October 31, 2011
Silly Goose: Mariano Rivera and the Myth of the Seven-Out Save
Believe it or not, most of our writers didn't enter the world sporting an @baseballprospectus.com address; with a few exceptions, they started out somewhere else. In an effort to up your reading pleasure while tipping our caps to some of the most illuminating work being done elsewhere on the internet, we'll be yielding the stage once a week to the best and brightest baseball writers, researchers and thinkers from outside of the BP umbrella. If you'd like to nominate a guest contributor (including yourself), please drop us a line.
Kevin Baker is a novelist and historian who is currently at work on a social history of New York City baseball, to be published by Pantheon.
I used to love watching Goose Gossage pitch. With that wonderful rising fastball he was one of the best examples ever of the pure power game. I remember a stretch in late August and September, 1980, when he retired 21 batters in a row, and he was throwing so hard that hitters for the Toronto Blue Jays literally seemed reluctant to step up to the plate—something I have never witnessed before or since.
When he was at his best, there was no real question of doing anything with his fastball. He seemed in complete control of the game, just playing catch out there. I can still recall sitting in a Manhattan dive called the Third Phase, watching a Monday night game in that same 1980 season when—in his second inning of work up at Fenway—the Goose gave up a hit and a scratch single to load the bases, but also struck out the side. He finished with a flourish, pumping three straight fastballs past Tony Perez, then leading the American League in RBI, to end the game. No one was going to beat him that night; the only doubt was whether he could rein in his speed enough to keep the ball over the plate.
That’s why I was happy to see him elected to the Hall of Fame, an honor he heartily deserved. Gossage’s approach to the Hall was as straightforward as his pitching style. He campaigned actively by pointing out how much harder relief pitchers were worked back in his day than now, and how much more difficult it was to run up impressive save totals.
His points were well-taken, and now that he’s in the Hall…I wish he would shut up.
Over the last few years, the Goose’s advocacy for the pitchers of his era has turned more and more into carping about the closers of today, and especially Mariano Rivera. This is usually followed by some boilerplate about what a great competitor Rivera is, apples and oranges, blah blah blah. But more and more, it’s become downright pissy. Worse yet, it’s begun to influence those highly impressionable young minds we call sportswriters and broadcasters.
“…when I pitched the ninth inning to save a three-run lead, coming in with no one on base, I felt guilty. I would go home and be embarrassed,” Gossage told Fox Sports online columnist Greg Couch last month. “Rivera is a great pitcher, but what he’s doing is easy. It really is.”
Let’s have a quick show of (liver-spotted) hands: What springs to mind when you hear Goose Gossage talking about how “easy” it is to get through an inning without giving up three runs?
George Brett hitting a three-run homer into the upper deck of Yankee Stadium to clinch the 1980 ALCS for the Royals? Yes, thank you, Yankees fans!
Can I get an “amen” from the Padres fans out there? Remember Kirk Gibson going deep to wrap up the 1984 World Series for the Tigers with a three-run blast…the inning after Lance Parrish had already homered off Gossage? You stay classy, San Diego!
That’s right, Rich Gossage’s two most famous moments in postseason history consist of him surrendering mammoth, three-run homers.
But wait, that’s not really fair. Sure, it wasn’t strictly the postseason, but it was Goose Gossage out there on the mound saving the “Bucky Dent game,” the 1978 playoff between the Yankees and the Red Sox.
In that game, the Goose was actually given a three-run lead at one point…and barely survived, surrendering hits or walks to six of the last eleven Red Sox he faced. Save for a terrific head fake by Lou Piniella in right field, Boston would probably have tied the game in the ninth.
Not so easy, holding on to a three-run lead.
Unfortunately, thanks to Goose, Fox’s Greg Couch joined all too many commentators in damning The Great One’s record-breaking 603 saves with faint praise.
Couch claims that he doesn’t want “to doubt the greatness of Rivera,” but that there is “no way of knowing” if he is the greatest closer ever, due to the faulty, “fabricated” statistic that is the “save.”
What would be a better one?
