Replacement Level Yankees Weblog

Disclaimer: If you think this is the official website of the New York Yankees, you're an idiot. Go away.

by Larry Mahnken

The only goal left in the regular season aside from Jeter's batting title run and Posada's MVP run (and shot for 30 HRs and 100 RBI), which really aren't important at all, is Home Field Advantage. As I said earlier, that's not really that important, either (one game that they might not play in a series they might not get to against a team they might not play there), and accordingly, Joe Torre doesn't seem to be putting any focus on it. He's setting his playoff rotation now, and will give some of his starters days off this weekend.

The playoff rotation appears to be Mike Mussina, Andy Pettitte, Roger Clemens and David Wells, in that order.

Pitching isn't the most important part of baseball, but the pitcher is the most important player on the field (unless Barry Bonds is on the field). The school of thought that you win behind pitching and defense is flawed, because you need to hit some, too, but you can't win without at least decent pitching, and if you've got a dominant pitcher on the mound, you're already halfway there.

There are some "analysts" (*cough* Joe Morgan *cough*) who think that "Wins" are the best way to analyze a pitcher. The thinking is, of course, that the point of the game is to win, so if you have the most wins, or the best winning percentage, then you're the best pitcher. There are too many flaws in this line of thinking to go into, but the core of the problem is that just because the pitching statistic "Win" has the same name as the positive outcome of a ballgame, doesn't mean that they have the same meaning or value.

A much better way to evaluate a pitcher is by looking at his ERA, which tells you how many runs he allowed per nine innings pitched. Of course, to evaluate the overall value of the pitcher, you have to incorporate innings pitched into the analyis, but when we're looking at skill, rate is more important than quantity (though it's still important).

There are, of course, problems with ERA. Ballparks have an extreme effect on run scoring, as sabermetricians figured out before they were called sabermetricians, and even the most casual fan realized once they started playing baseball in Denver. A pitcher in Denver and a pitcher with the Dodgers could pitch equally well, but the Rockies' pitcher will have stats that look massively inferior to the Dodgers' pitcher. So, complex Park Factors were created, ERAs (and other numbers, as well) were adjusted, and pitchers were evaluated on a more even scale. With adjustments for the league in which they pitched, comparisons could be better made between ERAs. Suddenly, Don Drysdale doesn't look like such a worthy Hall of Famer (don't bother arguing with me about it. Go read "The Politics of Glory" first).

But you're still dealing with a flawed metric. Bill James wrote that much of what is considered pitching is actually defense. This is especially true for the Yankees, who have one of, if not THE worst defense in all of Major League Baseball. It's so bad that even the normally rah-rah Michael Kay acknowledges it, to a degree.

It's really difficult to seperate pitching from defense--probably impossible. But, you can get closer, and probably the most succesful effort to date is Voros McCracken's DIPS (Defense Independent Pitching Statistic). McCracken theorized that pitchers have no impact on whether a ball in play turns into a hit; that it's entirely dependent on defense, park and--a concept anemic to traditonal baseball thinkers--luck. It turns out that Voros wasn't quite right--pitchers have some control of what happens to a ball in play--but his basic premise was right. If you take what happens on balls in play and throw it out the window, you'll have a much better picture of the true abilty of a pitcher.

Voros's formula for DIPS is a bit complex, I've uploaded a worksheet to calculate it on my Yahoo! Group if anyone wants to calculate it for themselves. It uses batters faced, home runs, walks, strikeouts, intentional walks and hit batsmen, and adjusts for Home Run Park Factor and whether the pitcher is a lefthander or a knuckleballer. It can be daunting, so to make the statistic more palatable to the less than hardcore stathead, he concocted a "Quick and Dirty" formula that just uses innings pitched, hits, home runs, walks and strikeouts (that's on the worksheet, too), and isn't quite as accurate, but is close enough. Rather than go through every pitcher on every staff and calculate their DIPS, then add up their DIPS IP and DIPS ER to calculate DIPS ERAs, I used the Quick DIPS to come to this conclusion:

The Yankees have, far and away, the best pitching staff in the American League, and the second-best in baseball--very possibly the best.

