We have moved to our new permanent home at HighHeatStats.com!

All existing content will remain, but all new content is posted at our new location.

More and more pitchers with short appearances

Posted by Andy
For the 3rd year in a row, there was a record-setting numbers of pitchers with fewer innings pitched than games. That means they averaged less than 1 IP per appearance:

Rk Year #Matching
1 2011 206
2 2010 191
3 2009 160
4 2007 158
5 2008 143
6 2005 142
7 2006 130
8 2004 120
9 2002 116
10 2003 105
Provided by Baseball-Reference.com: View Play Index Tool Used
Generated 11/2/2011.

Think about that--in just the last 9 seasons, the number of pitchers averaging less than an inning per appearance has nearly doubled.

Here are the total number of pitching appearances in those 9 years:

Year ▴ #Matching
2011 18752
2010 18786
2009 19099
2008 19013
2007 19294
2006 18694
2005 18037
2004 18272
2003 17818
Provided by Baseball-Reference.com: View Play Index Tool Used
Generated 11/2/2011.

You can find the total number of games played right here, meaning that in 2011, pitchers averaged 3.86 innings per appearance (which includes starters) and in 2003, pitchers averaged 3.67 innings per appearance.

It would seem, though, that more individual pitchers are being used these days--more injuries or callups or whatever.

When I get more time and am not putting up a brand new blog, I'll go back and figure this by relievers only, which should be more illuminating.

28 comments:

  1. Who knows... perhaps that changes, now that Tony La Russa is gone.

    ... as well as the Ranger bullpen phone

    ReplyDelete
  2. Are relievers who pitch less innings than games more or less effective than those who don't?

    ReplyDelete
  3. Abott @12:43pm -- To answer that question in a meaningful way would require a complex study, because of the self-selection bias involved.

    A reliever who gets hit up in his first inning of work is likely to be lifted before he completes an inning. Similarly, a reliever who sails through an inning is more likely than average to start another inning (although extended relief outings are becoming scarcer every year).

    ReplyDelete
  4. OK, here's more info.

    In 2011, starters averaged 6.0 IP per game.

    (see that here.

    That was 29299.1 IP over 4858 games (6.03 IP per start). Relievers pitched 14288 innings over 13894 innings (1.028 IP per relief appearance).

    In 2003, here are the numbers:

    28615.2 IP for starters in 4860 games (5.89 IP per start).
    14719.2 IP for relievers in 12958 games (1.136 IP per relief appearance.)

    So, it would seem that more relievers are being used to pitch fewer innings, as the starters are going deeper. I wonder if this is partially roster evolution....teams are carrying a lot of relievers as habit but as starters go longer maybe we see bullpens shorten by a guy or two and the average relief appearance gets longer again.

    ReplyDelete
  5. Running with Andy's idea:

    Since 2000, # of pitchers with IP less than Games, and at least 30 games:

    Year #Matching
    2000 57
    2001 71
    2002 71
    2003 74
    2004 75
    2006 91
    2005 92
    2008 97
    2007 107
    2009 119
    2010 135
    2011 135

    ReplyDelete
  6. Correction to my previous comment: at least 20 games (not 30).

    ReplyDelete
  7. Good point John. I wonder if you looked at guys who only pitched to 1 batter, what their numbers would look like.

    ReplyDelete
  8. Continuing with a breakdown of pitchers with IP less than Games, and at least 20 games, since 2000:

    Lefties:
    Year -- #Matching
    2011 -- 50
    2010 -- 54
    2009 -- 54
    2008 -- 45
    2007 -- 42
    2006 -- 45
    2005 -- 46
    2004 -- 41
    2003 -- 43
    2002 -- 44
    2001 -- 44
    2000 -- 36

    Righties:
    Year -- #Matching
    2011 -- 85
    2010 -- 81
    2009 -- 65
    2008 -- 52
    2007 -- 65
    2006 -- 46
    2005 -- 46
    2004 -- 34
    2003 -- 31
    2002 -- 27
    2001 -- 27
    2000 -- 21

    Looks like the "LOOGY" phenomenon has stabilized; there were slightly fewer lefties this year than in the prior 2 years who met this criterion.

