Yankees Magazine: Numbers Game

With more information than ever available after every pitch, who uses Statcast data and what does it mean?

July 12th, 2017
Whether by design or by chance, the 2017 Yankees boast what could be classified as a group of Statcast All-Stars, with players all over the field who have exceptional tools. (New York Yankees)

With the home team in front, 7-3, on an 89-degree afternoon in a game that is already more than two-and-a-half hours old, some of the intensity has drained out of Yankee Stadium as steps into the box against Baltimore reliever Logan Verrett with a 1-1 count.
Until Verrett delivers his next pitch.
For the third time in the at-bat, Orioles catcher sets up low and away, but Verrett's 85-mph slider drifts up and over the heart of the plate.
Thwack.

The crack of the bat echoes through the Stadium, and the ball is in flight.
Judge hit the ball hard -- we can hear that. He hit it far -- we can see that. But exactly how hard did he hit it? How far did it fly? And is this blast the farthest we have seen this season? In 2017, these questions are no longer up for debate.
"If you grew up watching Mickey Mantle, let's say, you might want to know: Did he hit the ball harder than Willie Mays?" asks MLB.com analyst Mike Petriello. "Or, did Dave Winfield throw the ball harder from right field than any other outfielder in the 1980s? Those are the kinds of things that you could argue about with your friends forever, but never really have an answer to."
Now we can answer those questions -- and many more -- thanks in part to seeds that were planted during a postseason game in Oakland 16 years ago when a certain Yankees shortstop made an unorthodox and unforgettable play in the field.
***
In 2001, Major League Baseball Advanced Media -- the technology arm of MLB -- was in the midst of developing PITCHf/x, which tracks pitch type, location and velocity, when Derek Jeter pulled off his iconic "flip play" in Game 3 of the American League Division Series against the A's. The play inspired statisticians at MLBAM to aim higher than where PITCHf/x could hope to take them, and they began to envision what would eventually become Statcast™.
With two outs in the bottom of the seventh inning, Jeremy Giambi on first and Oakland trailing, 1-0, Terrence Long doubled into the right-field corner off Mike Mussina. With Giambi steaming toward the plate, Shane Spencer picked up the ball and hurled it over the heads of first baseman Tino Martinez and second baseman Alfonso Soriano. Giambi was going to score.
But Jeter bounded across the infield, grabbed Spencer's errant throw and flipped it to catcher Jorge Posada, who tagged Giambi on his right calf for the inning's final out -- preserving a 1-0 Yankees lead in a potential elimination game. (The Yankees would hold on to win this game, as well as Games 4 and 5, to advance to the ALCS.) Jeter appeared to come from the area near second base, where he was waiting in case Spencer's throw targeted Long instead of Giambi.

Cory Schwartz, then the manager of stats at MLBAM, had the same question the rest of those watching the game had: How did Jeter wind up there -- hurdling across the first-base line -- from his shortstop position?
"There was no way to deconstruct that play and figure out how it happened," says Schwartz, now MLBAM's vice president of stats. "We became very interested in trying to track everything that was happening on the field to better understand the DNA of what creates the different outcomes we see, because in this case in particular, that was the difference between going on and eventually playing in the World Series versus going home for the winter."
Jeter wasn't the only variable that would have affected what Statcast™ could have uncovered. Statcast™ would have recorded movements by every actor involved in the play, from Mussina and Long to Spencer and Giambi, because Statcast™ does more than just record exit velocity, distance and spin rate. It combines Doppler radar and six high-speed cameras to create a data point for each player's movement on the field.
Had the ballpark in Oakland been outfitted with Statcast™ technology on Oct. 13, 2001, it would have tracked Jeter's starting position on the play, where he went when the ball was hit, and, most importantly, the velocity and trajectory of Spencer's throw, as well as the speed and angle at which Jeter ran that led to him intersecting the ball's path at precisely the right moment.
But the technology used for Statcast™ was years away from being ready. PITCHf/x wouldn't even debut until the Yankees' 2006 American League Division Series against the Tigers, before going league-wide the next year. Still, MLBAM set out to track every movement of every player on the diamond.
***
MLBAM introduced Statcast™ to the public in 2015, and with the new technology came a new language that the casual baseball fan had never heard before. There was exit velocity and spin rate, launch angle and release point. The terms were and still are a departure from wins, batting average, home runs and other pieces of the baseball lexicon that have remained the norm since Cy Young, Ty Cobb and Babe Ruth set the standards for those statistics more than 80 years ago.
How many at-bats did a scout need to see before he knew that Ruth possessed raw, sustainable power? A week's worth? A season's worth? Somewhere in between? Statcast™ helps front offices gauge players' ability in less time by giving them raw data instead of waiting for a season-long sample size.
"It definitely allows you to quantify a lot of the same things, a lot of the same tools, that scouts are evaluating," says Yankees Assistant General Manager Michael Fishman, who says that, with Statcast™, "You hopefully can get at what a player's true abilities are a little quicker by measuring the tool and the skill rather than waiting for enough results to be able to tell you what a player is."
Whether by design or by chance, the 2017 Yankees boast what could be classified as a group of Statcast™ All-Stars, with players all over the field who have exceptional tools. Judge belted the hardest hit home run of the Statcast™ Era, a 121.1-mph laser to left field on June 10. In July 2016, unleashed a 105.1-mph fastball that remains the fastest pitch thrown since Statcast™ was introduced. And in April 2016, threw a runner out at the plate with 105.5-mph rocket from left field -- the hardest throw by an outfielder ever recorded by Statcast™.

