Relief pitchers facing one batter
Posted by Andy on January 20, 2010
(For users new to this blog or the Baseball-Reference.com Play Index, please see the bottom of this post for a short tutorial on how the data in the graph below was generated.)
Let's take a look at relief pitchers brought in to face just one batter. Unlike my recent series of posts dealing with duration of relief appearances, the data in this post considers all relief appearances regardless of whether they were in save situations.
It's generally been accepted that use of specialty relievers, specifically guys who come in just to face one batter, usually a left-handed batter and a left-handed pitcher, has increased recently in baseball. Rob Neyer coined the term LOOGY, for a left-handed one-out guy.
Let's look at the data and see.
First let me explain what this graph shows. Let's start with the black line. It shows the fraction of all relief appearances that last just one batter. For all these data, the outcome of the plate appearance is not considered--could have been a strikeout, a home run, a ground-into-double-play, or anything else. The black line above shows that the fraction of relief appearances lasting just one batter was roughly constant at 6-7% from the mid 1950s to about 1990. It seems to have peaked around 1970 at 7.5% and was as low as 5% some years in that span but it was generally pretty flat. In 1992 there was a sudden jump to almost 9%. More on this as we look at the rest of the data.
The other 4 lines on the chart--LH, RH, AL, and NL--break out the data for each subcategory. The LH line, for example, shows the fraction of left-handed relief appearances that lasted one batter. (Do not confuse this to mean the fraction of all relief appearances that were by left-handers.)
We see, then, that left-handed one-batter relief appearances were roughly constant at about 10% of all left-handed relief appearances until 1992, when it suddenly jumped up to 15%. The figure has hovered around 16% ever since. Among right-handers, there was no significant change in 1992 and just a small increase in 1993. Overall, right-handed relief appearances have remained quite flat at just 5% lasting one batter.
The splits among AL and NL show that the increase in 1992 happened almost entirely in the AL. One-batter appearances jumped from 1991 to 1992 in the AL from about 7% to about 10%. (The reason why this jump is not numerically as large as the left-hander jump is because the AL numbers consider both LHP and RHP.) We see that the NL numbers also went up slightly in 1992 but not any more up or down than in an average year over the entire time period.
So it seems that, yes, coincident with the Steroids Era in 1992, the use of the LOOGY did indeed become significantly more common, at least in the AL.
We should be able to look at specific game searches to get more info on which teams were responsible for this change.
For 1991, here are the AL left-hander appearances lasting one batter:
| Rk | Tm | Year | #Matching |
|---|---|---|---|
| 1 | SEA | 1991 | 21 |
| 2 | BAL | 1991 | 19 |
| 3 | TEX | 1991 | 16 |
| 4 | DET | 1991 | 16 |
| 5 | CLE | 1991 | 16 |
| 6 | OAK | 1991 | 14 |
| 7 | CHW | 1991 | 13 |
| 8 | MIL | 1991 | 12 |
| 9 | BOS | 1991 | 10 |
| 10 | NYY | 1991 | 9 |
| 11 | CAL | 1991 | 9 |
| 12 | MIN | 1991 | 8 |
| 13 | KCR | 1991 | 8 |
| 14 | TOR | 1991 | 6 |
There were 177 such games in the AL in 1991.