    The grown in pitchers with IP less than G is being driven by righties.

    ReplyDelete
  9. The total number of appearance looks like it peaked a few years ago and has started to come down, even as the number of pitchers with fewer than 1 IP/G has continued to go up. That means the pitchers who ARE throwing more than an inning are lengthening their appearances.

    It may be that starter innings troughed and are now on the way up (I haven't checked because I'm in a hurry) or it may be that there are now more bullpen innings being taken by pitchers whose role is to pitch in longer stretches. Or both.

    ReplyDelete
  10. Most One-Batter Pitching Appearances in a Season, all MLB Pitchers Combined:
    2011: 1,220
    2007: 1,168
    2010: 1,160
    2009: 1,119
    2006: 1,118

    In 1999 there were 980 one-batter appearances.
    In 1989 there were 508 such appearances.
    In 1979 there were 441 such appearances.
    1n 1969 there were 459 such appearances.
    In 1959 there were 161 such appearances.
    In 1949 there were 121 such appearances.
    In 1939 there were 81 such appearances.
    In 1929 there were 55 and in 1919 there were 18.

    All, of course, according to b-ref's Play Index.

    The top guys all-time in number of one-batter appearances in a career:

    Mike Myers 314
    Jesse Orosco 235
    Don Plesac 175

    Most One-Batter Appearances 2009-2011:
    Randy Choate 94
    Tim Byrdak 64
    Trever Miller 63

    ReplyDelete
  11. Record-setting for the third year in a row? It looks like the ninth year in a row, judging from the table. Or am I missing something here?

    ReplyDelete
  12. Mike Myers: .204/.304/.323 in his one-batter apperances.
    Julian Taverez: .057/.074/.0151 in 54 one-batter appearnces.
    Mike Holtz: .255/.333/.309 in 105 one-batter appearances.
    Pat Clements: .467/.525/.567 in 40 one-batter appearances

    ReplyDelete
  13. I agree with Dvd avins. It's interesting that very short relief appearances are going up, yet overall relief appearances are going down. It seems likely to me that, as offense has gone down around the league in the past 5-10 years, managers are adjusting and allowing their starters to pitch deeper into games, but have not yet adjusted their bullpen strategies.

    Also, well done John Pazniokas, who not only beat me to the Tony LaRussa joke, he did it within 8 minutes of the original post.

    ReplyDelete
  14. This comment has been removed by the author.

    ReplyDelete
  15. Not sure if a move from 5.89 IP to 6.00 by starters over two years is significant, or just a blip, the 1 batter stats that have been dug out are a much more significant looking series , but there is obviously the possibility of a train of effects here. What is the number of pitchers on 25 man rosters? Assuming it has gone up for reasons extrinsic to the current question, as for example, a discovery about a better way of developing young arms than giving them 600 IP in the minors, that would increase the in-game attractiveness of platoon advantage pitching changes both because there are more available and the opposing manager has fewer options to counter with . To me , the most intriguing fact that this excellent thread has dug out is that all of the effect is from RHP. who'd a thunk it?

    ReplyDelete
  16. #14, I came to the same conclusion at the end of my comment above at #4...I definitely agree. Bryan hits it in 15 as well.

    BryanM, I'm sure that 5.89 to 6.03 over 9 years is quite significant--I've looked at the numbers in more depth, and they are a significant trend that leave a lot fewer innings for relievers these days.

    Also #14, would you consider going back to your Mike L name? People know you as that...

    ReplyDelete
  17. I've ranted about this before so I'll save the rehash but the first manager/GM smart enough to replace the 12th & 13th pitchers on the roster (who likely face fewer than 100 batters between them in the course of the season) with a solid bat and an infield glove are going to pick up 3 to 5 wins a year.