Now I know what some of you are thinking: "That's crazy talk, Larry! You're crazy! Get off the road!" The Yankees' team ERA of 4.09 is 11th in baseball, more than half a run behind Oakland and about a third of a run behind Seattle for third in the American League. Sure, they're good, but the best? Come on.

But Seattle plays in an extreme pitchers' park, and while the Yankees have one of the worst defenses in baseball, Seattle and Oakland have the best. Seattle's pitching staff is actually painfully average. Using Quick DIPS, we see that the Yankees' expected 3.76 ERA is a third of a run better than Oakland's, and nearly 2/3 of a run better than Seattle's. The only team ranked ahead of the Yankees is the Dodgers, with a 3.64 ERA, and they play in the National League, where pitchers bat.

So, the Yankees have a great staff, maybe the best. But how does it break down?

Well, the first starter in the postseason rotation is Mike Mussina. Moose finishes the season with a 17-8 record, a 3.40 ERA, and is a legitimate top-five candidate for the Cy Young Award. His Quick DIPS is even better, his 3.27 ERA is second among qualified starters for AL playoff teams to...oh, take a guess. The Yankees signed him shortly after winning the 2000 World Series because they knew that they had sneaked by, and had to get better to stay on top. They could have opted to sign Manny Ramirez instead (or A-Rod, but that wasn't going to happen), and when they were shut down by the Diamondbacks' pitchers in the World Series, it looked like they might have made the wrong choice. The signing of Jason Giambi put that talk aside, since the Yankees now had a hitter of Ramirez's quality, and there had been no pitcher of Mussina's quality on the market in the 2001 offseason (unless you're a true believer in Jason Schmidt).

Every pitcher has bad days, Moose had one today, and he had one in Game 1 of the 2001 World Series, and Game 3 of last year's ALDS. But he's also the pitcher who shut down the A's in 2001 when the Yankees couldn't lose, and a pitcher who has come through with clutch pitching performances time after time this season. He's the Yankees' ace, and they'd be hard pressed to do better. Anyone they sign this offseason is a #2. He's not an A+, that would be the Pedro and Prior. But matched up against Santana, Pedro or Hudson, I feel good about the Yankees' chances. He's an A.

The Game Two starter is Andy Pettitte, who won 20 games for the second time in his career this season. Of course, he won't, and shouldn't, get serious consideration for the Cy Young Award, but just like about everyone else on the Yankees, his ERA is inflated by the poor fielding in front of him. Quick DIPS rates Andy as the fifth best starter in the AL playoffs, and the best #2 on any playoff team.

He takes a lot of heat from me for being so inconsistent, but in the second half of this season, Pettitte seems to have gotten it together, and he's only thrown one clunker in the second half, against Boston earlier this month. Andy got a mistaken reputation as a big game pitcher early, then a mistaken reputation as a big game choker. He is, in fact, neither. You've got Good Andy and you've got Bad Andy, and while you never know who you're going to get, the situation doesn't matter. It all depends on whether his mechanics are right, and lately, they have been. He'll be matched up in the first game of the ALDS against Brad Radke, who has been untouchable in September. But I think whether the Yankees win or lose that game is entirely dependant on which Andy shows up. If it's good Andy, they win, Bad Andy they lose. He's not an ace, but he's a damn good pitcher, and I've made up my mind: I want him back in 2004, even though he'll probably be overpaid. There aren't many better options on the market. I rate him a B+

In Game 3 it's Roger Clemens, and if things don't go well, it might be his final game. I could use this space to talk about how great Clemens has been over his career, and how he's the best pitcher who ever lived. I could criticize the Red Sox for letting him go, I could discuss whether he deserves to choose the logo to wear on his HOF plaque. Instead, I'm going to actually evaluate Roger Clemens right now.

