In case you missed the FanShot from last Friday, the Astros hired a new pitching coach. Brad Arnsberg comes with a reputation as a young pitcher-killer, but was loved by all his staffs and seemed to be able to work some magic with 'reclamation projects.' The World Wide Web was abuzz with articles about his abundance, or lack thereof, of talent. A fine hat-tip to clack and AstroAndy for linking to them.
What I wanted to do, though, is to see if it was warranted. What we're really worried about here are young pitchers, right? Guys 24 years old and younger are in the prime area for suffering from over-use. There are two different models out there currently used to predict injuries to young pitchers. Both were researched pretty thoroughly, but it's hard to tell if they're good predictors of injury potential or just good at explaining what happened in the past.
When I went back to research Arnsberg, I looked at every staff he had from 2000 in Montreal to 2002 with the Marlins and through his five-year tenure in Toronto. I focused on all the pitchers he had that were 24 or younger. This is a bit of a departure from the model, since they stop at 25, but since two-thirds of his stops were of the one-year variety, I downgraded the age a year. The two methods I mentioned were Tom Verducci's theory on an increase in innings from year to year, most recently laid out here. The other was Pitching Abuse Points (PAP), first published back in 1998 by Rany Jazayerli here. I believe it's been tweaked a bit here and there since then, but it's a good starting place for the discussion. My results after the jump...
Arnsberg oversaw 28 different pitchers in their Age 24 season or younger. His biggest year for this was in 2000 in Montreal, when 10 different young pitchers got into games. The 2000 season also seemed to be a learning year for Arnsberg as to how best to utilize pitchers. Of the 10, only one threw more innings than the previous season (Javier Vasquez, +63 IP) and only one had more than 10 PAP per start (Vasquez, 16 PAP). Carl Pavano pitched seven fewer innings than in 1999, but had a PAP/Start of 4.8 and then suffered an injury the following season. The rest all had significant drops in innings pitched and none had more than 2 PAP/start. None of the relievers in his bullpen threw three straight days and only Pavano suffered an injury the following season. Knowing what we do now about Carl, it's easy to see it probably wasn't Arnsberg's fault.
The next year he surfaced was in 2002 with the Marlins. Five guys played at 24 or younger and two of the five had jumps higher than 30 innings. Both Justin Wayne and Nate Robertson pitched over 40 innings more than the previous season, but each threw the majority of their innings in the minors, not the majors. It's hard for me to pin these jumps on Arnsberg for that reason, since their workload was controlled by the minor league pitching coordinator, not the big league pitching coach. As for the other three, Josh Beckett threw 47 less innings than the previous season and averaged 2.1 PAP per start. Pretty reasonable for a 22-year old. Brad Penny had a 69 inning drop in his workload and had 0.8 PAP/start. The other guy in this group was Blaine Neal, a reliever who saw his workload go up by six innings and was used three straight days twice that season. One of those times, though was on both ends of a double-header, when he faced just one batter in the first game. No real injury trouble here, at least not by Arnsberg's work. Beckett had the blister problems, but that's not related to overuse and Penny did have arm trouble, but much later than this.
On to the Toronto years, where Arnsberg saw an impressive group of young pitchers hit the big leagues. In his first season, he had five pitchers in the age range we are looking at. Two drastically increased their inning total (Dustin McGowan, +70, Chad Gaudin +74) while one starter also saw an uptick into the danger zone (Gustavo Chacin +36). The other two guys, Brandon League and Shaun Marcum had smaller increases or decreases, but only Chacin had any PAP to speak of. Chacin's total of 3.7 PAP/start is high, but not distressingly so and his innings increase was slight in comparison to McGowan's. Still, Chacin did suffer a major drop-off in performance and had an injury or two thrown in for good measure. Of course, his delivery is a bit quirky and he came out of nowhere to begin with, so it's hard to say whether Arnsberg led to this flameout. Certainly, the other guys on this list went on to have solid years after 2005, so we may just be dealing with a small sample size.
In 2006, Arnsberg oversaw seven young pitchers, including newcomers Casey Janssen, Ty Taubenheim, Jeremy Accardo and Davis Romero. Accardo, Romero and Dustin McGowan all had inning increases but the rest saw their totals drop. Janssen was the only pitchers with more than 1 PAP/start and his was exactly one. Marcum saw his total drop by 35 innings and had a PAP of 0.8 while League had a slight drop in innings, but was used twice three games in a row. Taubenheim moved from a starter's role in the minors to the bullpen in the majors, which explains his 60-inning drop. None of the evidence here supports the conclusion that misuse led to any injuries.
