AL Cy Young Award Winner 2004

[Crash Course]

Who will it be? Who should it be? Why?

[KCBOOMER]

There simply is no discussion here. It is Johan Santana followed by a distant Schilling. The Cy Young is for the best pitcher. It is not a goofy "most valuable" award type thing.

[Crash Course]

Betcha it's close between the two - based on the Clemens selection, I could see Schilling getting a lot of votes.

[KCBOOMER]

I think the media upswell for Clemens began during the season. Schilling's started during the post-season. Most of the ESPN talking heads were really talking up Santana.

[Crash Course]

Santanna was about 28% better than Schilling this year. He should win.

[hopbitters]

Based on the NL CYA, I could see Carlos Santana getting the award. The only pitcher the voters know anything about is full of Milwaukee's Best, and I do not mean Ben Sheets.

[RascalJones]

Carlos Santana, now THAT would be a feat. ;)

He's so Smooth. ;)

[Crash Course]

It's unam: http://mlb.mlb.com/NASApp/mlb/mlb/news/mlb_news.jsp?ymd=20041111&content_id=912848&vkey=news_mlb&fext=.jsp

The only thing that bothers me with Santanna is that he was terrible for the first 1/3 of the season. Should consistency be a factor with an award like this? Or, is it, 20 years from now, the end numbers is all that anyone will recall?

[RedSeat]

Inconsistent performances will generally show themselves as they affect the overall stats. If he was able to put up the best overall numbers in spite of a poor start, more power to him.

That's about as close to a unanimous 1,2,3 as I remember seeing. And I agree with the order as well.

[KCBOOMER]

The season is what matters, not some portion of it.

[Crash Course]

I would have gone with Brad Radke 3rd, maybe even Mark Buehrle, over Rivera.

[RedSeat]

Is there any chance that Rivera's second place vote wasn't George King?

[Crash Course]

Dunno.
But, it will probably be leaked soon.

[mainsr]

This was one of the easiest votes in years. Almost no chance of the voters screwing it up.

[hopbitters]

This was one of the easiest votes in years. Almost no chance of the voters screwing it up.

Johnson led the NL candidates in ERA, ERA+, RSAA, BR/9, K/9, VORP, WSAA, FIP, starts, IP, NW/L, and perfect games and was second in K/BB and CG to Sheets and they screwed that up. They've always got a puncher's chance.

[mainsr]

Yeah, but Johnson didn't win 20, so that it made it much harder for the voters to get it right. He didn't "just know how to win" or "pitch to his run support."

[captain_napalm]

The season is what matters, not some portion of it.
Ya,

that's why they have AL & NL "pitcher of the month" awards :)

[Throwback]

I don't think a closer (or, more generally, any pitcher with under 100 innings) should be able to win the award unless you could practically double his ERA (more or less depending on actual inning totals) and still justify his winning. To me, ERA+ and IP are the only stats that matter a lick. Not Wins, because they say zippo about the actual skill of the pitcher, and not K's because they're no more effective than a popup (and potentially waste more pitches) and whatever benefit they gave the pitcher will show up in his ERA. If you apply that kind of penalty for low-inning pitchers, truly historic relief efforts like Eckersley's 0.61 ERA season would warrant strong consideration still, but otherwise the award would be firmly in the domain of the starters who do the bulk of the work.
Thus, I don't think Rivera should have received any votes at all, nor do I think Gagne should have beaten Schmidt last year.

[PianoMonkey]

I gotta agree with Throwback that IP needs to factor more strongly than it seems to in the CYA, but I must nonetheless take offense at the notion that K's waste pitches.

I would agree that the majority of the benefits of a pitcher's strikeout totals are already represented in their ERA+, but that doesn't mean that K's are no better than popups. I don't have the numbers (yet), but even if a pitcher manages to "waste pitches" on 10 K's a game, that would only realistically add like 6-7 pitches on average to a 5-K game, right? Even if Pedro does magically fall apart at pitch 106 (BTW, that was the most entertaining thread I've seen here in a while, Crash :D), those five extra K's usually won't make a lick of difference. To me, the slight increase in pitch counts is more than made up for by the immediate reduction in balls-in-play.

[Throwback]

I gotta agree with Throwback that IP needs to factor more strongly than it seems to in the CYA, but I must nonetheless take offense at the notion that K's waste pitches.

