studying.

"It would be nice to say it was this blueprint where we went through a checklist and banged everything out. But there was a learning curve. We were just working through things as a staff and just kind of evolved as the season went on."

But then, as Scott said, "A lot of things went right."

The Red Sox started the season at 17-2 implementing some of their new ideas. Offensively, hitters were seeing results of these new attack plans. And, perhaps most notably to the participants, the infielders were seeing great results out of the gate.

"We know because of the human nature aspect of it we needed some good luck early on," Scott noted. "I know with the infield shifts it was working really well, and that was probably by chance, but we needed that to get some buy-in. If we had a run of bad luck where a bunch of balls were beating shifts it would have been a problem. The outfield was a tougher transition. We didn’t really start doing it until early June. Alex was like, ‘They’re resistant, but it’s going to take some time so we should be patient.’

"You kind of go into it with that expectation, that this isn’t happening overnight. I give Alex a ton of credit. Not only is he really invested in this work but he has the right mindset and personality and has realistic expectations. He will be clear if something is important to him. He communicates it very well to the players the coaches that it is important, understanding that we aren’t going to make every play. For a pitcher, the human nature is to get frustrated the one time a ball beats a shift when it looks routine. You go into it knowing there is going to be an adjustment period, especially for players who had been here."

GETTING THE MESSAGE

Scott, Cora and Co. had plenty of ideas, and a lot more information than ever before. But that was just part of the equation.

Making it all translate to wins was another challenge, one which ultimately was a bit more challenging than anyone could have realized.

"Trying to figure out how best to communicate that without being overwhelmingly complicated, that was where there was more trial and error," Scott said. "It was like we were presenting some information and we would kind of tell them, ‘We’re accounting for X,Y and Z.’

"It gets complicated and it’s hard to wrap your head around it. You think how to adjust to it all. You don’t want to go in and explain, ‘Well, we used this mixed model.’ But we don’t want it to be a black box either. To find the best way to communicate is a good challenge that my team and I faced. I would say that was where we realized, ‘You know what, we present it this way, we could probably do something better that is more clear.’ That was really where things didn’t work, where we would present something, try and get feedback and then realize, ‘You know what? We’re still asking the guys to compute too much on the fly.’ Our whole goal is to help them in their preparation so when the players take the field they don’t have to do that much thinking and when the coaches are in the game and it’s game speed they don’t have to do too many calculations in their head. We’re trying to get to that sweet spot where you present the information in a way where you can pretty much glance at it and know what it’s telling you in a certain situation and then you can adjust with your baseball instincts on the fly rather than thinking you have to look at these six reports.

"What these guys have been doing for years is trying to calculate a ton of information in their head during a game, or before a game in preparation. It’s just really hard. I don’t think any human being can really do that. Our job was to simplify that."

This brings us back to the Betts catch.

Cora could have simply told the outfielders to carry the cards. He could have mandated outfield coach Tom Goodwin to implement the new way of doing things. Or the manager might just let the trio of outfielders go on their merry way and figure it out on their own. There was none of that. Instead, Cora would ask the analtyics folks to print up what actually happened and what most likely would have transpired if the new information was put into play. That was then presented to the particulars.

It had to be made clear that the spray charts and previous type of models presented the previous seasons were a thing of the past. This was next-level stuff, which Betts ultimately discovered in late June at Target Field.

"That was the work we did to make the pitch through Alex and the coaches," Scott said. "I think that was a good approach and that was Alex’s idea."

WHAT NOW?

By the time the Red Sox got to the postseason the system was in place.

While previous playoff runs were always thick with information, this was something different, especially when it came to the involvement of Scott and his crew.

"There is more a detailed focus on how to beat the other team, how to attack hitters or how to approach their pitchers," the analytics boss said. "That’s always been true but this year was different because we were using more of the kind of information that was coming from my department. It was a day to day conversation, with Alex reaching out to me, or talking to the coaches. That was different. There was way more dialogue about matchups and lineups."

Why was Eduardo Nunez playing against a right-handed pitcher? Or how about the thinking behind playing Ian Kinsler over Brock Holt? A lot of those calls were a product of the aforementioned conversations.

(Ironically, one of the more drastic changes, moving Jackie Bradley Jr. to left field just for Manny Machado's at-bats, wasn't a product of an analytical get-together. "When he did it I was like, ‘Oh, that’s a really good idea.’ But we didn’t talk about that at all," Scott said.)

Eleven postseason wins later, the system proved itself.

But the Red Sox know this is just the start. Competitors are catching up, evolving and, in some cases, primed to surpass introduce ideas the majority of baseball haven't even thought about. It's why the Sox have signed off on significantly bolstering the analytics staff. Because as well as things worked in 2018, it was still far from perfect ... other than winning that World Series trophy, of course.

"The demand coming from the major league clubhouse was a great experience but one of the things it highlighted was how thin we were," Scott said. "My job is to prioritize these things. The number of projects was so much larger than the capacity we had.

"The game isn’t a formula. It’s not a script. We have to present it in a way where they can work off of it. It’s kind of a baseline. But things are going to change as the game goes on."

Comments ()