Welcome to the third year of Statcast™. We're hoping it's going to be a fun one.
Over the first two years, terms like exit velocity, launch angle, and spin rate have entered the baseball vocabulary at a speed that's surprised even us, with slugging stars like Kris Bryant, Daniel Murphy and Justin Turner not only using the phrases freely, but actually using them to improve their games. Spin rate helped predict Justin Verlander's rebound, and it made Seth Lugo a must-know name.
The first two seasons, largely, have been about observations: How fast? How far? How much? We've tracked, for example, more than 10,400 miles of balls hit in play. Outfielders have run nearly 27 marathons -- that's over 700 miles -- to make catches. You know not only how far home runs go, but how hard they were hit. We've been able to quantify just how elite Giancarlo Stanton, Aroldis Chapman, and Billy Hamilton are at the specific things they excel at.
Now that we know what we can measure and what's useful, we've moved on into combining these observations into metrics, like when we introduced Barrels last year, identifying batted balls with the perfect combination of exit velocity and launch angle, since we know both are extremely important when it comes to success. As time goes on, we can distill that into measuring individual skills and a better look at a player's overall value.
That's for the future, though. What's new with Statcast™ in 2017? Here's just a taste.
Measuring Outfield Defense With Catch Probability
Catch Probability attempts to give a simple answer to a complicated question: Based on how far an outfielder has to go and how much time he has to get there, how often does a catch get made? We'll soon include direction and proximity to the wall, as well, to make it even more accurate.
Based on those inputs, when you see Byron Buxton ranging a great distance to make this spectacular catch, you don't have to guess how great it was. You can know. The nine percent Catch Probability tells you that more than 90 percent of the time, an outfielder given a similar opportunity does not make this play:
Video: DET@MIN: Statcast™ looks at Buxton's diving catch
Putting a difficulty on each ball allows us not only to tell stories around an individual play, but to show which outfields made the most Five-Star Plays (last year, it was the Rays and Reds), share the greatness of Hamilton's defense, look at which outfielders were the best overall (good news if you like Mookie Betts or Ender Inciarte), and explain why Andrew McCutchen really does fit better in right field. It also allows us to rate each catch as being "Five-Star," which is one percent to 25 percent, "Four-Star," which is 26 percent to 50 percent, etc.
Maybe you say that you didn't need numbers to know that the Buxton catch in the video above was great. Fair enough. But what about great plays that didn't look spectacular? Check out Kevin Kiermaier here, combining great speed with a good route to get to a ball that drops nearly 80 percent of the time, without even needing to lay out for it. The numbers tell a story of an elite play that the eye test may not have.
Video: TB@NYY: Kiermaier overcomes tough catch probability
Measuring Hitting and Pitching With Hit Probability
Hit Probability looks at exit velocity and launch angle to explain how likely a ball was to fall for a hit, and that can be very useful. For example, in last year's National League Wild Card Game, Brandon Belt did something extremely difficult -- he squared up a Noah Syndergaard pitch at 105.9 mph with a launch angle of 25 degrees, the kind of crushed ball that turns into a hit 95 percent of the time.
Video: Statcast™ looks at hit probability on Belt's fly out
In this case, Curtis Granderson made an excellent play to rob him, so Belt gets an 0-for-1 in the box score, but whether or not Granderson was able to get there really has little to do with Belt. He showed tremendous skill, even though it didn't become a hit, and we can use the 95 percent Hit Probability to credit him for that.
By taking the accumulated value of all batted balls over the season, and adding in real-world strikeout and walk numbers, we can tally seasonal numbers to get to Estimated OPS and find some interesting things. For example, Kyle Hendricks was an elite starter even without the great Cubs defense behind him. The Dodgers had baseball's most dominant staff in 2016, and Miguel Cabrera might actually be a better hitter than you already think he is.
Quantifying Outfield Depth
One of last year's more interesting stories was how the Cubs moved Dexter Fowler 21 feet deeper in center field, which may have contributed to why defensive metrics viewed him as having improved in 2016. Meanwhile, the Pirates pulled McCutchen in to play more shallowly, which they later admitted was a mistake, though it's not as simple as "deeper is better," of course.
Knowing where outfielders are positioned to begin each play also allows the opportunity to add some interesting context to great moments. For example, remember when Adam Jones made his instantly-iconic catch to rob Baltimore teammate Manny Machado in the World Baseball Classic? Of course you do:
Video: Petriello, Meyers discuss Jones' WBC 2017 catch
What was most interesting about that play is that Jones was baseball's shallowest center fielder in 2016, tied with McCutchen and Denard Span at 307 feet from home plate, on average. On that play, perhaps respecting Machado's power, he was 321 feet from home, which is a big difference. We've tracked 589 Jones catches over the last two years, and this play was tied for his ninth-deepest start point on a putout, making this one in the second percentile for him as far as starting depth on a catch. Does Jones get to that ball if he had an additional 14 feet to go? Maybe he does. Maybe he doesn't. It's a data point that adds additional context to a great play.
There's so much more, of course. We were able to use data to add context to potential or reported health issues from Bryce Harper, Carlos Correa, and Gregory Polanco. We're working on an exciting new way to judge foot speed on the bases and in the field. We'll be able to better quantify the different types of shifts, and what happens when they're in place. We'll be able to reenvision when a potential basestealer should or shouldn't go, based on lead distance, catcher throws, and pitcher release time, or help explain if a third-base coach made the right call to send a runner.
There's no shortage of ideas, obviously. The only question is which order to do them in. We're looking forward to more exciting developments as this year and future years unfold.
Mike Petriello is an analyst for MLB.com and the host of the Statcast podcast. He has previously written for ESPN Insider and FanGraphs.