A History of the LOOGY:  Part Two

Last time we explored the murky depths of the origin of the LOOGY. We discovered its early-1960s crawl from the sea of pitcher-deployment patterns, and followed its long slow evolution — including a bout with near-extinction in the 1970s! — and left it following 1986, a year in which the LOOGY appeared to be on its most solid footing yet, but still anything but fully-established. Remember that there had been a grand total of ten LOOGYs deployed over the 1986 season, the most yet.

And, oh yes: let’s review our working definition of a (L)eft-handed (O)ne (O)ut (G)u(Y) season:

– A southpaw who appears in at least 20 games
– With fewer than 1.20 innings per appearance
– And fewer than .20 Saves per appearance

Okay. So, based on that definition, what happened in 1987?

Boom! Goes the LOOGY!

What happened was a veritable ka-boom. Twenty LOOGY seasons graced (cratered?) the 1987 landscape; twice the record-setting 1986 level. Eighteen of the 26 major league teams deployed LOOGYs in 1987; two of them deployed two. Oakland Athletics’ manager Tony LaRussa, who hadn’t invented the LOOGY, but had been something of an instigator in its mid-1980s revival, actually wasn’t a particular ringleader in 1987; LaRussa contributed just one among the twenty. The LOOGY had suddenly crossed the threshold from fringe indulgence to broadly accepted alternative.

There was regression in 1988, down to eleven LOOGY seasons from nine teams. But over the next few years, it became clear that ’87, while a high point, hadn’t been a fluke: there were 17 LOOGY seasons in 1989, 15 in 1990, and 18 in 1991. The LOOGY was solidly established. And then came 1992.

In 1992 the LOOGY population soared to 29. Only five teams in 1992 didn’t deploy at least one LOOGY, and one team — the Houston Astros, managed by Art Howe — used three pitchers (Al Osuna, Rob Murphy, and Rob Mallicoat) in the LOOGY mode, the first time that had ever happened.

Since 1992, the LOOGY has never looked back. How about at this point we just display the aggregate LOOGY totals for the entire period of 1987-2004:

Year   #  Per Team  Avg G  Avg IP  IP/G  Avg Sv   Sv/G
1987  20    0.77      45     48    1.07     2    0.054
1988  11    0.42      45     46    1.01     3    0.072
1989  17    0.65      49     53    1.09     2    0.050
1990  15    0.58      47     46    0.99     3    0.064
1991  18    0.69      55     55    1.00     3    0.060
1992  29    1.12      52     47    0.91     2    0.043
1993  39    1.39      46     42    0.92     1    0.032
1994  36    1.29      40     37    0.92     1    0.025
1995  44    1.57      38     33    0.85     1    0.020
1996  45    1.61      45     39    0.88     1    0.030
1997  42    1.50      53     45    0.85     1    0.024
1998  46    1.53      48     41    0.86     1    0.022
1999  46    1.53      51     44    0.86     1    0.015
2000  46    1.53      50     44    0.87     1    0.025
2001  51    1.70      55     47    0.85     2    0.029
2002  57    1.90      47     40    0.85     1    0.020
2003  51    1.70      52     43    0.83     1    0.018
2004  52    1.73      51     43    0.84     1    0.011

Two patterns have been abundantly clear over this period. The first is the dramatic increase in the magnitude of LOOGY deployment. The 1992 usage level seemed enormous at the time, but immediately became dwarfed by what happened in subsequent seasons. In the space of fifteen years, the LOOGY has gone from something that a typical bullpen might or might not include, to something found in every bullpen, indeed at the frequency of nearly two per team. The rate of increase in LOOGY deployment has slowed — it could scarcely have continued at the reckless pace of the early 1990s — but it isn’t obvious that it has yet finally plateaued, and certainly LOOGY usage has shown no signs of returning to anything resembling previous levels.

The second pattern is in the stints of these increasingly-used LOOGYs, and that pattern is shorter and shorter. The average innings worked per game of the fifty-something LOOGYs in each of 2001 through 2004 has been between 0.83 and 0.85; that was a rarified hyper-specialized mode achieved by a total of just 14 LOOGYs in history before 1987.

