January 14, 2014
Baseball's Seven Wonders
Mariano Rivera's Postseason Performance
First in a series.
It’s practically impossible for a pitcher to allow less than a run per nine innings. Since 1920, there have been about 20,000 pitcher seasons that have covered 50 innings or more. Of those 20,000, eight produced a sub-1.00 ERA. That’s ERA, of course; ERA record holder Fernando Rodney allowed four unearned runs, which pushes him over 1 run per nine. Unearned runs also disqualify Dennis Eckersley (1990), and Bill Henry (1964), and Dennys Reyes (2006), and Chris Hammond (2005), and Eric O’Flaherty (2011), and Jonathan Papelbon (2006). That leaves one pitcher, out of 20,000, who threw 50 innings in a season while allowing less than one run per nine innings. And that guy (Rob Murphy) threw 50 1/3.
It’s practically impossible. So now think of Mariano Rivera’s postseason career as two seasons. The first goes from Oct 4, 1995 to Oct 18, 2001, and spans 71 innings. Rivera allows seven runs, six of them earned; 0.89 runs per nine, 0.76 ERA. The second “season” goes from Oct 21, 2001 to Oct 6, 2011, spans 70 ⅓ innings. Rivera allows six runs, five of them earned; 0.77 runs per nine, 0.64 ERA.
Nope. This is impossible. Rivera essentially does something twice that only one pitcher in history managed once (and just barely). He does it in mostly in an era of unrestrained offense. He does it exclusively against the best teams in the game. He does it in an environment that requires him to throw more than one inning in more than half his appearances, and to throw at least two innings in more than a third of them. Mariano Rivera’s postseason dominance is one of the best-known phenomena in baseball. It’s also probably underappreciated.
When the 1995 regular season ended, Mariano Rivera was an afterthought in the Yankees’ bullpen. He didn’t have a hold in his career. After moving to the bullpen, he had made eight relief appearances, all but one of them in mop-up work or long relief. And then the division series starts and, forced to use him in the 12th inning of Game 2, the Yankees watch Rivera go 3 ⅓ scoreless innings, striking out five, throwing 71 percent strikes, getting the win. Then, in Game 5, with the bases loaded and the game tied in the eighth inning, with John Wetteland having allowed seven runs in the series, with Steve Howe having allowed a 5.54 ERA in the second half, with David Cone walking in the tying run with his 147th pitch of the night, Buck Showalter is desperate. He brings in Rivera.
This is one of the best games of the 1990s for all sorts of reasons. Really, you have to watch this game, start to finish. This game’s got everything: A moose doing the worm; a seven-foot sasquatch coming in as a reliever; a rookie Alex Rodriguez grounding out with two on to end the bottom of the ninth; a Macklemore song; Vince Coleman; Andy Pettitte warming up in the bullpen; a prepubescent Derek Jeter on the bench; four to seven Hall of Famers, depending on the future; fans hexing the pitchers; 10 jacked homeless guys in old-fashioned bathing suits. But more than anything—even more than a rookie Alex Rodriguez grounding out with two on to end the bottom of the ninth—it’s got the birth of Mariano Rivera as a bullpen stopper. Remember: He has essentially never pitched in truly high-leverage relief before this. “It's not only tough to throw strikes right now. It's tough to breathe,” the announcer warns. But Buck Showalter brings him in, bases loaded, tie game, bottom of the eighth, and he strikes out Mike Blowers on three pitches. Showalter leaves him in there to start the ninth, even. In all, before an intentional walk to Ken Griffey, Jr., Rivera throws eight pitches, all of them fastballs, and check out this command (blue is the target):
There’s a hit in there—Vince Coleman took a single the other way—but the gas is real and he’s dotting the strike zone. It’s not just that Mariano Rivera was a phenomenal reliever who was also phenomenal in the postseason; he’s a phenomenal reliever who was born in the postseason. Dude was nothing until this series.
Here’s a clip of Rivera pitching in the postseason a few years later, only now he’s throwing a cutter. You might argue that the central narrative of Rivera’s career—that he’s the greatest one-pitch pitcher ever, thanks to a cutter that God randomly gave him one spring day—is wrong. After all, Rivera didn’t start throwing the cutter until mid-1997*. His best season was probably 1996. After that season, Sports Illustrated named his fastball the second-best pitch in the game. Even before the cutter, he was essentially a one-pitch pitcher throwing nothing but fastballs, he had the pinpoint control, he had a fastball with natural movement, and he had easy velocity that exploded on hitters. “Others may throw harder, but Rivera's 93-mph heater is the hardest to hit. Why? The Yankees righthander lulls batters with an easy, slow delivery. ‘And for some reason, his ball just explodes at the end,’ says Red Sox catcher Mike Stanley,” SI wrote.
