Red Sox management lessons: What Alex Cora’s “Coralytics” means to you

Congratulations to the Boston Red Sox for winning the World Series. What made the difference? Was it the heart and talent of the players? Or was it the cold, hard Moneyball analytics that have made baseball as much science as art?

It was a combination of both. This is “Coralytics,” the new approach of Red Sox first-year manager Alex Cora. And once you understand how Coralytics works for Alex Cora, it might change your approach to how you manage your team.

The Boston Globe described Coralytics in an article by the stats-loving baseball writer Alex Speier, titled “What’s Coralytics? When manager combines what’s best for player with synergy of analytics.” The article includes a description from Scott Boras, baseball superagent and longtime friend of Cora’s:

“Coralytics understands that you have to have the synergy of analytics plus the psychology of a player,” Boras said at Dodger Stadium prior to Game 3 of the World Series. “[It is] the idea of the filter of analytics to communication to the player in the correct way, the adaptation of understanding that, ‘I will take the character and psychology of the moment, and apply that more so than the analytic definition.’ Knowing that, when to use one and then the other, is Coralytics.”

Boras praised Cora for his ability to synthesize all kinds of information coming from the team’s scouts and analysts, and then to selectively — but not slavishly — apply the information based on what he was seeing in a game and what he knew of his players.

While numbers can offer context for a decision, in the estimation of Boras, they need to be part of the decision-making process rather than the entirety of it. To Boras, Cora’s gift is a feel for when he’s well-served to apply or resist the general conclusions of big data when employing his roster.

Traditional baseball analytics — as Michael Lewis so vividly described in his book about the Oakland A’s, Moneyball — is a decision-making tool. It tells you which players have more potential. It tells you whether to let the guy that’s up swing the bat, or to put in a pinch-hitter, based on their respective tendencies with the current pitcher. And analytics can also inform strategies for a whole team. For example, analytics encouraged managers to tell players to have long at bats, to wear out pitchers. As we got more information about exit velocity, analytics started the launch-angle revolution, in which players swing slightly upward to approach a 25-to-35-degree angle of flight, to improve the likelihood of hitting home runs.

But Coralytics is different. In Coralytics, the manager and his staff use analytics, not just to make decisions, but to give individual players the information they need to maximize their own potential.

Some examples from the Red Sox:

• Alex Cora stuck with hitter Jackie Bradley, Jr., during the season, despite his low batting average. Bradley is one of the best center-fielders ever, but when he put balls in play they were often caught. Cora and his staff helped Bradley to make adjustments that improved his average in the second half of the season, and he became the MVP of the series in which the Red Sox won the American League pennant, with timely hits and home runs.
• Especially in the National League parks, pinch-hitting is a crucial discipline — when the Red Sox were in Dodger Stadium, they frequently had to substitute for poor-hitting pitchers and catchers. The Red Sox prepared their hitters who weren’t in the game with work in the batting cage at Dodger Stadium, simulating the hard-throwing and tricky relief pitchers they would likely be facing in the game. Result: when those pinch hitters entered, they weren’t cold, they were warmed up, prepared, and primed to strike. Pinch hitters like Mike Moreland and 22-year-old Rafael Devers got timely hits, including home runs.
• The Red Sox prepared their pitchers with knowledge of hitters’ hot and cold zones — so they could avoid putting pitches in exactly the wrong spot and getting pasted. In some cases, the Sox and their pitchers actually preferred walking batters, putting pitches in places where a small mistake was more likely to be out of the strike zone than in a hitter’s sweet spot.

The crucial element of how to use all of this data is communication. As Boras told the Globe’s Speier:

“He is your manager. He is that autonomous person that directs you. When a player has that familiarity he was raised with, when it’s familiar to him in the big leagues, there’s a comfort there. When there’s a disruption of that, if he’s playing for the Oz and all he sees is the curtain, then the manager has a completely different feel. That’s the best way to describe it.

“There is no Oz in Coralytics. And I think the manager is also accountable. The players know he is accountable.”

The traditional manager-player relationships has two parts. First, the manager is a friend and source of support to the player — to buck him up. Second, the manager makes decisions about who to play when and who to sit, decisions that can seem arbitrary to the player. These two roles are in opposition. But if there is a communication, this all changes. The manager shares with the player exactly the information that the player needs to succeed. It’s not “We love you,” but “We want to help you, and here is some information that you can work on to get better.” When a manager is sharing like this, players understand why they are playing, why they are not, and what their role is.

There is no better example of this than the winning pitcher of last night’s clinching game, David Price. Price came into the playoffs with an unfortunately reputation for losing when it counted — his teams had won zero times out of ten in games he started in the playoffs. But Cora knew that Price was a lionhearted pitcher with incredible ability. The coaching staff had already transformed Price during the season, moving his starting point to the first-base side of the pitching rubber. That move had paid off.

Now the pitching coaches had noticed that in playoff games against the Yankees and Astros, Price was releasing the ball lower than was typical for him — 5.7 feet off the ground, compared to 5.85 feet during the season. The result was lower velocity and pitches that lacked the deceptive movement that baffled hitters. So before a decisive game with the Astros in the League Championship Series, the coaches helped Price make an adjustment, and the result was a dominant performance in which Price helped the Sox to clinch the series.

Price went on to make major contributions to winning the World Series, including last night’s game that won the whole thing for the Red Sox.

Consider the two analytics approaches here:

• Traditional analytics: “David, you aren’t getting it done in the playoffs. Analytics shows you’re not the best choice. We’re going with someone else.”
• Coralytics: “David, I think we can see the problem you’re having. If we can adjust your motion to get your release point higher, you’ll get more guys out, because your stuff will be nasty. Let’s work on that so you can help us win some games.”

What about confidence? Confidence is the intangible that makes players better. But as Coralytics shows, confidence is not about telling someone they’re great, when they’re actually not. It’s about showing them what they can work on to get better. I’m sure David Price’s confidence was much higher in the games he just won. He knew he could do it, not because someone patted him on the shoulder and told him he could do it, but because he had something to work on to get better. The simple act of making a substantive change based on data increases confidence and generates more success.

These players love their manager. Their manager loves data. These two ideas are not inconsistent, they are strongly connected. But it only works if you can find the right data, explain it to the player, and then help them to act on it, so they’re going to get better.

What does this mean for you?

Data is supposedly the new fuel that makes businesses better. You can use data to figure out how to change your Web site or what businesses to buy or which products to work on. And that’s great.

But you can also use data to terrorize your staff. Get the average handle time of those calls down by 30 seconds or you’re fired. Your bonus depends on our Net Promoter Score.

When you set metrics like that, the result is not just resentful people, but people who game the system to boost their metrics. And that’s not helping anybody.

Neither is the gladhanding and cheerleading. Nobody buys that crap when you’re spouting it while culling people for poor performance at the same time.

Have you considered how to use metrics to identify what your staff can improve? Have you used metrics to help boost the performance of individuals?

Have you determined what it will take to make your people better — individually — and showed them why these improvements will matter? Have you showed them how they can be better?

This is hard to do. It’s not how businesses think. But it is, perhaps, an area in which AI can help in the future.

If you can figure this out, your business results will improve — and your people will love working for you. They’ll not only be better at their jobs. They’ll be motivated.

This takes effort. But if it works for Alex Cora, it ought to work for you, too.