I would agree that the majority of the benefits of a pitcher's strikeout totals are already represented in their ERA+, but that doesn't mean that K's are no better than popups. I don't have the numbers (yet), but even if a pitcher manages to "waste pitches" on 10 K's a game, that would only realistically add like 6-7 pitches on average to a 5-K game, right? Even if Pedro does magically fall apart at pitch 106 (BTW, that was the most entertaining thread I've seen here in a while, Crash :D), those five extra K's usually won't make a lick of difference. To me, the slight increase in pitch counts is more than made up for by the immediate reduction in balls-in-play.
First of all, I have to express my alarm that we've now used the word "lick" in two consecutive posts (or is it 3 now :confuse: ). Anyway, I'm hardly trying to say strikeouts are a bad thing, but I think most pitchers (or especially most managers) would agree with me that they'd rather get a first-pitch popup on the infield than a 4 or 5 pitch strikeout, and yet these don't show up on the stat line, even though runners are even less able to move up on them than on a strikeout. All I was really trying to say was that it's a glamour stat that really doesn't tell you anything about the pitcher's ability, just about his style.

[mainsr]

Well, yes, but. Certainly watching vintage Maddux retire a side with five pitches--popup, grounder, grounder, no balls--was breathtaking. An out's an out. That being said, there are plenty of empirical analyses that indicate that strikeout pitchers last much longer than others, and more recent work has shown that to some degree, everything a pitcher does other than Ks, BBs, and HRs is somewhat random.

[Throwback]

Perhaps strikeout pitchers are just used to throwing more and thus are better conditioned, sort of like the starters of old.
As for the random theory...that's just crazy. Who told you that, Bill James? If that's the case, finesse pitchers should be entirely interchangeable, and Maddux and Glavine are nothing more than statistical anomalies...their entire careers are just huge coincidences. You don't think pitchers can utilize location to usually get the kind of contact they want? You don't think a good sinker on the outside corner down isn't going to tend to get a nice double-play grounder more often than other pitches?

[huskerdru]

Well, yes, but. Certainly watching vintage Maddux retire a side with five pitches--popup, grounder, grounder, no balls--was breathtaking. An out's an out. That being said, there are plenty of empirical analyses that indicate that strikeout pitchers last much longer than others, and more recent work has shown that to some degree, everything a pitcher does other than Ks, BBs, and HRs is somewhat random.

I believe Voros McCracken is the best-argued purveyor of such empirical analyses...http://www.baseballprospectus.com/article.php?articleid=878 here's his DIPS thesis, which is based in large part on the random-nature-of-balls-in-play hypothesis.

Whether most of it is shown in ERA (or ERA+) already, the simple fact is that more strikeouts means fewer batters putting balls in play. Whether random or not, fewer balls-in-play = fewer men on base = fewer runs.

[mainsr]

Thanks, Huskerdu, you're right; it's Voros McCracken's research I'm citing. Sorry, Throwback, there is empirical evidence to back up McCracken's position. I'll try to dig up some data to support it, specifically in the cases of Maddux and Glavine. They are certainly not finesse pitchers. Hang on.

[Throwback]

Thanks, Huskerdu, you're right; it's Voros McCracken's research I'm citing. Sorry, Throwback, there is empirical evidence to back up McCracken's position. I'll try to dig up some data to support it, specifically in the cases of Maddux and Glavine. They are certainly not finesse pitchers. Hang on.
Well I just checked the numbers, and it's true that their strikeout numbers are hardly abysmal, and thus perhaps I pigeonhole them a bit too much by assigning the "finesse" tag to them. However, although at times both of them have at least approached 200 Ks in certain seasons, neither has come very close to a 1:1 strikeout-IP ratio, which to me, a modern power or strikeout pitcher should at least flirt with from time to time in their career in order to be good. Whatever data you trot out there, the best it can do is to declare Maddux and Glavine to be either exceptions to the rule, or statistical anomalies. They've always succeeded by putting the ball in locations that get them the type of batted ball they desire. If this isn't a key part of pitching, then guys who can't consistently break 90 MPH should have no place in the majors, because since they don't strike people out much, their performance should be inevitably determined by luck of the draw.

[mainsr]

Whatever data you trot out there, the best it can do is to declare Maddux and Glavine to be either exceptions to the rule, or statistical anomalies.

Glad to see you have an open mind.

OK, this is going to be crude, because I don't have the inclination to do all the translations, so let's just do this: For each pitcher, I subtracted BB, K, and HR from BFP. I also subtracted SH and HBP. I then calculated hits divided by the remaining BFP, which is a proxy for balls put in play.