Instead of waiting for Judge to hit 20 home runs, Chapman to strike out a batter per inning or Hicks to record a dozen outfield assists, Fishman can draw a conclusion in a much shorter time span. "It can take a pretty large sample for traditional stats to tell you the true story on a player," he says. "You see all the time a good player who has a bad year. And a lot of those times, over the course of an entire season, things just didn't even out for him."
Some Statcast™ metrics forecast a player's skill faster than others. A player's hitting ability can be predicted in as few as 40 to 50 batted balls, according to Petriello. For example, Judge was first in baseball in average exit velocity through July 12, at 97.2 mph, and ranked third in the American League in batting average at .329.
For pitchers, on the other hand, the measurements that Statcast™ takes are not always a direct indicator of future success.
"Having a high spin rate by itself does not automatically make you a great pitcher, in the same way that being able to throw hard does not make you a great pitcher," Petriello says. "There's lots of other stuff that goes into it: location, deception, all this other stuff. [Spin rate is] a skill that you can't be taught, most likely, so if you have this skill, it's a great starting point."
Spin rate measures the revolutions per minute that the ball makes after it leaves a pitcher's hand, and with a fastball that spins on average around 2,500 rpm, the Yankees' is a strong case study.
Fastball spin rate correlates to a high swing-and-miss percentage, with an rpm of 2,400 on any pitch qualifying as above average. In 2016, the 10 pitchers with the best spin rates in baseball all induced an above-average frequency of swinging strikes.

Through mid-June, Betances had thrown his four-seam fastball 45 percent of the time and generated a 43 percent swing-and-miss rate. He also owned the second-best K/9 mark among all pitchers with at least 10 innings pitched in 2017. But it's not just his spin rate that allows him to have success. Betances also throws hard, keeps runners off base and doesn't get rattled in late-game situations. The high spin rate is just one of many factors that have helped him succeed.
"It's all a part of the process, trying to gather as much information for every decision, whether it's a player transaction or a promotion in the Minors," Fishman says. "Trying to put the puzzle together as best we can using information from all of our scouts, coaches and data."
While front offices use Statcast™ tools to evaluate and compare players, the measurements it generates also have tangible effect on the field because they provide a new trove of information for the Yankees' coaching staff as it readies for each series.
"Our analytical department does a really good job breaking it down for us, so we can understand it, and it will not all look numerical," says Yankees third base coach Joe Espada, who handles much of the team's infield instruction. "They'll provide spray charts; they'll provide suggestions on where to play certain guys. And we spend some time looking at video also to kind of help us prepare for the series."
Before Espada puts together a defensive plan for the infield for each series, he confers with pitching coach Larry Rothschild about how the Yankees hurlers plan to attack opposing hitters.
"We have a lot more shifts now catered toward how pitchers pitch to guys," says first baseman Chris Carter, who was designated for assignment on June 24. "Now you see a different shift every time guys come up, or sometimes with two strikes a different shift or whatever the situation is."
That would explain why might leave his shifted position on the right side of second base and move back to the left side of the infield when the Yankees get two strikes on a left-handed, pull-happy opposing hitter. Such a move by Espada might indicate -- based on the hitter's spray chart and the way the Yankees plan to attack -- that the batter is less likely to pull the ball with two strikes.
While baseball has been infiltrated by more data than ever, the Yankees' coaches prefer to carefully filter what the players take with them into each game. "I don't like to give my guys a lot of information," Espada says. "I like to look at it and I give it to them as simple as possible, so when they get on the field it comes natural for them."
The more players can relax, the more Espada can sit back and watch their raw talents surface.
"I'm an Aaron Judge coach and an Aaron Hicks coach, but I admire seeing their tools, their strength, their arm strength," Espada says. "It's always great to see guys perform at such a high level. Judge's power -- he's probably the strongest guy in the Major Leagues -- I get to watch it every single day. I think fans, they want to see more."
As soon as Judge connects with Verrett's slider, Didi Gregorius jumps up. He's standing on a bench, leaning against the roof of the dugout and cupping his left hand against his forehead as if he is looking over the horizon -- which is essentially where the ball lands.
Judge sends it over the left-center field bleachers, where it touches down an astonishing 495 feet from home plate. It got there because he hit the ball 118.6 mph at a launch angle of 28 degrees. It is the longest home run hit in 2017 and trails only a 504-foot bomb by Miami's as the longest homer hit in the Statcast™ Era.

Each of those blasts qualifies as a ball that was "barreled," a metric MLBAM introduced late in 2016. For a batted ball to fall into this category, it must have an exit velocity and a launch angle that, on average, results in at least a .500 batting average and a 1.500 slugging percentage. For a ball hit with an exit velocity of over 100 mph, its launch angle must be between 24 and 33 degrees.
Will barrels someday replace home runs? Probably not -- but that's not the goal. The goal is to comprehend why a pitch, play or game unfolds the way it does, not to turn baseball into a pastime played on a spreadsheet.
"Ultimately what this all boils down to is that the metrics have to be relevant and repeatable and have context," Schwartz says. "We want to look at things that help understand what we're seeing on the field as they're happening. That's really what will make a metric resonate with fans and why certain things have caught on."
Thanks to Ruth, home runs caught on nearly a century ago and never went out of style. So as the Judges of the world continue to hit the ball harder and farther, Statcast™ can help fans, players and statisticians alike agree on one simple principle.
"Hitting it hard is great," Petriello says. "Hitting it hard at the right launch angle is better."