    ReplyDelete
  18. This comment has been removed by the author.

    ReplyDelete
  19. This comment has been removed by the author.

    ReplyDelete
  20. I thought this when Hatvig first posted that idea on the old blog---

    That opinion is a reflection, I think, of how the league is changing to resemble the 1980s more, when it was the norm for all teams to carry 10 pitchers, or maybe 11. Hartvig is seeing less value in the 12th and 13th guys because they are being used less and matter less, and if the offense trend continues for another year or two, I have no doubt we'll see at least one team that goes to a 11-man staff and puts an extra hitter/defender on the bench.

    ReplyDelete
  21. Andy-will you take "Michael"? Seems to be the best I can do, unless I open a 4th gmail account.

    Of course, by this time, I've used up all my insights...

    ReplyDelete
  22. I'll accept whatever you want...the other name is fine too. I just want people to be able to remember who you were from the old blog--you were a great and prolific commenter over the last few months of the B-R blog, so some continuity would be nice.

    ReplyDelete
  23. I assume Orosco has the record for most seasons with more games than innings. He did it before it was cool in 1988 and he just kept on doing it about 15 more times after that.

    I'd love to see the top table broken up by lefties and righties. I'd assume that most of the pitchers with more games than innings are lefties.

    ReplyDelete
  24. The 12th and 13th pitcher may be a good idea gone too far, as Hartvig believes, in which case surely some innovator ( ie not Dusty Baker) will discover this and exploit it. I personally believe, although I have no data to back this up, that hitters are facing better pitching throughout the game as managers are not tempted to try to squeeze one more out out of an obviously tired starter in order to save the bullpen, and the next innovation will be to acquire a pivot man , decent bat, can play a few positions competently ( Skip Schumaker) to free up a roster position for another bat/Lf/DH, and that this will be acknowledged role, like setup man today.

    ReplyDelete
  25. @24, Bryan: ”hitters are facing better pitching throughout the game as managers are not tempted to try to squeeze one more out out of an obviously tired starter in order to save the bullpen”

    Bryan, I’m inclined to agree with you (also without having studied it). But I think there’s a self-fulfilling factor in that. The increase in # of relievers carried and the number of changes made during a game, and the decline in # of bench hitters, means that a reliever probably has the platoon advantage more often today than in decades past.

    ReplyDelete
  26. Followup to #25:

    As a very rough approach to the question whether today's relievers are more likely to have the platoon advantage, I looked at lefty-lefty matchups for 3-year periods. Note that these data are for all pitchers, not specifically relievers.

    Percentage of all PAs that are LHP vs. LHB:
    2009-2011 - 8.0%
    1999-2001 - 6.5%
    1989-1991 - 7.0%
    1979-1981 - 7.4%

    This indicates some small support for the thesis, although it could easily be overwhelmed by other facts not studied (such as the % of overall PAs overall that are represented by LHPs and LHBs).

    ReplyDelete
  27. Followup to #26 and #25: Trying to account for yearly variations in the number of total PAs represented by LHBs and LHPs overall, I divided lefty-lefty matchups by the average of (PAS by LHBs plus PAs against LHPs).

    The 3-year averages of this adjusted percentage:
    2009-2011 - 22.6%
    1999-2001 - 19.7%
    1989-1991 - 18.7%
    1979-1981 - 21.0%

    I realize that there's probably a simpler method, and/or someone else has already calculated the percentage of PAs time that a reliever has the platoon advantage.

    But this rather blunt study still suggests to me that today's relievers are, in fact, more likely to have the platoon advantage than their counterparts from decades past.

    ReplyDelete
  28. You won't find many pitchers with appearances this short:

    http://www.baseball-reference.com/players/h/herreda01.shtml

    (Listed at 5' 6"....)

    ReplyDelete