Now here is the same list for 1992:
| Rk | Tm | Year | #Matching |
|---|---|---|---|
| 1 | OAK | 1992 | 38 |
| 2 | BOS | 1992 | 36 |
| 3 | SEA | 1992 | 30 |
| 4 | CLE | 1992 | 29 |
| 5 | BAL | 1992 | 22 |
| 6 | MIL | 1992 | 21 |
| 7 | MIN | 1992 | 20 |
| 8 | DET | 1992 | 20 |
| 9 | TEX | 1992 | 19 |
| 10 | NYY | 1992 | 14 |
| 11 | CHW | 1992 | 13 |
| 12 | CAL | 1992 | 12 |
| 13 | KCR | 1992 | 9 |
| 14 | TOR | 1992 | 5 |
The total jumps to 288 games, about a 60% increase. Oakland added 24 such games from 1991 to 1992, and Boston added 26 games. Here are the leaders among individual players in 1992:
| Rk | Player | Year | #Matching | W | L | GS | CG | SHO | SV | IP | H | ER | HR | BB | SO | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Tony Fossas | 1992 | 32 | Ind. Games | 1 | 2 | .333 | 2.16 | 0 | 0 | 0 | 1 | 8.1 | 5 | 2 | 0 | 2 | 6 | 0.84 |
| 2 | Jesse Orosco | 1992 | 19 | Ind. Games | 1 | 0 | 1.000 | 0.00 | 0 | 0 | 0 | 1 | 5.0 | 2 | 0 | 0 | 1 | 5 | 0.60 |
| 3 | Vince Horsman | 1992 | 19 | Ind. Games | 1 | 0 | 1.000 | 0.00 | 0 | 0 | 0 | 0 | 4.0 | 7 | 0 | 0 | 1 | 1 | 2.00 |
| 4 | Mike Munoz | 1992 | 18 | Ind. Games | 0 | 1 | .000 | 2.08 | 0 | 0 | 0 | 0 | 4.1 | 3 | 1 | 0 | 1 | 2 | 0.92 |
| 5 | Derek Lilliquist | 1992 | 18 | Ind. Games | 1 | 0 | 1.000 | 1.80 | 0 | 0 | 0 | 2 | 5.0 | 3 | 1 | 0 | 1 | 3 | 0.80 |
| 6 | Rick Honeycutt | 1992 | 16 | Ind. Games | 0 | 1 | .000 | 2.08 | 0 | 0 | 0 | 0 | 4.1 | 2 | 1 | 1 | 1 | 1 | 0.69 |
| 7 | Mike Flanagan | 1992 | 14 | Ind. Games | 0 | 0 | 5.40 | 0 | 0 | 0 | 0 | 3.1 | 3 | 2 | 1 | 1 | 3 | 1.20 | |
| 8 | Scott Radinsky | 1992 | 11 | Ind. Games | 0 | 0 | 0.00 | 0 | 0 | 0 | 2 | 3.1 | 1 | 0 | 0 | 0 | 4 | 0.30 | |
| 9 | Dennis Powell | 1992 | 11 | Ind. Games | 0 | 0 | 4.50 | 0 | 0 | 0 | 0 | 2.0 | 1 | 1 | 0 | 4 | 1 | 2.50 | |
| 10 | Kevin Wickander | 1992 | 10 | Ind. Games | 0 | 0 | 5.40 | 0 | 0 | 0 | 0 | 1.2 | 5 | 1 | 0 | 0 | 0 | 3.00 | |
| 11 | Mark Guthrie | 1992 | 10 | Ind. Games | 0 | 0 | 0.00 | 0 | 0 | 0 | 1 | 3.1 | 2 | 0 | 0 | 0 | 4 | 0.60 |
Tony Fossas leads the way with 32 such appearances for Boston. Vince Horseman and Rick Honeycutt were both pitching for Oakland.
