The Importance of Being Gagne


Last night, Eric Gagne's streak of 84 straight saves finally came to an end when he surrendered a two-run lead to the Diamondbacks.  It's a pretty impressive streak and certainly one players on the opposing teams were aware of.  Much like when Mariano Rivera or Trevor Hoffman or any of the other great closers came out of the bullpen during their peak years, the opposition knew the game was over.  But is his streak historic?  Will it stand for the rest of our lives and beyond?

Well, it certainly appears that it has some legs.  The closest previous streak was 54 straight saves set by Tom Gordon.  Gagne's exceeds that by 55% and the next closest ones are around half as long as Gagne's.  Streaks with that much distance over the closest competitors - like DiMaggio's consecutive games with a hit, or Cal Ripken's consecutive games played - usually stand the test of time.  DiMaggio's streak is 27% longer than the next longest.  Cal Ripken's is 24% longer than Lou Gehrig's, and Gehrig's was 63% longer than the next best.  DiMaggio's is now 63 years old and Gehrig's lasted more than 50 years.  So if the margin over the second best is the primary criteria for longevity, there's a good chance that this record could stick around a while.

Probably the most amazing thing about Gagne's streak wasn't the fact that he was able to sustain a lead of 1-3 runs 84 consecutive times, because with most good teams that margin will more often be three runs than one.  What makes Gagne's streak in particular so impressive is that in 35 of those games he had only a one-run advantage when he entered the game.  That's not a surprising statistic given how anemic the Dodger's offense has been the last three years, but it does add some magnitude to his achievement that he was able to get those last three outs with zero margin of error in over 40% of his opportunities.  

The big problem with Gagne's record is that it was set during an era that was particularly conducive to setting such a record and that era is still relatively young.  Baseball players have been playing in consecutive games and hitting in consecutive games ever since the game began.  That's almost 130 years of professional baseball in which to establish these kinds of streaks.  Over such a long period of time, there are bound to be some extreme examples like DiMaggio and Ripken.  And because baseball has been played for such a long time and in such a manner that these streaks can occur, we have a high degree of certainty that these kinds of things don't happen very often.

This is not true with Gagne's streak.  The one-inning closer is still a relatively new development.  Many people point to Tony LaRussa's use of Dennis Eckersley as being the starting point of the trend.  That dates all the way back to... 1988.  Baseball has been played this way for 16 years.  For all we know, Gagne's streak could be the equivalent of Babe Ruth's first 50+ home run season in the advent of the live ball.  Ruth had nearly doubled the record he had just broken the year before and many thought no one would ever come close to topping his 1920 total.  But Ruth bested it the very next season with 59, and 54 has been tied or topped 20 times since, including three times by Ruth himself and by three other players before he died in 1948.

If Gagne had set such a streak before LaRussa, when closers routinely pitched two, sometimes three innings then we'd have a much larger sample of performances by which to compare it to.  Even then, the save only became an official statistic in 1966.  So even if we ignore the way closers are now used, we're still looking at a stat and a style of baseball that has been around for less than 50 years.  Could Bruce Sutter or Goose Gossage or Dick Radatz or Rollie Fingers have established a comparable streak had they been allowed the luxury of pitching just one inning at a time and rarely entering a game with anyone on base?  Given the kind of years those guys had, it certainly seems conceivable.  Maybe even likely.  Not to take anything away from Gagne's achievement, but if it's possible that there were/are comparably talented pitchers, then Gagne isn't unique and therefore this isn't a unique streak.  He's just a great closer who pitches in an environment that is designed especially to showcase it.  This also means that there will be other guys like Gagne sooner than later.  Who knows, maybe the next great streak will be set by a guy who's pitching in the majors right now, but isn't closing yet... just like Gagne was when he started.

Gagne became a closer because he couldn't break into the 2002 Dodger rotation that was already over-loaded: Hideo Nomo, Kevin Brown, Andy Ashby, Odalis Perez, Omar Daal and Kaz Ishii.  In retrospect that doesn't sound like an overly impressive rotation, but they did finish with the 3rd best ERA in the majors, albeit helped by their home park.  Gagne had struggled as a starter because he had trouble fooling hitters the second time through the line-up and was put in the bullpen as much out of necessity as it was that his manager thought he'd be a great closer.  This gives us a clue as to who, or at least what kind of pitcher, might challenge and possibly break this streak. 

His success as a closer is largely attributable to the fact that he has 3 good pitches he can rely on.  If one isn't working, he always has two others he can throw.  What if the Dodgers had decided eight years earlier that Pedro Martinez should be a closer rather than a starter... could he have been as good as Gagne?  How about John Smoltz in Atlanta before he had all his arm surgeries?   How good of a closer could Josh Beckett be?  In no way am I suggesting that any starter can do what Gagne has done, but the three guys I cited in particular appear to have both the stuff and the mental toughness, that "closer's mentality" if you will, to come up big when the stakes are highest.  The problem with Gagne's streak is that more often than not, the best candidates to break it aren't given the opportunity; they are put into the rotation instead.  But if a situation were to arise where a starting pitcher with good stuff and control has to be put into the bullpen - either due to injury concerns, lack of endurance, no place in the rotation to put him, whatever - then it's certainly conceivable that he could be as successful as Gagne.  So the question isn't really "can it be broken", but "will it be broken".

Establishing and maintaining a streak that relies on 100% success in one-run games requires some luck.  The luck factor is reduced somewhat if the closer always gets to control the situation he comes into, i.e. beginning the inning with no one on base.  But luck is always a factor in those games.  One lucky swing or, as what happened in the game in which Gagne's streak was broken, one unlucky bounce and the streak is over.  But given a team that can limit the role luck plays by scoring lots of runs, where three run leads are far more common that one-run leads, and a closer who (like Gagne) was a guy with a starter's repertoire but one attribute lacking to stick in the rotation then it's not only possible that this record will be broken, but probable.  It probably won't happen soon, the way Mark McGwire's single season home run record lasted only three years, but it will be lucky to last 20 years.