Here's an example: Clemens in 1996 (last year in Boston, went 10-13): 1032 BFP, 216 H, 19 HR, 4 SH, 4 HBP, 106 BB, 257 K. That means out of 1032-19-4-4-106-257=642 balls in play, he allowed 216 H, a .336 pace.
In 2001 (Cy Young, 20-3): 918 BFP, 205 H, 19 HR, 4 SH, 4 SF, 5 HBP, 72 BB, 213 K. 601 balls in play, 205 hits = .341. No significant difference.

But your point is that some pitchers, by virtue of good ball placement or whatever, are always ahead of the game. Let's take six: Maddux and Glavine (control guys), Clemens and Johnson (strikeout guys), Jeff Fassero and Jamie Navarro (mediocrities)

Year by year, min. 300 BFP
1988: Maddux .286, Glavine .297, Clemens .315
1989: Glavine .280, Maddux .290, Clemens .310, Johnson .311, Navarro .317
1990: Johnson .290, Clemens .301, Maddux .310, Glavine .330, Navarro .337
1991: Glavine .283, Clemens .292, Maddux .297, Navarro .300, Johnson .313
1992: Maddux .261, Navarro .273, Glavine .277, Clemens .288, Fassero .302, Johnson .304
1993: Fassero .287, Maddux .287, Glavine .300, Johnson .309, Clemens .312, Navarro .345
1994: Maddux .260, Clemens .286, Fassero .301, Johnson .332, Glavine .339, Navarro .387
1995: Maddux .258, Glavine .296, Navarro .307, Johnson .325, Clemens .346, Fassero .358
1996: Maddux .295, Glavine .311, Fassero .325, Navarro .329, Clemens .336
1997: Glavine .276, Maddux .294, Clemens .304, Fassero .317, Johnson .325, Navarro .372
1998: Maddux .280, Clemens .289, Glavine .294, Fassero .333, Johnson .361, Navarro .378
1999: Glavine .332, Johnson .342, Clemens .342, Maddux .345, Fassero .404
2000: Glavine .297, Maddux .300, Clemens .318, Johnson .368, Fassero .370
2001: Glavine .309. Maddux .314, Fassero .332, Clemens .342, Johnson .352
2002: Glavine .299, Maddux .304, Johnson .333, Clemens .358, Fassero .368
2003: Maddux .316, Clemens .331, Gllavine .331, Faserro .378, Johnson .399
2004: Glavine .289, Johnson .294, Clemens .301, Maddux .339, Fassero .348
CAREER: Maddux .300, Glavine .302, Clemens .313, Johnson .329, Navarro .333, Faserro .339.

Big caveat: Since I didn't adjust for park effects, I'm understating the average for Glavine, Maddux, and, when they played in Seattle, Fassero and Johnson, while Clemens' is overstated. Given that, I'm not sure you can say that Glavines' figures are better than the others. I concede the point about Maddux, but I think the inconsistency in these numbers from year to year suggests that there is a pretty large random element.

[Throwback]

Big caveat: Since I didn't adjust for park effects, I'm understating the average for Glavine, Maddux, and, when they played in Seattle, Fassero and Johnson, while Clemens' is overstated. Given that, I'm not sure you can say that Glavines' figures are better than the others. I concede the point about Maddux, but I think the inconsistency in these numbers from year to year suggests that there is a pretty large random element.
Well I'm not saying that randomness isn't a major factor. One only needs look to seeing-eye singles and texas leaguers to prove that hypothesis. My contention is that randomness is not the governing principal on balls put in play. If it were, there would be no way for a non-strikeout pitcher to have consistent comparable success with a strikeout pitcher, since only strikeouts, HRs, and BBs could set them apart. I also would not argue that strikeout skill and overall skill don't often go hand-in-hand, but to assume that they always must would be a mistake. Keep in mind, also, that your BA numbers don't take into account important run-related numbers like double plays, runners advanced, and extra-base hits.

[mainsr]

My contention is that randomness is not the governing principal on balls put in play. If it were, there would be no way for a non-strikeout pitcher to have consistent comparable success with a strikeout pitcher, since only strikeouts, HRs, and BBs could set them apart.

Substitute pitchers for a pitcher and it's pretty much true: Non-strikeout pitchers do not have consistent comparable success as strikeout pitchers, period. There are exceptions, of course, but in general, that statement is true. Check out this:

There were 40 AL pitchers who qualified for the ERA title.

The top 10 strikeout pitchers (in terms of K/IP) had a combined 4.06 ERA.

The next 10 had a combined 4.49 ERA.

The next 10 had a combined 4.63 ERA.

The next 10 had a combined 4.76 ERA.