Well, Couch quotes approvingly a definition that’s been bandied about a lot recently. That is, a save of “seven outs or more”—or over two innings. Goose Gossage has 52 of these in the regular season, he informs us; Mariano Rivera…one, Trevor Hoffman, two. Optimally, what a real closer does, according to Couch and Goose, is to “come in during the seventh inning, bases loaded, one-run lead.”
“I used to love that,” says Gossage. “They used to use and abuse us, but think of the pressure. You couldn’t even let them put the ball in play.”
I’m sorry, but just when did we start handing out style points for degree of difficulty? This is baseball, not gymnastics or figure skating. The idea is to win. If Mr. Couch covered music, would he be sneering, “Nice concerto, Mr. Heifetz. But let’s see you play it while crossing a high wire—riding a tricycle?”
I’ll concede that there are plenty of problems with the current save statistic. And some day, in baseball’s equivalent of punctuated equilibrium, a manager will climb out of the antediluvian ooze and try using his best relief pitcher in the most critical moment of the game, whether or not that’s in the ninth inning. (Although this is expecting a lot of prescience from the poor manager and it still leaves that pesky ninth for someone to get through.)
But what the record shows is not that Mariano Rivera should be used more like Goose Gossage It’s that Goose Gossage should’ve been used more like Mariano Rivera.
The basic idea here is that a relief pitcher is a weapon, and like all weapons, it makes sense to use it as wisely and efficiently as possible. Someone—I think it was Roger Angell—compared the closer to the cavalry of Napoleonic era warfare, designed not to make foolhardy frontal assaults, but to exploit breaches in the line and turn an opening into a rout. I think that’s a pretty fair analogy. And when it comes right down to it, the weapon that was Goose Gossage was all too often sent charging into the guns, in acts of idiotic bravado.
Let’s examine first the mythology of the seven-out save—we’ll call it a “Supersave.”
Why seven outs and not, say, six, or nine? The whole idea seems at least as arbitrary and fabricated as the original save stat. For that matter, Gossage’s 52 Supersaves become a lot less impressive when you take into account the fact that he was a major-league pitcher for 22 seasons, and a reliever for 21 of them.
Thanks to the brilliant statistical work of Baseball Prospectus’ own Bradley Ankrom, we can report that the Goose was in fact only the master of the extended save for a few seasons, most of them near the beginning of that very long career.
He was at his best in 1975 when, as a 24-year-old hurler for a poor White Sox team, he converted 11 of 13 Supersave opportunities, throwing 141.2 innings. After flopping as a starter for the Sox the following year, he came back in 1977 to convert seven of nine Supersave chances, while throwing 133 innings.
Pretty spectacular. But this was clearly a young man’s game, and the Goose would not last at it. In 1978, at age 27, Gossage racked up six more Supersaves—but also blew six such chances. He had only two more good years at this sort of work—1980, when he converted nine of 11, and 1984, when he was six of seven.
After that, for the last nine years of his career, he racked up exactly four more saves of seven outs or more, while blowing two. Nor was the save rule always unkind to him; one of these “Supersaves” consisted of pitching four innings—beginning with an eight-run lead.
At the same time, the Goose would, according to my count, blow 25 Supersaves—or about one-third of all his opportunities, a ratio that would be unacceptable to most teams today.
Yet this was pretty much in keeping with Gossage’s entire career record. Looking it over—thanks again to Mr. Ankrom’s industry—the first thing that jumps out at you is just how many games Goose Gossage managed to lose, compared to all leading closers today.
Even in 1977-78, two of his very best years, spent pitching for a hard-hitting Pirates club that won 96 games and a World Champion Yankees team, he lost a total of 20 games coming out of the pen, and blew 22 saves—in other words, almost one-third of the 129 total games those two teams lost. Nor was this an anomaly. Throughout his career—spent largely with winning clubs—Gossage ran up double figures in blown saves in six of the 13 seasons when he was either his team’s primary closer or at least shared the role.
Mariano Rivera, by contrast, has never lost more than six games in any one season, in his 15 straight years as the Yankees’ closer and another as their set-up man. He has blown more than six saves only once, in 1997, his first year as a closer, when he gave it up nine times.
