October 12, 2004
Looking Both Ways
I've always been jealous of the great Sherri Nichols for coining a cool baseball term that included her name. The Nichols Law of Catchers' Defense states that a catcher's defensive reputation moves in inverse proportion to the quality of his hitting. It's frighteningly accurate, and responsible for the careers of players like Brad Ausmus
at one end and the career paths of guys like Mickey Tettleton
at the other.
Heck, even Jazayerli's Law of Backup Catchers--given enough chances, a backup catcher will have a season in which he hits .300 in less than 200 at-bats--gets run, and all Rany has on me is height and a name with more Scrabble value.
So, in a blatant attempt to get listed in a glossary someday, I'm introducing Sheehan's Starter Selection Theory:
Any starting pitcher is a good starting pitcher if you score 12 runs.
We can analyze the decision to start a pitcher on three days' rest, or start a lousy pitcher instead, or go with Johhny Wholestaff for a game, until we're blue in the face. None of that matters when your offense goes nuts.
The Astros were the latest team to provide support for the theory, backing Roy Oswalt with four runs in six innings and finishing with an even dozen to--all together now--win a postseason series for the first time in franchise history. It's amazing what not having to face the best starters in the league will do for your chances.
Oswalt was the second straight Astros starter to go on three days' rest, following Roger Clemens on Saturday. It was also the third time in this playoff round that a starter had done so. What's weird is how similar the results were. Each starter scuffled through exactly five innings, allowing one, two and two runs respectively while escaping rallies along the way. Each pitcher got good run support, each leaving with at least a two-run lead. Twice, the pitcher's relief corps blew the lead; last night, the Astros got four innings of one-run ball without using Brad Lidge and won the game.
While the results were there for Phil Garner, I don't think he got value for his decision. Clemens and Oswalt threw 10 innings in their two starts. On normal rest, you could probably have pencilled Clemens in for seven innings, with Oswalt possibly available for an inning or two of relief. The net gain was small, and exposed the Astros to comparable risk--the back end of their staff--as they would have faced in starting Peter Munro in Game Four.
When we talk about the effects of using pitchers on short rest, we are generally concerned with their effectiveness. However, a big part of the edge top starting pitchers have is that they tend to work deeper into games. If starting them on short rest gives back that edge, it shifts the balance even more towards not doing so, and certainly not when the decision involves a team that's ahead in the series, as the Astros were in this one.
Having used Clemens and Oswalt to get by the Braves, the problem the Astros now face is that they will have to either use Munro in Game Two against the Cardinals--after opening the series with Brandon Backe--or again use Clemens on short rest. Neither option is attractive. Even if you can make the case that starting Clemens on Saturday (and committing to Oswalt on Monday) was the best way to beat the Braves, it was clearly a short-term strategy that may have increased their chances of winning one series while reducing their chance at a championship.
The biggest strategic point to come out of Monday's game was that Lidge didn't pitch in it. Given that he threw just seven pitches on Sunday, he'll go into the NLCS well-rested, and available if Phil Garner needs to stretch him out to win one game. The Astros' best chance to win this series will be stealing a game in St. Louis so that they can go home with Clemens and Oswalt available to start Games Three and Four on full rest. An effective, durable Lidge is going to be critical to any win in Missouri.
While the final score was 12-3, making an extended analysis seem a bit silly, there were two key moments in the game. The first was in the bottom of the fourth, after Charles Thomas walked to give the Braves first and second with two outs. Bobby Cox let Jaret Wright bat, which I thought was an inexcusable decision. I know why he did it--the injury to John Thomson left the Braves shorter on pitching--but it was a mistake:
- The Braves were down three runs, and conceivably might not get a better chance to tie the game
- Wright wasn't pitching that well, despite a low pitch count
- Wright wasn't going to pitch much longer, maybe six more outs at best
- Wright can't hit
In that situation, you have to take the chance of starting up the bullpen in exchange for the increased run potential. That was a huge moment in the game, with Oswalt not pitching very well and runners on base, and Cox let him off the hook in exchange for what turned out to be four more outs. Even without Thomson and with Paul Byrd
probably unavailable, the Braves had a lot of effective relievers available who hadn't seen much work in the series. Forget what happened later in the game; at the moment Wright stepped to the plate, there's no way that the expected difference between two innings of him and two innings of Juan Cruz
could have been worth more than the gap between having Wright and his 311 OPS at the plate instead of Julio Franco
or Eli Marrero
That decision didn't cost the Braves the game, but it reduced their chance of winning it.
The other key moment came in the middle of that big seventh inning. With two outs and Jose Vizcaino on second base, Craig Biggio lined a ball to right field that J.D. Drew just missed picking out of the air. He one-hopped it, allowing Vizcaino to score on the single.
The difference between Drew catching that ball and not doing so was maybe four, maybe six inches. A half-second at most.
The playoffs take place at the intersection of performance and something else, something we have yet to describe adequately. Biggio hit a ball hard. Drew made a good play, coming up just a hair short. The next two Astros hitters took advantage of their opportunities and ended the Braves' season.
Four inches. A half-second. And with that, some reputations are cemented, while others are remade.
Am I the last person still writing about that series? Probably. It seems everyone's attention has already turned to the East Coast, where the Yankees and Red Sox will play between four and seven games over the next week and a half.
