10.09.07
ALCS rotations
This year, for the first time, there is an extra off day in the League Championship Series between games 4 and 5. This is not a travel day, nor does it appear to be motivated by weather concerns. Most likely, the networks asked that the late-series games be staggered to avoid scheduling two deciding games on the same night. Under the new schedule, that can only happen if the NLCS goes a full seven. This is good for the networks, good for the bullpens, and irrelevant to most fans. But what implications does it have for strategy?
Previously, it was almost impossible for a team to get around using four pitchers in the LCS. You could choose to pitch your ace in games 1, 4, and 7, with the latter two starts coming on short rest (three off days between starts), however you would still require a fourth pitcher unless your other two starters were also able to pitch on short rest. Since pitchers working on short rest are generally both less effective and less able to work deep into the game, this would almost always be a losing strategy. The last time this was tried was when Bobby Cox used Greg Maddux on short rest in the 2001 NLCS in game 4; Maddux was good for only three innings and six runs against a thoroughly average Diamondbacks lineup. Arizona went on to win the NLCS in five and the World Series after that.
Now slip in the new off day between games 4 and 5. Suddenly the gap between games 4 and 7 is extended to normal rest, as are the gaps between games 2/5 and games 3/7. It becomes possible to use a three-man rotation in the LCS with exactly one start made on short rest. Neither would this schedule handicap a team in the World Series, since the starter in game 7 of the ALCS is lined up to work games 3 and 7 of the World Series. (The NLCS game 7 starter could potentially work games 2 and 6 of the World Series on normal rest, as that series finishes a day earlier.) Give me a team with an innings-horse at the top of the rotation, and I’ll give you a team that could potentially benefit from this strategy.
Consider, for example, the Indians rotation. Sabathia, a true workhorse, made 34 starts for 241 innings this year including four complete games. Even that is less than the six complete games he threw in 2006. Sabathia hasn’t even thrown that many pitches, as manager Wedge has held him to a hard pitch limit of 120. Sabathia has not attempted to start on short rest in several years, however 27 of his starts this year were on four days of rest. Clearly he doesn’t feel the need for an extra day very often.
While we don’t have much empirical data to work with, Sabathia could reasonably throw ~100 pitches (6-7 innings) in game 1, another five innings in game 4, and then stretch out to the full 120 pitches in game 7. This is a young fastball pitcher with a proven record of durability, not a 35-year-old Maddux who relies on location and finesse to succeed. I wouldn’t expect him to be at his best in the middle game, but he should still be a stronger option than Paul Byrd. The worst case scenario? Sabathia struggles in the early innings, is pulled quickly before the game gets out of hand, and Byrd works the next five. Sabathia would still be well rested for game 7. This plan does place a greater load on the bullpen, however the Indians have some excellent middle relievers (as we saw in the Yankees series). Perez, Lewis, Betancourt, and Fultz can easily take a game from the fifth through the eighth as many times as needed in a playoff series.
The Red Sox face the same question. Beckett and Sabathia are reasonably similar pitchers. Beckett was worked lightly in the first half of the season, and has made half his starts on an extra day of rest. In the second half, however, he regularly ran his count up to the same ~115 pitch vicinity with no signs of ill effect. Neither do his splits show any obvious benefit from that extra day — his ERA on four days rest is a stellar 3.09. Again, we have little history to support our case; but Beckett is as good a candidate as anybody in the game to pull this off successfully.
For the Red Sox, however, questions about their other pitchers might also play a prominent role in the decision. Are Matsuzaka and Schilling better off with an extra day of rest? Is the middle relief depth sufficient to cover the sixth and seventh innings on those days that Beckett is restricted? And most importantly, how comfortable does the team feel with Wakefield taking the mound for a start?
In fact, Matsuzaka has been much more effective when pitching with extra rest. He has a 5.29 ERA when working on four days of rest, a 3.80 ERA when working with one or more extra days between starts. The Japanese leagues use a longer rotation with extra rest between every start, so it wouldn’t be surprising if Matsuzaka has had trouble with that adjustment. Middle relief? I count Okajima and Delcarmen, but have little confidence in the bullpen after that. The Indians are weak in the ninth inning, while the Red Sox may be weakest in the sixth and seventh. And Wakefield? All sources insist that his side sessions have gone well and he will be ready to take the mound next week.
It isn’t clear whether either team will attempt the strategy that I describe, though I believe it makes more sense for the Indians than for the Red Sox. The Red Sox have greater depth in the rotation, less bullpen depth, and a reputation for using their pitchers cautiously. They enter the ALCS as the favorites, making it less likely that they will take a risk in testing the unknown. If both teams choose the same option, then the decision in some sense becomes a non-factor in the series outcome. But if they choose differently? Whichever way the series turns, this could become the focal point of the post-mortem.
Just reported: The Red Sox will pitch Beckett, Schilling, and Matsuzaka in that order. Wakefield is the likely starter for game 4 at this time.