It’s a big reason why Mo’s failures are so memorable. He reached almost ridiculous heights of efficiency in 2008 and 2009, blowing one and two saves, respectively, out of a total of 86 opportunities. It’s why his lifetime percentage of saves is a mind-blowing, all-time high of 90 percent, while Gossage’s is only 73.5 percent.
Even in his heyday, the Goose routinely squandered between a quarter and a third of his save opportunities. In 1977, he managed to save only 72 percent of the leads he was sent in to preserve. In 1978, just 69 percent, in 1982, only about 77 percent; 1983, 63 percent; 1984, 69 percent; 1986, 66 percent, 1987, 65 percent; 1988, 56.5 percent…after which even unevolved managers decided they’d just as soon find more novel ways to lose games.
Goose’s best years in terms of the percentages of games he would win and save out of the pen? Well, unsurprisingly, these tended to be in the seasons when he was used the most judiciously, if only by accident.
Limited to 36 appearances and 58.1 innings in 1979, when Cliff Johnson broke his thumb in a showerroom brawl, the Goose saved a career-high 90 percent of his 21 save opportunities. In 1981, limited to just 32 appearances and 46.2 innings by an owners’ lockout, he was nearly untouchable, saving 87 percent of his opportunities (20 of 23), losing only two games, surrendering just 22 hits, two home runs, and 14 walks, and compiling an ERA of 0.77.
You’d think somebody would have noticed just how much better the Goose did with more rest, and someone did—namely Dick Howser, the most perspicacious of all Gossage’s managers during this period. Howser limited him to “just” 99 innings in 1980, a full season…in which the Goose went 6-2 and saved 33 of 37 games, including nine Supersaves.
Unfortunately for Gossage and the Yankees, Howser was fired after that one season (in no small part because of Goose’s insistence that he could throw a fastball past George Brett, a concept he never would rid himself of). Then in his thirties, the Goose was once again given over to the care and maintenance of managers who refused to make much allowance for his age or condition.
I thought I remembered him being used in a particularly egregious fashion in 1983, which marked one of Billy Martin’s later and uglier incarnations as Yankees manager. Thanks once more to Bradley Ankrom, I was able to check if this was really so. Sure enough, it was.
To be sure, Gossage’s overall innings totals remained more limited. But day to day, his use seemed more mindless than ever. Here, for instance, was the Goose’s first appearance: entering the second game of the season with one out in the bottom of the eighth inning and the Yankees trailing Seattle, 6-2.
Why, exactly? To beat the spread? Because Martin had dinner reservations at the Space Needle?
This set a pattern. Gossage would enter seven more games that year with his team trailing in the seventh inning or later, four of which they were losing by more than one run. Only once would they rally to win.
If there ever was a manager willing to indulge a pitcher’s desire to get out on the mound and stay there, it was Billy Martin. Throughout 1983, Gossage—now 32 years old—would attempt six more Supersaves but convert only one of them.
It wasn’t that Goose was finished, or close to it. He still threw hard, still allowed just 82 hits and 25 walks in 87.1 innings; still struck out 90, won 13 games, and saved 22 more while compiling a 2.27 ERA. But he did blow 13 saves—in a year the Yankees finished seven games out of first—and clearly seemed less able to get outs when he wanted them.
This leads us to the other part of Gossage and Couch’s blather about what a “real” save should look like. That is, how “thrilled” Goose always was to come in with runners on base, particularly “bases loaded in the seventh inning.”
Throughout that 1983 season, it struck me that Gossage—now an older pitcher who probably required more time to get ready—gave up more hits and walks than ever to the first batter or two he faced, then seemed to settle down. Yet Martin almost never seemed to use him to start an inning.
The record confirms this, too. Of his 57 appearances in 1983, just four of them started an inning. In all but four of these, there was already at least one man on base. As for the “thrill” of coming in with the bases loaded…it was limited to the other team’s dugout. Goose faced that situation exactly four times all year—and every time, he surrendered hits that scored one or two runs.
Over the entire course of his career, Gossage would enter regular-season games with the bases loaded 48 times—far more than any closer, or even set-up man, is likely to do today. In those games he performed well…but slightly worse than he did in the rest of his appearances, compiling 16 saves and a win but also blowing eight saves.