Roger Clemens isn't the pitcher he once was. He's not going to strike out 15 men, and probably not even 10. But he doesn't walk many men, and when he needs it, he can still reach back and blow someone away. The problem is he can't rely on that anymore, and on the days he doesn't have it, he can't just throw heat and get away with it. Usually, the Yankees can expect him to give them a chance to win the game, and if he had gotten better bullpen support early in the season, he probably would have had a shot at 20 wins.

He's a good choice to pitch third, if he's pitching well, the Yankees will probably beat whoever he's facing in that spot, while it might be wasted against Santana or Pedro. On the other hand, if he's got nothing, he's going to lose, and it doesn't matter who's out there. That's the risk you take with a 40 year old pitcher, and for the Yankees, it's worked out so far. He's been good a lot more than he's been bad, and I'll rate him a B.

I'm not sure why, but Torre has picked David Wells to be his fourth starter. I'm sure it's because of experience, and maybe the trust issues that Torre has always struggled with. Boomer has a Perfect Game, and was dominant in the 1998 postseason, and with Joe that counts for something, he knows that he can handle the pressure. But it's not the pressure I'm worried about with Wells, it's his back. For the first half of the season, Wells was quite good. He had only 4 walks through June--that's FOUR--and his ERA was 3.41. He had a good shot at 20 wins, was a worthy All-Star candidate, and maybe even a Cy Young candidate. Then his sciatica started bothering him.

And when it was bothering him, he simply couldn't pitch well. His control was off, and he fell behind batters--and then he started having to come over with his stuff, and he got hurt. For nearly two months, through seven starts, Wells didn't win a single game. Mel Stottlemyre questioned his work ethic, and the general assumption was that he had worn out his welcome in New York.

But on Sunday, September 7th, in probably the most important game of the year for the Yankees, he beat the Red Sox, giving up only 1 unearned run into the 8th. He's pitched "okay" since then, giving up 10 runs in 22 innings, but that one start against Boston was apparently enough. Possibly because he felt he had no other option, Torre named Wells his fourth postseason starter.

Wells isn't bad, of course. The Red Sox would very much like to have him on their roster for the playoffs, where he could be their 2nd starter. But if his back is sore--and you know he won't tell anyone beforehand--he's toast, and if he's pitching in the first round, that means that the Yankees are either on the brink of elimination or trying to avoid a 5th game--and you really don't want a potential implosion on the mound. I'm not saying he's going to fail, but he might, and if he does, it will probably be miserably. Healthy, he's a B-; hurting, he's an F. I'll rate him a C, and hope we see the B-.

Torre probably stuck Wells in the rotation because he didn't feel he had a better option. But I think he did: Jose Contreras. Believe it or not, Quick DIPS rates Contreras as the best starter the Yankees have. He's not of course, but he's almost certainly better than Wells, even a healthy Wells. His style (strike out a lot, walk a lot) is well suited to the Yankees' defense--and he doesn't give up many home runs, either. The problem is, he's made nine starts, and excluding the start against Detroit where he got hurt, he only had one bad start--against Boston. But then, Wells had an even worse start against Boston on the Fourth of July, Clemens had three of them, and Pettitte had one, too. But then, Contreras hasn't pitched well against them yet, either, doesn't have any rings, doesn't have a perfect game. He's a risk, that's for sure, but I don't see him as any more of a risk than Wells. I'd rate him an optimistic B, hope that the Yankees get to the ALCS and he replaces Boomer, and for next year, feel a little better about the rotation.

After looking at how I've rated the Yankees so far, you start to realize what a great team they have. It's not as good as it has been, but it's still just about the best team around.

Tomorrow, the area on the team that started as a total disaster and now is starting to look okay: the bullpen.

--posted at 12:01 AM by Larry Mahnken / |

Larry Mahnken and SG's
	Replacement Level Yankees Weblog
"Hey, it's free!"