So Wait a Minute

This raises an issue. It’s more than clear that a fundamental revolution has occurred in the rate of LOOGY usage. But the simultaneous ramp-up in the intensity of specialization of these LOOGYs forces us to acknowledge that, in the current era, a LOOGY isn’t a LOOGY isn’t a LOOGY. The stint-length criterion we’ve been employing to identify LOOGYs — under 1.2 innings-per-appearance — was useful in selecting them out of the broader bullpen population into the mid-1990s, but since then, it’s increasingly necessary to narrow our scope of inquiry, in order to focus on just the most specialized of lefty relievers.

So let’s take another pass at the LOOGY population since 1987. This time let’s look only at those guys who last week we described as “Hard-Core LOOGYs”: those with at least 20 games, fewer than .20 Saves/Game, and less than 1.00 Inning/Game.

Hard-Core LOOGYs (Or, The Legacy of Jack Spring)

Year   #  Per Team  Avg G  Avg IP  IP/G  Avg Sv   Sv/G
1987   4    0.15      38     34    0.90     1    0.020
1988   4    0.15      42     38    0.90     4    0.083
1989   3    0.12      41     37    0.90     1    0.016
1990   6    0.23      47     39    0.85     3    0.053
1991   8    0.31      43     36    0.83     1    0.023
1992  19    0.73      57     48    0.83     3    0.044
1993  25    0.89      44     36    0.81     1    0.022
1994  23    0.82      39     31    0.81     1    0.020
1995  37    1.32      39     31    0.81     1    0.021
1996  33    1.18      46     38    0.82     1    0.032
1997  31    1.11      55     43    0.79     1    0.026
1998  33    1.10      49     38    0.78     1    0.015
1999  32    1.07      54     42    0.79     1    0.014
2000  36    1.20      50     41    0.81     1    0.023
2001  45    1.50      54     45    0.82     1    0.027
2002  46    1.53      47     38    0.80     1    0.022
2003  42    1.40      53     42    0.79     1    0.013
2004  43    1.43      52     41    0.79     0    0.009

The Hard-Core LOOGY was still a true rarity in the late 1980s. But usage suddenly exploded beginning in 1992, and shows little sign of slowing down. Indeed, the proportion of LOOGYs in the 1.00-to-1.19 innings/game category has increasingly shrunk. The type of pitcher who was, until a little over a decade ago, a special case among the LOOGY population — fairly considered a “hard-core” LOOGY — has become in fact the normal LOOGY. Such is the measure of how dramatically bullpen usage has changed in a short time.

So if in recent seasons the Hard-Core LOOGY has gone mainstream, who is the specialist-among-specialists today? For that, we have to select for the type of usage pattern that last week we described as “Really Hard-Core LOOGY:” these are the LOOGYs with fewer than 0.80 innings per appearance.

Really Hard-Core LOOGYs (Or, the Legacy of Bill Henry)

Year   #  Per Team  Avg G  Avg IP  IP/G  Avg Sv   Sv/G
1987   0    0.00       0      0    0.00     0    0.000
1988   1    0.04      39     31    0.79     2    0.051
1989   0    0.00       0      0    0.00     0    0.000
1990   1    0.04      37     26    0.70     1    0.027
1991   2    0.08      46     29    0.63     1    0.022
1992   7    0.27      60     40    0.67     2    0.033
1993   9    0.32      46     30    0.65     1    0.017
1994   8    0.29      39     27    0.69     1    0.022
1995  11    0.39      43     29    0.69     1    0.021
1996  13    0.46      52     37    0.72     2    0.031
1997  14    0.50      63     43    0.69     1    0.016
1998  18    0.60      46     31    0.66     1    0.014
1999  17    0.57      51     34    0.67     1    0.013
2000  15    0.50      53     36    0.69     0    0.005
2001  15    0.50      51     34    0.67     1    0.010
2002  21    0.70      42     28    0.67     1    0.018
2003  19    0.63      56     39    0.69     1    0.011
2004  20    0.67      56     38    0.68     1    0.009

Finally we reveal the avant-garde. As in the patterns of the larger populations, 1992 was the year of sudden emergence from the shadows for this most extreme usage pattern. And as in the patterns of the larger populations, the sudden emergence of 1992 has been followed by sustained growth, and it would be completely premature for us at this point to conclude that its growth is finished.

Who are these Really Hard-Core LOOGYs of our modern age? Early-to-mid-90s examples were John Candelaria, Tony Fossas, and Paul Assenmacher; more recent specimens include Mike Holtz, Jason Christiansen, and Ricardo Rincon. The most prominent ones, who have really spanned the entire era, have been Jesse Orosco, Dan Plesac, and, of course, His Majesty, the Big Kahuna, the King of LOOGYs: Mike Myers.