Over the period of 1992-1995 here are the leaders:
| Rk | Player | #Matching | W | L | GS | CG | SHO | SV | IP | H | ER | HR | BB | SO | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Tony Fossas | 77 | Ind. Games | 2 | 2 | .500 | 3.05 | 0 | 0 | 0 | 2 | 20.2 | 12 | 7 | 1 | 4 | 17 | 0.77 |
| 2 | Jesse Orosco | 51 | Ind. Games | 1 | 0 | 1.000 | 1.88 | 0 | 0 | 0 | 3 | 14.1 | 8 | 3 | 0 | 3 | 14 | 0.77 |
| 3 | Derek Lilliquist | 47 | Ind. Games | 3 | 0 | 1.000 | 2.19 | 0 | 0 | 0 | 5 | 12.1 | 8 | 3 | 0 | 4 | 6 | 0.97 |
| 4 | Scott Radinsky | 44 | Ind. Games | 3 | 0 | 1.000 | 4.09 | 0 | 0 | 0 | 3 | 11.0 | 10 | 5 | 0 | 2 | 16 | 1.09 |
| 5 | Rick Honeycutt | 43 | Ind. Games | 1 | 2 | .333 | 5.73 | 0 | 0 | 0 | 0 | 11.0 | 10 | 7 | 2 | 2 | 4 | 1.09 |
| 6 | Jim Poole | 42 | Ind. Games | 1 | 1 | .500 | 0.79 | 0 | 0 | 0 | 0 | 11.1 | 6 | 1 | 0 | 5 | 9 | 0.97 |
| 7 | Vince Horsman | 36 | Ind. Games | 1 | 0 | 1.000 | 2.08 | 0 | 0 | 0 | 0 | 8.2 | 10 | 2 | 0 | 1 | 4 | 1.27 |
| 8 | Billy Brewer | 33 | Ind. Games | 2 | 0 | 1.000 | 1.93 | 0 | 0 | 0 | 1 | 9.1 | 1 | 2 | 0 | 5 | 6 | 0.64 |
| 9 | Bob Patterson | 27 | Ind. Games | 0 | 1 | .000 | 1.04 | 0 | 0 | 0 | 0 | 8.2 | 5 | 1 | 0 | 1 | 11 | 0.69 |
| 10 | Mark Guthrie | 26 | Ind. Games | 1 | 0 | 1.000 | 0.00 | 0 | 0 | 0 | 1 | 8.2 | 3 | 0 | 0 | 2 | 9 | 0.58 |
| 11 | Paul Assenmacher | 26 | Ind. Games | 1 | 0 | 1.000 | 1.42 | 0 | 0 | 0 | 0 | 6.1 | 4 | 1 | 0 | 2 | 8 | 0.95 |
There's Fossas again, plus the combo of Horseman and Honeycutt also ranks right up there.
It seems that Tony LaRussa didn't just solidify the idea of a closer who pitches strictly the 9th-inning but also solidified the idea of a specialty left-handed reliever brought in to face just one batter.
Incidentally, in 2009, the leading team was by far the Tampa Bay Rays, who themselves accounted for 47 of 206 such games in the AL. That included 29 games from Randy Choate, 14 games from Brian Shouse, and 4 games from J.P. Howell.
Tutorial
Here is a brief rundown of how the data for this post was generated. First, I did a bunch of Pitching Game Finder searches. For example, to find all relief appearances that involved just one batter faced, I clicked the bubble for "reliever" under the Role section and on the right I set the additional criteria of Batters Faced = 1. In this top yellow section, I clicked the bubble for "Find Players with Most Matching Games in a Season". This gives the totals for each year. When the results table came up, I sorted it by year. Then I copied the data and pasted it into Excel. After that, I went back and removed the Batters Faced = 1 criteria. This gave the results of ALL relief appearances. I put that data into Excel too. Then, by diving the totals for each year, I calculated the data in the black line on the chart. Finally, I went back and did similar searches for all the lines on the chart by clicking, for example, left-handed pitcher, or AL only, etc.
For the tables in the second part of the post, these are also Pitching Game Finders, but this time I selected just 1991 or 1992 as the date then grouped the results either by player or by team.
January 20th, 2010 at 11:25 am
But the so-called steroid era doesn't begin until 1993. The big jump in offense occurred in 1993 and again in 1994. I'm perplexed as to why this huge jump in lefty specialists would occur in one season, and basically remain there ever since. This is fascinating.
January 20th, 2010 at 11:57 am
I recall that some metrics indicate the start of the Steroids Era in 1992. It was certainly either 1992 or 1993.
January 20th, 2010 at 11:57 am
The other thing I forgot to mention is that even though the initial spike in behavior was seen in the AL in the early 1990s, the trend has now been equally spread over both leagues.
January 20th, 2010 at 12:39 pm
Awesome article! The jump in specialists seems to coincide pretty well with the first usage of pitch counts.
January 20th, 2010 at 1:08 pm
Lucas, what do you mean? Pitch counts certainly existed prior to '92, but they definitely didn't start controlling the game until around the turn of the century. (And of course, they still are rarely considered for relievers.)