But you might not argue that. The cutter was something else:
So that’s Ryan Klesko breaking three bats in one plate appearance. So many broken bats in Rivera's career. “Before the 2009 World Series, the Phillies' second baseman Chase Utley, a lefthanded hitter, custom-ordered a bat specifically to use against Rivera -- and he grounded into a double play with it.” Custom-ordered a bat specifically to use against Rivera! Other hitters, meanwhile, go the opposite direction: “Hitters have been known to use their batting practice bats against him rather than risking their gamers. ‘I admit I've thought about it,’ Gabe Gross says. ‘It's like when you have a long carry over water in golf. Do you drop that old ball in your bag or do you go ahead and hit the brand-new Titleist?’"
Rivera’s postseason performance is interesting partly because his strikeout rate plummets. In the regular season, he struck out 8.2 batters per nine innings; in the postseason, 7.0. But his BABIP drops from .265 to .219. You wonder, is it just luck that turned him from a regular season superstar into a postseason immortal? A hundred and forty innings isn’t, after all, that many innings. But then you see broken bat after broken bat, and you realize that DIPS can’t possibly apply to anything hit off a broken bat. Can’t possibly.
*The best detail about Rivera’s cutter is that he and his pitching coach spent a month trying to fix it; when the cut showed up while he was throwing before a game, it was an annoyance. Finally, they figured it wasn’t going anywhere, so he just accepted it.
Here are the greatest career postseason Win Probability Added totals since 1920, from Sean Forman at Baseball-Reference:
The best game of Rivera’s postseason career, by WPA, was his first, against Seattle, the three innings in extras. The next best was later, in 2003. It’s the game everybody remembers as the Aaron Boone game, but at the same time Boone entered the game Rivera did. He threw three shutout innings, 48 pitches. That’s the Aaron Boone game, but Boone’s WPA for the game was .356. Rivera’s was .432. It was the Mariano Rivera game, really.
It’s obviously hard to measure the greatness of a reliever against the greatness of a starter, mostly because starters go lots of innings and relievers throw only one. It’s hard even with Rivera, but no reliever of the generation was less of a one-inning guy than Rivera. Those three innings against Boston marked the third time in the series he threw at least two innings; he would throw at least two innings seven times in the 2003 postseason. In his career, 33 of his 96 postseason outings went two innings or longer. Not four outs or longer; six outs or longer.
Thirty-one of his postseason saves were four outs or more; the next highest in history is seven (Gossage). From 1997 to 2011, spanning Rivera’s career, the entire league had 45 postseason saves that went four outs or longer. In his multi-inning stints, covering 106 innings, he had a 0.51 ERA.
In his career, Rivera held batters to a .262 OBP. In the postseason he held them to a .212 OBP. The odds of that happening by chance alone are something like one in 200. The odds of his BABIP being that low by chance alone is around one in 50. It can’t just be luck.
"I saw him down in 2001," Jorge Posada told Sports Illustrated later. "He was a long time in front of his locker after the game was lost. We all went by.”
That’s the most famous failure of Rivera’s career. With a one-run lead in Game 7 of the World Series, Rivera allowed the tying run—in part because of his own throwing error on a sacrifice bunt attempt—then loaded the bases, then allowed a game-winning single by Luis Gonzalez.
About that error: Posada gives Rivera an excuse. “Do you remember that it rained a little bit? The ball was wet. He threw it, and it slipped out of his hand. Things you don't remember. They were trying to close the roof." It sounds convincing—they were trying to close the roof! Mechanical failure! Makes perfect sense!—but remember also that Arizona has that little strip of dirt between home plate and the mound. The ball never touched grass. Could it have become wet and slippery in the three bounces it took to get to Rivera? Posada knows better than we do. But it’s hard to accept.
Which sets up Luis Gonzalez. The first pitch is in on his hands, and he pops it foul; almost sounds like a broken bat. And here Tim McCarver has the greatest moment of his broadcasting career:
And that’s what happens. Five seconds later. Bat breaks, ball barely clears the infield. Romantics talk about how perfectly calibrated baseball is, but that’s not true at all: Luis Gonzalez gets a hit because he couldn’t hit it well enough to make an out. Rivera, better at doing one thing better than any pitcher in history, does that one thing, and it burns him. It’s nuts. Stupid broken bat. Probably sitting in the Hall of Fame now, and you wonder: was it one of Gonzalez’s gamers? Or did he bring his batting practice bat out there, resigned to the inevitable?