Derek Zumsteg has done a bang-up job with the capital-P preview of this one, so I'll just make a few general points:
- The Red Sox rotation edge is really, really big. The name value of the Yankee starters hides this a little bit, but in any of the four scheduled matchups, would you actually prefer the Yankee starter? Kevin Brown has been a very good pitcher and makes a lot of money; I don't know that he's better than Bronson Arroyo right now, especially given Brown's back and hand problems. Tim Wakefield hasn't pitched well in the second half, but Javier Vazquez has been much, much worse.
Forget daddy-baby roles: Pedro Martinez against Jon Lieber is a mismatch. The Sox can run seven left-handed batters out there against Lieber, who even in his good years struggled against lefties. There's simply no game in this series in which the Yankees will have the edge in the mound matchup.
The Yankees really miss Orlando Hernandez, who isn't likely to see action in this series. He went from rehabbing to one of the better starters in the AL to out of action in about 12 weeks, which is an interesting year, to say the least.
- The Sox are much deeper than the Yankees. It's easy to forget that the Yankees are carrying six or seven dead roster spots, since a number of those spots looked good against the Twins. Hernandez might not even be able to pitch. Tanyon Sturtze and Esteban Loaiza, ALDS performances aside, are poor pitchers. Felix Heredia has never been successful as a lefty specialist, yet that's his role here. John Flaherty, like all Yankee backup catchers, plays once per postseason. Enrique Wilson has no tactical value. Tony Clark does, but Joe Torre can't see past Ruben Sierra, so Clark rarely plays.
The Sox have no dead roster spots, although Derek Lowe is mostly an emergency pitcher. Their bench provides everything but power, as the mid-season trades gave them additional speed and defense that allows them to get plodders Kevin Millar and Trot Nixon to the bench in the late innings. Dave Roberts and Pokey Reese can pinch-run in close games, which will be important not just for base-stealing attempts, but for trying to score on hits to these parks' relatively small outfields.
- The Red Sox are the better team. The Yankees' three-game edge in the standings at year's end didn't reflect the two teams' relative performances. The Sox were better at most aspects of the game, and clearly have the stronger roster.
Being the better team doesn't matter all that much when you're playing a best-of-seven.
I think these will be the keys to the series:
- The Yankee rotation. The Yanks got three good starts and one mediocre one against the Twins, which is encouraging until you remember that the Sox outscored the Twins by about a run a game this year. Mike Mussina is on his game, and will have to steal at least one win from Curt Schilling for the Yankees to advance. After that, it's hard to tell whether the Yankee starters are pitching well or whether it was just an easy week. I'm leaning towards the latter.
- Being Daddy. The Yankee "dominance" of Pedro Martinez is overstated. They did hit him hard in two September starts, but before that, he had a 3.65 ERA against the Bombers since 2002. That's not vintage Martinez, but remember that everyone's ERA should be higher against the Yankees.
The Yankees have always had a good approach against Martinez. Knowing that he's no longer a horse, they've worked counts against him early and made him throw a lot of pitches. With his drop in velocity this year, he can't just decide to throw a lot of strikes to counter this; his massive jump in home-run rate this year is attributable to the fact that he can't throw fastballs by people any time he wants anymore. The Yankees will have to keep working counts and taking advantage of that drop in velocity against him. The Martinez/Lieber matchups are the key games for the Yanks. If they can win even one of them, they'll have stolen a game that could mean the series.
- The Red Sox bullpen. You can rely on Keith Foulke, whether the Nation believes it or not. It's the guys in front of him who will have to be effective. Last year's series went 7.3 games in large part because Mike Timlin and Scott Williamson were unhittable. With Williamson out, Timlin has to shoulder more of that load or get help from Curtis Leskanic, who despite his shaky rep has allowed just two runs in 14 scattered innings since coming off the DL in August.
- The bottom of the Yankees' lineup. A key difference between the two teams is that the Sox have eight good hitters and Orlando Cabrera, while the Yankees don't go as deep, especially without Jason Giambi. The offense can break down through the Sierra/John Olerud/Miguel Cairo segment, killing rallies or leading to nine-pitch innings.
Torre has already solved some of the problem by turning to Kenny Lofton in Game One, sitting Sierra down. The Bald One's home run on Saturday notwithstanding, he has no business starting against right-handers. Against the Red Sox' all-righty rotation, the more Lofton plays, the better chance the Yankees have at long innings and crooked numbers.
- The matchup problem. As Mike Scioscia discovered, there are times when you want to be able to get a left-hander into the game against these Red Sox. David Ortiz and Trot Nixon are much different players against southpaws, and they bat in close proximity to one another. The Yankees have just one lefty, Heredia, and he's both not very good and not really a lefty specialist. It's an even bigger problem for the Yankees in that the Angels at least had four very good right-handed relievers they could use; the Yankees have two.
At some point, Heredia is going to have to get one or both of these guys out in a big situation. Whether or not he can do it will swing a ballgame.
At the start of the year, I picked the Red Sox to win it all. While the path they took to this point was a bit different that I expected, they're right where I thought they'd be: the best team in baseball, eight wins away from a title.
Then again, it's a short series. But as Jonah Keri said to me, as we did a post-mortem on the Dodgers/Cardinals tilt, "Sometimes the better team does win, even in a short series." Red Sox in six.
Joe Sheehan is an author of Baseball Prospectus.
Click here to see Joe's other articles.
You can contact Joe by clicking here