Let’s Take Stock

So where are we, then? Why don’t we review a couple of the questions we asked at the beginning of last week’s piece? Was the LOOGY:

Invented by Tony LaRussa?

Our examination of the development of this phenomenon over the past 40-something years clearly demonstrates that, for whatever else we might credit or blame TLR, invention of the LOOGY is not among it. The very Kansas City/Oakland Athletics organization for which LaRussa signed and played as a Bonus Baby infield prospect in the 1960s/early 1970s was experimenting with LOOGYs in that era: Alpha LOOGY Leo Kiely in 1960, Don Mossi in 1965, Joe Grzenda in 1966, and Paul Lindblad in 1968 and 1970. We might infer that LaRussa took note of this, and was influenced by it, but by no means is it anything close to the case that LaRussa invented the LOOGY.

But LaRussa has been the single manager most prominent among the few who stimulated the LOOGY revival in the 1980s, and among the many who led its explosion in the 1990s. Of all the managers involved in the long and still-unfolding story of the LOOGY, none has more significance than Tony LaRussa.

Was the LOOGY:

First embodied in Tony Fossas?

Obviously not. But Fossas was unquestionably a very major figure in the history of the LOOGY. Fossas spent all or part of 12 separate seasons (1988 through 1999) in the major leagues, appearing in 567 games, and not only did he never make a start, he was never at any point deployed in any manner other than a LOOGY, and for the vast majority of that time he was Hard-Core. In fact Fossas spent the majority of his major league career as a Really Hard-Core LOOGY. He was the first significant pitcher with such a career profile.

Those were our questions of fact. How about the matters of opinion we raised at the outset last week?

Is the LOOGY properly perceived as:

Scourge of the modern era? Bizarre mutation metastasizing in today’s bloated, hyper-specialized bullpen?

Or perhaps … clever innovation by progressive managers? Sensible adaptation of resources to meet contemporary challenges?

Well, obviously the answer is that the LOOGY phenomenon can’t be properly understood in isolation. No LOOGY is an island! The LOOGY boom has been a startling development, no doubt, but it is only one element in the set of interdependent dynamics of bullpen construction and deployment. To understand the revolution in the usage pattern of left-handed relief pitchers, it’s necessary to see it in relation to the corresponding revolution in the usage pattern of all other relief pitchers — and most especially, the usage pattern of the bullpen’s top reliever, the role that used to be called the Ace Reliever, and is now called the Closer.

So that’s what we’ll do. Next week we’ll present the historical data illuminating the dramatic changes in how teams have come to get the ball to the key relief pitcher. We’ll assess the wisdom of the LOOGY revolution in that context.

Keeping These Facts in Mind

But we will leave this week’s installment with two factual observations regarding LOOGYs.

1. A basic precept of the rules of baseball has a huge bearing here: a relief pitcher entering the game is required to face at least one batter to the conclusion of the at-bat, but the offensive team is allowed to pinch-hit at will. What this means is that, while a LOOGY is almost always brought in to face a left-handed batter, the offense frequently responds by pinch-hitting a right-handed batter. (Another basic fact of life is relevant here too: most batters, like most people, are right-handed.) What this means is that LOOGYs in practice don’t actually face only left-handed batters. Even the most extreme of hard-core LOOGYs face a significant proportion of righties, and overall, most LOOGYs face right-handed batters most of the time. (Mike Myers, the hardest of Hard-Core LOOGYs, with 0.61 innings/appearance over his 684-game career through 2004, has faced 48% right-handed batters, and the percentage is much higher than that for nearly every other LOOGY.)

It is the case that no category of southpaw pitchers enjoys the platoon advantage more frequently than LOOGYs. But it must be acknowledged that even though the LOOGY is a role designed entirely as a means of exploiting the platoon advantage, it does not do so, to a very great extent of the time.

2. LOOGYs as a group are not especially effective pitchers, as measured by ERA+ or WHIP. Clearly there are two key factors contributing to this: one is the selection bias determining the LOOGY population; the best left-handed pitchers are not deployed as LOOGYs. The other is the fact that even though LOOGYs face a high proportion of left-handed batters, a high proportion of those left-handed batters are the elite left-handed batters in baseball.