The whole notion of using Gossage this way in the first place is baffling. Why limit your big power pitcher by constantly making him pitch out of the stretch? Never mind pitch counts; do you really have so little idea of when your starter (or another reliever) is running out of steam?
Goose coming in mid-inning in 1983 suffered all five of his losses and all 13 of his blown saves, compiling a 2.32 ERA. His four appearances starting an inning are too small to be statistically meaningful, but it is interesting to note that while he gave up five hits and two walks in those six innings, he surrendered just one run and saved a game.
The moral here is, take care of your tools—or your weapons—and they’ll take care of you. “Abused,” as he claimed, for most of his career, Goose was in serious decline as a relief pitcher by his early thirties…whereas Mariano continues as one of the very best relievers in the game at 42.
The abuse of the Goose was particularly unnecessary when you consider the fact that he played most of his career with very capable bullpen mates—Kent Tekulve and Terry Forster on the Pirates; Dick Tidrow, Sparky Lyle, Ron Davis, and Dave Righetti on the Yankees; Craig Lefferts on the Padres, etc. It wasn’t a case of desperate managers trying to eke out an extra win with no one else to turn to.
Goose and Couch assert that when Gossage appeared on the scene, “the bullpen was just a junk pile of washed-up starters who couldn’t throw nine innings anymore, or guys who weren’t quite good enough to start.” But like so much else of which they speak, it ain’t necessarily so.
Managers had been dabbling intermittently with the idea of specialty relievers since the days of John McGraw, and by the time Goose Gossage came up in 1972, there had been quite a few good ones. That is, men who were outstanding pitchers, expected from an early age to throw mostly or solely in relief: Joe Page, Hoyt Wilhelm, Clem Labine, Ryne Duren, Elroy Face, Lindy McDaniel, Dick Radatz, Luis Arroyo, Ron Perranoski, Pedro Ramos, Phil Regan, Wayne Granger, Clay Carroll, Dave Guisti, Tug McGraw, Sparky Lyle, Rollie Fingers, to name just a few.
These closers threw different pitches and had different backgrounds, but they all had one thing in common: either they burned out after a few wildly successful seasons, or they suffered mysterious “off years” throughout their careers.
The answer to the “mystery” was, of course, that they were overworked. Managers were so thrilled by this new weapon, one that would preserve their every lead—or so it seemed—that they couldn’t help themselves from overusing it.
“Never save a pitcher for tomorrow. Tomorrow it might rain,” was Leo Durocher’s famous adage, and it became their watchword, even though it was never supposed to apply to relievers throwing on a daily basis.
Goose Gossage had enough arm strength, enough bulk, and enough mental toughness to endure much longer than this generation of abused pitchers, and he deserves all the accolades he’s won. But too often, his remarkable gifts were wasted—the baseball equivalent of blindly throwing cavalry at artillery batteries (a tactic that would be immortalized as, “The Charge of the Light Brigade”).
The very idea of a relief pitcher is that of someone who has one extraordinary pitch—and one only. If they have more, they are being wasted as a reliever and should be moved into the starting rotation. Overusing a reliever not only weakens his arm over the long run of the season or his career, it also provides hitters with the opportunity to adjust to his specialty pitch—and given enough opportunities, at least in the same game, major-league hitters will adjust to any pitch you can throw.
Here’s one more statistic to throw the trade-off between save and Supersave into full relief: over the course of his career, Goose Gossage threw some 600 more innings than Mariano Rivera (although 224 of these came in Goose’s one season used almost exclusively as a starter, while River threw only 67.1 in 1995, when the Yanks gave him 10 starts).
However, by the year he turned 35, Gossage was throwing fewer innings per season than Rivera was for every year he was the same age. And throughout their respective careers, Mo almost always logged more appearances.
So the cost of Goose’s Supersaves was more blown saves, fewer appearances, and a foreshortened career. Tell me again: Why is it better to have an athlete try to do something he can’t do as well when he overextends himself, especially when that something makes him less available to his team and less effective over the course of his career?
Still, even if you could convince Couch and Gossage that it’s infinitely more rational to use relievers the way they are used today, rather than in romantic times of yore, they would argue that this only confirms Goose’s opinion that he was “abused” by his managers. It doesn’t answer the main thing they claim to want to know, which is: Who’s better? Who’s the best?