Andy, assuming the Steroid Era just refers to the rise in scoring, there's no question it occurred in '93-'94. Except for the blip in '87, from 1980-1992 MLB scoring was between 4.00 and 4.41 runs/game. In 1992 it was 4.12. In '93 it went up to 4.60, in '94 to 4.92, and it has never fallen back below the '93 level. HR, which had never been more than 1 per game except for '87, went over 1 again in '94 and have never dropped below.
I looked at all the AL teams in '91 and '92 to see if I could figure out why this big jump in LOOGYs occurred. An interesting thing I noticed is that only three teams had the same manager for the entirety of both seasons (Det - Anderson, Min - Kelly, Oak - LaRussa). I thought that perhaps this change in strategy was a result of all the new managerial blood arriving, but it does not seem to be the case. Even the three holdovers all increased their LOOGY usage a lot, according to Andy's charts above.
January 20th, 2010 at 1:24 pm
Both scoring and homers were down in 1992 compared to 1991 in both leagues. 1992 is tied with 1988 as the lowest scoring NL season since 1968. I don't think it was due the "Steroid Era".
Fascinating stuff, though! Usually, you don't get a sustained jump like that without some sort of rule change. You keep coming up with these great plots.
I'd be curious to see what led to this jump. Anecdotally, I can imagine one or two teams having success with a LOOGY followed by everyone noticing and following suit. I don't know if I see what happened here. I wonder what caused so many managers to start using lefty-specialists at the same time.
January 20th, 2010 at 1:28 pm
As far as I knew, pitch counts were first tracked in the late 80s, and I figured that the increase in one-out guys was due to that; however, that wouldn't explain the sharp jump in 1992.
January 20th, 2010 at 1:46 pm
I think this is a great piece of work as well. I really think DavidRF has it right. Somebody thought of it, had success with it, and others tried it. Could we back up the team stats before 1991 to see where the first blip was.
Separately, Mark Guthrie appears to have been the most effective of these specialists for that time period...25 appearances / 25 outs / 0 ER. He allowed a few baserunners, so he obviously induced a few double-plays as well.
January 20th, 2010 at 1:58 pm
Lucas, I don't know when pitch counts were first "officially" tracked. Actually, as far as I know they're still not officially tracked, I _think_ those numbers are just kept by independent third-party stats providers, not by MLB. The Dodgers are known for recording pitch counts in the '50s and '60s (that data is on this site), and I'm sure other teams or independent observers did as well, though the numbers may not have been published or saved. Based on the numbers available on B-R, I think you are right that pitch count data started being widely recorded in the late '80s or early '90s.
January 20th, 2010 at 2:35 pm
I would have guessed that LaRussa's use of Rick Honeycutt popularized the LOOGY usage. It's nice to see that my memory was close to the actual truth. It looks like it wasn't LaRussa alone though.
I thought John Sickels coined the "LOOGY" term.
January 20th, 2010 at 3:41 pm
I thought it was Neyer but I might be wrong--Neyer might have just popularized it. Certainly at that time both guys were working closely on ESPN.com.
January 20th, 2010 at 4:44 pm
If you look at those 1991 & 1992 AL LOOGY numbers, you see that almost all teams increased their usage, most of them by significant margins. Three teams (BOS, OAK, and MIN) more than doubled their number of LOOGY appearances, three more (CLE, MIL, and NYY) increaed their usage by more than 50%, and SEA wasn't far off that mark at +42.9%. That's half the AL teams increasing their LOOGY usage by nearly 50% or more. Add to that the fact that only one team in the entire league (TOR) decreasded their LOOGY usage (only by 1 - from 6 to 5 - well within normal fluctuations), and only one team (CWS) had the same number of LOOGY appearances (at 13), and you have a genuine lemming effect.
I, too, would like to see the numbers for individual teams in the few seasons leading up to 1992. Perhaps they will reveal the head lemming. To butcher a quote (supposedly) by Picasso, "Good lemmings copy, great lemmings steal."