So, even though their usage pattern provides them with the best possible conditions — high proportion of platoon advantage batters faced, and extremely short stints, necessitating no pacing and exacting no in-game fatigue — it would be unrealistic of us to expect LOOGYs to be really shutting their opponents down. Nonetheless the bottom line remains: LOOGYs don’t really shut their opponents down. Even the very best LOOGYs, the elite deployed in the most extreme hard-core pattern — the Myers, Plesac, Orosco class — don’t produce rate stats that compare with those of the elite pitchers deployed in the more challenging roles of Closing or Starting. And, especially as the number of pitchers deployed as LOOGYs mushrooms, the typical LOOGY is less and less accomplished, resembling an elite pitcher less and less. The LOOGY is a class of pitcher deployed in a very high proportion of high-leverage situations, but without a particularly impressive record of effectiveness.

1987-2004 Really Hard-Core LOOGY Fun Facts

Total Really Hard-Core LOOGY Seasons, 1987-2004: 191

Most Really Hard-Core LOOGY Seasons Pitched, 1987-2004:

1. 10, Mike Myers (1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004 SEA, 2004 BOS)
2. 8, Dan Plesac (1997, 1998, 1999 TOR, 1999 ARI, 2000, 2001, 2002, 2003)
3T. 7, Tony Fossas (1992, 1993, 1994, 1995, 1996, 1997, 1998)
3T. 7, Jesse Orosco (1992, 1995, 1997, 1999, 2001, 2002, 2003)
5. 6, Mike Holtz (1997, 1998, 1999, 2000, 2001, 2002)

Most Really Hard-Core LOOGY Seasons Managed, 1987-2004:

1. 13, Tony LaRussa (1992 Honeycutt, 1992 Horsman, 1993 Horsman, 1994 Leiper, 1996 Fossas, 1996 Honeycutt, 1997 Fossas, 1998 Painter, 1999 Radinsky, 2000 Christiansen, 2001 Christiansen, 2004 King, 2004 Kline)
2. 9, Dusty Baker (1994 Frey, 1996 Creek, 1996 Poole, 1997 Poole, 2001 Christiansen, 2002 Eyre, 2003 Guthrie, 2004 Mercker, 2004 Remlinger)
3T. 8, Mike Hargrove (1997 Assenmacher, 1997 Morman, 1998 Assenmacher, 1998 Morman, 1999 Assenmacher, 1999 Rincon, 2003 Groom, 2003 Ryan)
3T. 8, Jim Riggleman (1995 Casian, 1995 Nabholz, 1996 Casian, 1996 Patterson, 1997 Patterson, 1998 Heredia, 1998 Patterson, 1999 Heredia)
5T. 7, Lou Piniella (1993 Plantenberg, 1995 Guetterman, 1998 Fossas, 1998 McCarthy, 2002 Creek, 2003 Malaska, 2003 Venafro)
5T. 7, Joe Torre (1991 McClure, 1992 McClure, *1995 Fossas, 1996 Howe, 1996 Polley, 1998 Lloyd, 2000 Choate)

* Managed partial season

Fewest Innings/Appearance, 1987-2004:

1. 0.46, Jesse Orosco 2001
2. 0.47, Rich Rodriguez 2002
3. 0.476, Jason Christiansen (STL) 2000
4. 0.478, Tony Fossas (SEA) 1998
5. 0.482, Jesse Orosco 2002

Fewest Innings/Appearance With At Least 40 Appearances, 1987-2004:

1. 0.48, Jesse Orosco 2002
2. 0.49, Jesse Orosco 1999
3T. 0.50, John Candelaria 1992
3T. 0.50, Tony Fossas 1992
5. 0.54, Mike Myers 2002

Most Really Hard-Core LOOGY Appearances, 1987-2004:

1T. 88, Mike Myers 1997
1T. 88, Sean Runyan 1998
3. 86, Ray King 2004
4T. 83, Scott Eyre 2004
4T. 83, Mike Myers 1996
4T. 83, Kelly Wunsch 2000

Most Really Hard-Core Appearances With Zero Saves, 1987-2004:

1. 86, Ray King 2004
2T. 80, Ray King 2003
2T. 80, Tom Martin 2003
4T. 76, Buddy Groom 1999
4T. 76, Bob Patterson 1997
4T. 76, B.J. Ryan 2003