In this sense, Couch is right—it’s arguing apples and oranges, and ultimately unknowable.
To wonder if Mariano could have carried Goose’s old pitching load without breaking down is about as useful as wondering if Roy Halladay could have pitched four hundred innings a year, the way Joe McGinnity and Happy Jack Chesbro did over a hundred years ago. (The answer is…yes, probably, if you started all the games in the late afternoon and handed Mr. Halladay a ball he could spit on, rub any sort of gunk on, and not replace until it had become a wobbly, soggy, gray mess.)
Certainly, Rivera doesn’t have Gossage’s bulk…although as a young set-up man back in 1996, he did throw 107 innings, firing almost entirely fastballs that moved as much as the Goose’s ever did. Could Mo have kept up that pace year in and year out, even with his famous cutter? Who knows? Gossage certainly didn’t; his effectiveness falling off as dramatically as his yearly innings by his early-to-mid thirties.
Of course, what Couch and Gossage are driving at is how good Goose would have been in the modern era of relief pitching, free to just come in at the start of the ninth, with no one on base. I suspect he would have been spectacularly successful…although again, who knows?
Rivera has pitched, after all, almost entirely in an era of bandbox ballparks and souped-up sluggers. While it would not surprise me to learn that any ballplayer today has used performance-enhancing drugs, it seems unlikely that Mo has ever done so, considering the course of his career, his body type, his declining velocity over the years, and his religious convictions.
This would mean that he has played his entire career with a handicap unlike anything that Goose was ever subjected to. Would a fastball pitcher with a wild streak, stubbornly maintaining that he could throw his ball past anyone, anytime, really have fared so well in an age of steroidal hitters who specialize in working pitch counts? Just how many of all those impressive Gossage innings included popping up bandy-legged shortstops on the first pitch, or getting batters to fly out to the far reaches of stadiums built mainly for football?
Maybe Gossage would’ve made adjustments. The great ones usually do…although it’s hard not to forget the famous footage of the Goose talking Dick Williams out of making him walk Kirk Gibson intentionally in the eighth inning of that 1984 World Series finale, while over in the other dugout, Sparky Anderson bet Gibson ten dollars that Gossage would pitch to him. Team be damned—it was all about how hard the Goose could throw.
There is one further indication of how Mariano Rivera might have fared in the Gossage era, and that’s his prodigious postseason record. During the regular season, along with those 603 saves, Mo has a 75-57 record and a lifetime ERA of 2.21—the best ever compiled in the live-ball era, depending on how you want to measure it—along with just 934 hits and 275 walks in 1,211.1 innings, and 1,111 strikeouts, figures so gaudy they’re almost absurd.
But in his postseason appearances, which by now have amounted to an extra season, or maybe two seasons, of pitching, Rivera is even better…much better. Against the best teams in baseball, with everything at stake, he’s run up an 8-1 record, with 42 saves in 47 attempts, allowing 21 walks and 81 hits against 110 strikeouts in 141 innings and compiling an ERA of 0.70.
Yet the most salient fact about all those playoff games is how dramatically Rivera changed his usual pitching habits in them.
You want seven outs? Mariano has provided four such appearances in the postseason; in none of them did he allow a run or an inherited runner to score. They included a couple of the most memorable playoff games in history; his coming out, a 3.1-inning victory over Seattle in the 1995 American League Division Series, and the three unforgettable innings he pitched to win the “Aaron Boone game” against Boston in the ALCS finale in 2003.
You want two-inning appearances? Rivera has run up 29 of those in the postseason, garnering four wins, 14 saves, and three holds.
You want more than one inning? Mo has another 24 one-inning-plus playoff appearances to his credit, earning another 16 saves and a hold.
In other words, 57 of Rivera’s 96 playoff appearances have been for more than one inning. In them, he has run up half of his eight postseason wins and almost three-quarters of his 42 postseason saves.
You want inherited runners? In nearly a third of his playoff appearances—30 out of 96—Rivera has entered the game with runners on base; a total of 48 of them, 14 of them on second, 11 on third. He has prevented all but eight of them—or one-sixth—from scoring.
