A trivial statistical walk through the 2010 bullpen

I was reading The Bill James Handbook 2010 and took especial notice of the Relief Pitching section. The statistics given here aren't available or aren't easily analyzed at my usual online places and I thought that taking a look at those here would perhaps let us understand a bit more about the future performance of the Brewers bullpen based on last year.

Unfortunately, there are some things and players I can't look at for various reasons. The 2009 version of the Handbook had a great chapter about defining roles for pitchers via statistics. My 2009 book is in storage. For our purposes, it's important to know that Mitch Stetter is defined as a setup man although we all know from seeing him in action that his role in actual use tended to flip from setup to situational specialist depending on the game. The only other LHP in the 'pen in 2009 were R.J. Swindle, who was used situationally for the small time he was there, and Chris Narveson, who wasn't. I'm not going to spend a lot of time dwelling on players who are no longer with the team, who are injured for 2010, or who didn't have much or any data to analyze in 2009. This means that I can't really get a good look at Chuck Lofgren with these metrics.

Mitch Stetter led the major leagues in inherited runners when entering a game, with 67. According to his game logs, he entered 5 games with 3 runners on base and 17 games with 2 runners on base. Out of those 67 runners, 14 scored.

Hoffman really did have a great year. If you use the more traditional statistics, 2009 was the second-best year of his career. Hoffman was 4th in the majors last year in clean outings, defined as appearances where a pitcher does not allow a run or allow an inherited runner to score; 47 out of 55 of Hoffman's games were clean. I note that in 4 of the 8 games in which Hoffman allowed runs, the runs were scored by inherited runners. (I also note that Hoffman gave up hits in 25 appearances and walks in 10; in 5 games he also gave up hits and walks.) Hoffman also had the 3rd lowest opposition OPS for pitchers appearing in 50 or more games, with .481.

Todd Coffey led the staff in games pitched with 78, followed by Mitch Stetter. LaTroy Hawkins appeared in 65 games with the Astros. Yes, this is a commonly available statistic but I threw it in here just to show how much Coffey and Stetter were used as the primary setup man and situational lefty, respectively. Hawkins' 2009 performance was quite comparable to Coffey's. If used correctly, both should be able to give each other a break and prevent the sorts of problems with tiring and rest that happened at the end of 2009. Chuck Lofgren may be a viable alternative to Stetter just in case.

The Brewers need a new long man in 2010 and just may need new mopup guys as well. They used a variety of pitchers in 2009 when starters couldn't make it through the 6th inning. Mark DiFelice led the team in early entries with 21, followed by Seth McClung with 18. The obvious long man candidate is Chris Narveson. 11 of Narveson's 17 appearances in relief came in or before the 6th inning. He pitched for more than 1 inning in 7 of those appearances. As for mopup, the other early-entry pitchers still with the team with undefined roles are Carlos Villanueva and Chris Smith. As we'll see in the leverage section, it's very obvious that Smith was already filling this role in 2009.

Todd Coffey led the team in long outings, defined as games in which pitchers threw 25 or more pitches. Most of those long outings for Coffey were multiple-inning appearances. When they weren't, he was giving up a lot of hits. Claudio Vargas had 8 long outings. Not surprisingly, those were multi-inning appearances but instead of giving up hits, he was giving up walks as well.

Leverage index, which is pretty much available everywhere, tells you the importance of the situation in which a pitcher is used. Average is 1.0. Hoffman's was 1.9. Coffey's was 1.3. Narveson's was 0.5. Smith's was 0.4. Looking at Narveson's game logs, it's evident that he mostly came in when either a win or a loss (mostly a loss) was a forgone conclusion. Smith pitched in a whole bunch of lost causes; the Brewers lost 29 out of the 35 games in which he appeared.

Relatedly, closer usage on the Brewers in 2009 was very interesting. Most of Hoffman's saves were defined as "easy," meaning that he entered the game with the tying run not on base or at the plate and with three outs or less to go. Yup, this is the good old "it's the 9th inning, it's a save situation, stick in the closer" situation. Todd Coffey, on the other hand, had four opportunities for tough saves where he entered with the tying run on base. He only converted one of those opportunities. It's obvious that when the going got tough in 2009, one of the tough guys to go to was Coffey.

The bullpen should be the least of fans' worries in 2010. Fans are probably right to worry that Hoffman and Hawkins can't keep their performance up to the absolutely elite levels they were on in 2009, and that they're overpaid, but I don't see any reason to worry that things will be significantly worse on all levels for the bullpen in general for 2010. The only wildcards in this process are the performance of David Riske and some of the minor league players, or what may happen if someone like Suppan or Parra get shifted into the bullpen in place of players like Narveson going to the rotation. The Brewers have a recent history of getting more performance out of minor league free agent pitching than one would expect; DiFelice, Narveson, Smith and Mike Burns spent significant time with the Brewers last year. I see no reason to also expect that the 2010 signings may not make impact in 2010 or 2011 if they perform well in AAA and stick around. Now, to worry about that rotation....