In 2007, there were only four pitchers in our age-range, including two holdovers, League and Taubenheim. Newcomers Jesse Litsch and Josh Banks both saw increases, though Litsch saw his total rise by 29 innings, just underneath the magic threshold. Litsch also had just 0.7 PAP/start, meaning he wasn't throwing a ton of pitches. League saw his total drop by 59 innings, in part due to injury (I think, though I can't find it anywhere). Taubenheim had his total rise by 15, but was not used in more than two straight games, so he didn't really get overworked.
Litsch was the only pitcher considered young in 2008 and he saw his total innings drop by 11, but his PAP jumped up to 4.2 per start. More distressingly, though was how he was used down the stretch. From August 21 to September 29, Litsch started nine games and threw 61 innings and 980 pitches. He totaled 88 PAP for an average of 9.7 per start, a huge jump over the beginning of the season. He went 5-4 over this stretch and had an average game score of 62.4. Those were both reasons why he pitched so much, I'm sure. He was doing so well that the Jays wanted to keep him pitching to keep them in the race, as the team got to 12 games over .500 on Sept. 12. However, Litsch was pretty much abused during this stretch. He was used once on two days' rest, throwing 49 pitches over three innings and was used once on three days' rest in the back half of a doubleheader. Litsch threw 110 pitches over 5 1/3 innings.
In 2009, Litsch was shut down in April and finally had Tommy John surgery in June. I think it's safe to say that his usage pattern at the end of 2008 may have led to this, but we can't be certain. Arnsberg's last year with Toronto saw five pitchers in our age-group, including Litsch. Ricky Romero, Brett Cecil, Mark Rzepczynski and Brad Mills made major-league debuts and all but Mills saw increases in their inning totals. None saw jumps of over 30, though they all had more PAP than in Arnsberg's past. Romero tallied 6.3 PAP/start and saw his innings jump 14, while Brett Cecil had a jump of 24 innings and 2.1 PAP/start. Rzepczynski made the jump straight from Double-A and saw an increase in 28 innings while having a PAP of 3.8 per start.
What does this mean? Arnsberg gets blamed for the injuries to Marcum and McGowan, but in their formative years with the club, you can see he really didn't do much to hurt them. Only in 2008 did his usage patterns significantly change, and that also coincided with a new manager, Mr. Old School Cito Gaston. This is a classic example of a guy from a bygone era trying to throw pitchers out there like they did in the '70's and hoping it will work. In 2008, for example, Gaston used pitchers on short rest six times, while the American League average was just 3.7. In 2009, he was right in line with the league average of 3, but it still shows that he used his pitchers differently than most of the league. Arnsberg didn't like it, and clashed with his manager as was chronicled here and here. I'm not sure you can blame Arnsberg for what happened to Litsch but you definitely can't blame him for Marcum and McGowan. B.J. Ryan also can't be blamed on him, nor can A.J. Burnett, as both guys had either a history of injuries or a jump in innings before coming to Toronto.
At least one place is mad Arnsberg left Toronto, while there are a couple other places that still levy criticism. I'm not sure how Arnsberg will do with Houston, but I do know he's got a pretty good track record of keeping the young guys healthy. He'll have a tough time on his hands next season, though, as Bud Norris saw a 95 inning jump in his workload, though his PAP/start was just 0.5. Yorman Bazardo was also in the danger zone with a 34 inning jump, and his performance doesn't need any regression. The only worrying part of the bullpen was Sammy Gervacio being used three straight days three different times and once for four straight games. Of course, the 24-year old only threw more than 20 pitches in a game twice, so it's not as worrisome as it appears.
To review, Arnsberg coached for four different managers and only under Gaston did his track record go astray. Out of all his young pitchers, just three were inordinately abused: Chacin, Litsch and Vasquez. Now, we can see that Vasquez was just built to hold up to a higher workload, similar to Roy Halladay. Still, it doesn't excuse the abuse in the first place. Is Gaston more to blame for the last couple of seasons in Toronto? We can't REALLY tell, but after reviewing all this, I certainly feel better about the Astros' new pitching coach.