And yes, he’s come into postseason games with the bases loaded. He did it in his first year in the playoffs against Seattle, age 25, and he did it just this fall, in the ALDS against Detroit, age nearly 42. In each case, he struck out the next batter to end the inning.
Gossage’s record in the playoffs, while much more abbreviated, is also outstanding. In 19 appearances, he had two wins and eight saves, with an ERA of 2.87, and 21 hits, 10 walks, and 29 strikeouts in 31.1 innings. He inherited runners on six different occasions, ten in all, and allowed only one to score. In 1981, easily his best postseason, the well-rested Goose did indeed come into the seventh inning of a game the Yankees were winning 1-0, in the second game of the special ALDS that year, and retired Robin Yount and Cecil Cooper in a bravura performance.
That was the only time he ever did it. But he also managed to blow three of eleven save opportunities in his postseason career, as well as effectively taking San Diego out of that last game of the 1984 World Series. In 1980, he wouldn’t have even had to face George Brett had he not given up a two-out single to Royals’ shortstop U.L. Washington (who’s “embarrassed” now?).
Yet somehow these blips have dropped from most sportswriters’ memories, while one after another felt obliged to bring up the fact that Rivera “had his failures” in the postseason, almost as if he were the Greg Norman of relief pitching.
It’s instructive to take a look at those “failures.” Mariano has blown all of five saves in the postseason. One of these was Sandy Alomar’s famous, opposite-field home run in the 1997 ALDS that barely cleared the right-field fence—and only tied Game Five of that series.
One was the even more famous Yankees meltdown in the seventh game of the 2001 World Series, where Mo—after pitching a scoreless eighth inning—was victimized more by the fielding of himself and his teammates than his pitching (and when Joe Torre foolishly decided to move his infield in, behind a pitcher who specialized in weak pop-ups to the near outfield).
The other three came within a space of 13 days in the 2004 playoffs, after a season in which Rivera had set career marks in appearances and saves, with 69 and 53, respectively. In the midst of this period, he had to make a hurried flight to Panama and back to deal with the tragic death of his cousins in a pool accident.
Nonetheless, Joe “Breaker of Pitchers” Torre decided to call on Rivera seven times in this span, including three two-inning stints and three more of five or six outs. One of the blown saves was when Rivera gave up the tying runs to the Twins in the ALDS, in a game the Yanks later won in extra innings.
The other two, of course, came against Boston. One was Game Five of the ALCS, in which Mo gave up a sacrifice fly to tie the game after coming in men on first and third—something that tells you most of what you need to know about the problems with the save statistic. The other was the famous “Dave Roberts game”—although here again, Rivera, in his second inning of work, gave up the tying but not the winning run.
In other words, in 96 tries, Mariano River has never given up an earned run that lost a ballgame in the postseason. Used as he was “supposed” to be used—that is, the way he was used through most of the season, brought in at the start of the ninth inning—he has never surrendered a lead in he postseason, period. The only postseason contest where Rivera was really even hit hard was Game Two of the 2000 World Series against the Mets when, rushed in to save a floundering Jeff Nelson, he gave up a two-run homer to Jay Payton, and nearly another one to Todd Zeile. The Mets almost broke through—almost.
Throughout his career, Mariano Rivera has been the most brilliant of weapons, a stiletto expertly applied to win a great many ballgames with his one, unhittable pitch. What he has done is unique in the history not just of baseball, but all athletics: appearing for a decade-and-a-half, only when the game is on the line, and succeeding nine times out of ten in preserving victory. No other athlete in a team sport has ever performed so consistently under pressure.
Yet when jerked out of the security of his usual role and used in a very different role—when his managers have tried to use the stiletto as a meat axe—he has actually picked up his game. Taxed beyond his usual endurance, at the end of a long and wearing season, and against the best teams and hitters in the game…he has performed better than ever.
If they want to contribute something, Goose Gossage and Greg Couch should take up the worthy cause of getting some of Goose’s other contemporaries into the Hall with him (Sparky Lyle, anyone?). In trying to denigrate what Mariano Rivera has accomplished, they only make themselves look foolish.