Another Look at Bullpen (mis)Management

It was brought to my attention that my own research suggests that the Phillies’ bullpen should be more worn out than the Mets’. Yet, the Phillies relievers continue to outperform those wearing orange and blue — as well as every other team in the NL.

So what gives? Do the numbers lie? My theory incorrect?

Let’s look at this in further detail and try to figure this out, comparing the two teams head to head.



From the above, it’s true that the Phillies top seven most-frequently used relievers have accounted for 94% of the total relief innings pitched by the team. The “top seven” Mets relievers, in contrast, have accounted for only 83%.

There are a few flaws with my “top seven” theory, and some things jump out. First being, the fact that one of the Mets’ “top seven” — Jorge Sosa — hasn’t been on the team since May. The second is the number of appearances, which aren’t addressed by my percentage of innings.

The Phillies have only two people who have appeared in more than 60 games. In comparison, the Mets have four, with two of them about to hit 70. So while the Phillies’ top relievers have carried most of the innings load, they haven’t appeared as frequently.

Another issue is the 16 more relief innings that the Mets have had to cover — that’s nearly the equivalent of two full games. So the Mets relievers are getting into more games, and pitching more innings.

Something else strikes me, which goes hand-in-hand with the frequency factor. Looking at the Phillies’ relievers, there are three pitchers with less innings pitched than appearances. One of them is their closer Brad Lidge, but we know he’s not a “matchup” guy. So in essence, the Phillies use only two guys for matchups — J.C. Romero and Tom Gordon (who is currently on the DL). And actually, Gordon wasn’t really a ROOGY, but more of a setup man — he threw at least one inning in nearly all of his appearances.

Compare that to the Mets, who have used Feliciano, Smith, Schoeneweis extensively as “one out guys”, and began using Duaner Sanchez similarly of late. On the one hand, you’d think that pitching to only one batter shouldn’t be such a toll on their arm and body. But on the other hand, they’ve had to warm up in the bullpen for each of their appearances, plus how many other times when they didn’t get into a game? Is it possible that all those pitches thrown in the bullpen take just as much, if not more, of a toll on the human body? Understand that pitchers often “get up” several times in a ballgame. Is anyone counting how many pitches each reliever is throwing — pregame + bullpen + game + between-innings warmups? Every one of those pitches takes something out of a pitcher’s body.

Further, Jerry Manuel made a big deal about “establishing roles” in the bullpen. Looking at the numbers — and watching the games — I’m not sure I recognize a pattern that would suggest roles, other than Billy Wagner as the closer. With Wagner now out, and a month of the season left, all roles go out the window — it’s all hands on deck. Newsflash: it’s been a roleless, “all hands on deck” bullpen since May.

What makes things scarier is that Manuel has publicly stated several times that he will be “going with the hot hand” out of the bullpen. I don’t need to mention that this strategy used to be employed by Willie Randolph, who learned it from Joe Torre. What I do need to mention is that it makes zero sense. HITTERS go on hot streaks — not pitchers. The only time pitchers go into “slumps” is when they are overused and fatigued. So by “going with the hot hand” you are effectively creating the “slump”.

With first place — and his job — on the line, Jerry Manuel has begun managing out of fear. If Smith can’t get a batter out, Schoeneweis comes in. If Scho doesn’t get an out, Feliciano comes in. If Feliciano gets an out, he stays in until someone gets a hit. And so on. This is not a strategy, and certainly not “management”. It’s panic.

Unfortunately, bullpen management is not something that you can suddenly start to do in the last 30 games of the year — it is something that begins on Opening Day. While I still think the Phils are due for a bullpen collapse in September, it appears that, so far, Charlie Manuel’s Phillies have managed their bullpen arms for the long haul, while the Willie Randolph/Jerry Manuel Mets have put the “pedal to the metal” since the beginning. As we move into September, when “every game counts”, the Mets just may push that pedal through the floorboard, snap the accelerator cable, and thereby stop the engine.

Joe Janish began MetsToday in 2005 to provide the unique perspective of a high-level player and coach -- he earned NCAA D-1 All-American honors as a catcher and coached several players who went on to play pro ball. As a result his posts often include mechanical evaluations, scout-like analysis, and opinions that go beyond the numbers. Follow Joe's baseball tips on Twitter at @onbaseball and at the On Baseball Google Plus page.
  1. hdarvick August 27, 2008 at 2:17 pm
    The Mets will have Nelson Figueroa in the bullpen tonight for long relief. His only 2008 Mets long relief stint was on April 30th when he relieved Oliver Perez in the 2nd inning vs Pittsburgh, got the 3rd out, then pitched the 3rd, 4th, and 5th innings allowing 1 run. Mets lost that game 13-1 (the old Ollie allowed 7 runs in the 2nd).
    Figueroa’s major problem on the Mets was his walk to strike out ratio. It’s improved dramatically: on the Mets 24 walks, 29 strike outs in 38 2/3 innings; on New Orleans 33 walks, 97 strike outs in 113 2/3 innings.
  2. Micalpalyn August 27, 2008 at 2:51 pm
    Hopefully dirty is placed on the DL.
  3. isuzudude August 28, 2008 at 6:52 pm
    Taylor, I lasted about 4 paragraphs into that article before getting completely lost. Can’t bullpen management and projecting relief pitching be made any simpler?
  4. joe August 28, 2008 at 7:09 pm
    Thanks for the link, Taylor. What’s your take from that article compared to the argument presented here? It’s interesting but how can we use it to project over the course of one season?

    Isuzudude, the key is to go to the bottom and read up (the LAST four paragraphs instead of the first four). 😉

    But there are a few problems with that Hardball Times article, the main one being that the measurement / comparison is in ERA. I tend to believe that ERA can be an AWFUL indicator of measuring a relief pitcher. Case in point: Scott Schoeneweis, who has had a sparkling ERA all year but has been terrible in regard to allowing inherited runners to score.

    Also, the bulk of the study is based on years prior to steroid testing. I will be curious to see the same study done only from 2006-. I’m convinced that PEDs were used just as frequently by overused relievers as anyone else, and now that they don’t have that crutch, I think overuse will be a bigger factor going forward.

  5. Taylor September 2, 2008 at 2:32 pm
    True it doesn’t address the effect within the season. It simply suggests that the bullpen’s performance THIS year is not due to overuse LAST year.

    I remember seeing another article where someone did a study of relief pitchers velocity and movement using the velocity and movement data from the’s gameday database.

    They were trying to determine whether pitching too much decreases velocity and whether having more rest tends to increase it. The conclusion was yes. Pitching too often decreases velocity and movement and having rest, but not too much rest, improves velocity.

    I’ll try to find that article and post the link. I looked for it but couldn’t find it so far.

  6. joe September 2, 2008 at 2:42 pm
    Taylor, that’s great research you’ve done, thanks.

    This is definitely a subject we will re-visit in the next week or so.

    One of the real problems that many people (and managers) don’t understand is that a lack of velocity doesn’t necessarily mean there isn’t a problem / injury. Very generally speaking — and this is from medical experts, not me — a drop in velocity suggests an issue with the shoulder, while a lack of control suggests an issue with the elbow.

    When Aaron Heilman lost his setup role to Duaner Sanchez in the first half of 2006, it was because he was pitching a LOT like he has recently — velocity was fine, but control was not. He did recover with a fine August / September, but after the season quietly underwent minor elbow surgery (just a scope, not TJ).

    One must wonder if he’s having the same issue now?