Belli's 'hidden' web gem saved G5 for LA

Eighth-inning grab had 45 percent Catch Probability

October 27th, 2020

An underrated part of the absolutely bonkers-bananas play that ended Game 4 of the World Series on Saturday night is that the center fielder who got the shenanigans going by kicking the ball away, Chris Taylor, wasn't even supposed to be there. Neither was the man he replaced, starter AJ Pollock.

No, it was supposed to have been , who had started every previous game of the postseason in center for the Dodgers, but who had been shifted to DH shortly before the game due to a stiff back. Taylor often heads to left field late in games for his defense; Pollock spent years roaming center for Arizona. They're both capable. Neither, however, is Bellinger. It's easy to argue that one simple stiff back set in motion all of the insanity that ended Game 4.

The Dodgers suffered no such problems in their 4-2 win in Game 5 on Sunday night. Bellinger was back in center, and while he went only 1-for-4 -- his run-scoring first-inning RBI barely left the infield -- it's what he provided on defense that counts. A day after a late-inning ball to center nearly sidetracked the entire season, a similar one hit to the man who was meant to be there may have saved it. And there is a good chance you probably didn't think much of the play because Bellinger made it look routine, even though it was not.

Here's the setup, and this is important. It's the bottom of the eighth, and the Dodgers are up, 4-2. Manager Dave Roberts has made his second ready-to-be-criticized pitching move of the night; after removing Clayton Kershaw in favor of rookie righty Dustin May in the sixth inning, he replaces May with rookie lefty Victor González after May had allowed a single and a flyout.

A wild pitch by González pushes Kevin Kiermaier to second, and then a walk to pinch-hitter-for-the-pinch-hitter Mike Brosseau puts González in the unenviable position of facing The Best Hitter Alive, Randy Arozarena. González, to his credit, got Arozarena to weakly fly out to Bellinger, setting the stage for the seemingly reborn Brandon Lowe, Tampa Bay's best regular-season hitter, who had homered three times in the previous three games.

This was, so far as an advanced stat like leverage index goes, the third-most pressure-filled plate appearance of the entire game, 3.25 times more important than the average plate appearance. But leverage index doesn't know which players are involved, or what had happened the night before, or the complicated playoff history of the Dodgers' starting pitcher -- and it certainly didn't know about the three-decades-and-change run without a ring in Los Angeles. It felt even bigger than that.

Lowe took a slider low and inside for a questionable first strike, then lined a high-and-inside sinker into shallow right field. Bellinger was playing deep, both for himself and for Lowe; at 327 feet away from home, he was farther than his average starting point against lefty batters (318 feet) and what Lowe had seen across the entire year against lefty pitchers (322 feet).

The extra distance mattered. Bellinger had 74 feet to cover to get to the ball, and he had 4.4 seconds of time to do it. That's a combination that comes out with a Catch Probability of 45 percent -- which is to say that when outfielders are presented with similar opportunities, they successfully make the play slightly less than half the time. This is what we call a "four-star catch," out of a possible five stars.

"But wait!" you're thinking, probably because Bellinger caught it on the run and didn't even have to dive. But one of the notable things about Statcast is that it can show us how certain catches that look difficult are sometimes easier than they appear, while certain routine-looking grabs (like this one) are more than meets the eye.

Take, for example, Harrison Bader of the Cardinals -- a truly elite defender himself -- failing to get to a similar play where he had to go 73 feet in 4.4 seconds, back in July:

Here's another one from earlier this year, where Cincinnati's Jesse Winker had 4.3 seconds to go 72 feet. He didn't get there. He didn't come close.

Here's something that hits a little closer to home, perhaps. Back in 2018, then-Angels catcher René Rivera hit one of these types of balls to shallow right-center, where the center fielder had 4.3 seconds to go 75 feet. He was unable to make the play. You might close your eyes and wonder if this is what Lowe's ball might have looked like if Bellinger was unable to go, for this center fielder was ... then-Arizona star AJ Pollock.

So, what made those plays different? The first thing you think of is speed, and that's true to some extent. Bellinger's top running speed was 29.8 feet per second, which is taken by finding the one-second window where he was going his fastest. Bader got up to 28.9 ft/sec. Pollock was 27 ft/sec. Winker was a well below-average 23.9 ft/sec. There's a story here that is simply, "Bellinger ran faster."

But there's a story here about reactions, and instincts, too. There's a lot that happens before the fielder can even get up to that top running speed, and so often that first step or two is what defines whether the play is going to be made, even if you can't really ever see it on the broadcast.

One way we have to measure that is with the Statcast metric known as "jump," i.e. "did you get a good jump or not," and what it's trying to do is express how many feet in the right direction did the fielder cover in the first three seconds, compared to MLB average on similar plays. Think about it this way: You remember the diving catches, but you don't usually notice if the dive was made possible by a great jump, or made necessary by a poor one.

Bellinger's jump was a great one.

Bellinger, Game 5: +4.6 feet better than average
Bader, 7/25/20: -3.2 feet worse than average
Winker, 7/29/20: -6.3 feet worse than average
Pollock 8/22/18: -2 feet worse than average

That, for the most part, is it. Why did Bellinger get there without having to leave his feet, while Bader didn't really come close? Because after the first three seconds, Bellinger was nearly eight feet closer to the ball, before top running speed even comes into play. (And, of course, he did run faster at top speed, also.) That particularly poor jump by Winker might explain why his top speed was so unimpressive; he'd already lost his chance to make the play before top speed was required.

We've cherry-picked a few examples, obviously, but there are plenty of others. (There's also examples that got caught, too; here's Bader getting one last year, and here's Lewis Brinson doing it the year before, and so on.) But the point here should be that even though Bellinger did not have to leave his feet, this was a very difficult opportunity. It's one that's made less than half the time.

You'll always remember this World Series for what Mookie Betts did in Game 1, for what Brett Phillips did in Game 4, for what Arozarena continues to do at all times in all games. The most memorable play in Game 5 was Manuel Margot attempting to steal home, improbably. But the most important play was the one Bellinger made, the one that might get lost in the noise.

If Lowe's ball had dropped, the Rays probably pull within one run at 4-3 and have runners on the corners for Margot, forcing Roberts to make another tough pitching decision. Instead, Bellinger's play got them off the field. It helped them hang onto their third win of the World Series. One step closer to erasing 32 years in the wilderness.