Same thing in the NL: 45 qualifiers -
Top 11 were 3.16
Next 11 were 3.86
Next 11 were 4.23
Next 11 were 4.37

You'll find it over and over again. The best strikeout pitchers are the best pitchers, period. The worst strikeout pitchers are the best pitchers, period. There are exceptions. But the rule is ironclad.

[huskerdru]

James discusses this in an essay in the NBJHBA. His hypothesis is, essentially, that no pitcher in history has had sustained success (I forget how he defined that, but something like 5+ years above average, or whatever) with a K/9IP of worse than about 5.00 (or just under). Maddux is at about 6.3 for his career, which is quite good, though clearly not excellent - it's good enough, though

The main exceptions to his low-strikeout-pitchers-have-short-careers rule that he cites are Jimmy Key, Tom Glavine, Tommy John...calling them the guys who

"define exactly how many strikeouts you absolutely have to have to win consistently in the major leagues. If you're left-handed, and you do everything well, you can win consistently even if your strikeout rate is a little below the league norm."

Otherwise, he pretty much scoffs at the notion that any pitcher can sustain success with a low K/9IP rate...except Lew Burdette, his proverbial exception to the proverbial rule. Burdette survived a low K/9IP with his excellent K/BB ratio. Otherwise, no one else sice WWII "breaks" the "rule." I've tried...everyone I think of who has pitched well with a low K rate ends up, upon empirical assessment, either: 1) Actually possessing a higher K/9IP than I ever would have guessed - like Maddux; or, 2) mostly sucking - like Kirk Rueter.

The essay is at the end of the 1970s chapter...

[Throwback]

Well if you set the standard low enough, of course it will ring true. In today's game, hitters often flail so wildly that they will miss both well-placed pitches and not-so-well-placed ones. If a guy doesn't have about 5 Ks per 9, it probably indicates that his pitches are ending up where the hitter wants them more often than where the pitcher wants them. So by saying that a guy with worse than slightly under league average K/9inn can't be consistently dominant, James is basically saying guys that suck can't be consistently dominant. What an astute observation.

Besides, I've never denied a general correlation between good K numbers and good ERA. I've always said that Ks contribute to ERA, because a K is a really good way to keep runners from moving up and scoring. However, a great pitcher can get his results from groundouts in the right situation, double plays, popups, and pitching to his defense. Strikeout pitchers are, however, at an inherent advantage because they don't need to do quite as well in their non-K batters as the others do.

Because of this advantage, strikeout pitchers in general will be better, but non-strikeout pitchers still have the opportunity to make up their disadvantage with exceptional finesse, so they can not be counted out. Think of it this way: people indicted by a grand jury are way more likely to be guilty of a felony than the general population, but indictment is certainly not automatic grounds for a conviction in the individual cases.

[PianoMonkey]

non-strikeout pitchers still have the opportunity to make up their disadvantage with exceptional finesse, so they can not be counted out.

I'd definately agree with this, but I'd say the number of so-called "finesse" pitchers who actually deserve the title (i.e. those that are capable of routinely inducing non-random out-producing contact) are exceedingly rare. My baseball team is currently (or was, at the end of this season) home to approximately 1000 innings of "finesse" pitching, so I know what I'm talking about here.

[Throwback]

I'd definately agree with this, but I'd say the number of so-called "finesse" pitchers who actually deserve the title (i.e. those that are capable of routinely inducing non-random out-producing contact) are exceedingly rare. My baseball team is currently (or was, at the end of this season) home to approximately 1000 innings of "finesse" pitching, so I know what I'm talking about here.
Agreed, also. Isn't the world a beautiful place?

[PianoMonkey]

Not to make the world ugly again, but just to clarify: I mean, there's been like maybe 3 or 4 of them (not counting knuckleballers--who ever does?)

[huskerdru]

Well if you set the standard low enough, of course it will ring true...So by saying that a guy with worse than slightly under league average K/9inn can't be consistently dominant, James is basically saying guys that suck can't be consistently dominant. What an astute observation.

Well, I think you've missed a lot of his argument...so don't knock him for my inability to articulate it well!! James didn't set the standard...he arrived at it through analysis. I threw the league average thing in there on K/9IP...would have to re-read his essay to understand better what his threshhold is. The point, though, is that it's more of an absolute than a relative threshhold...that is, it's not so much about league average, as about being above (or below) a particular number (which could be a lot worse, or better than, league avg). That threshhold represents the point at which a low-K pitcher must have some exceptional "other" skills/qualities (e.g., pinpoint placement-which few do, good speed changes and differential, good scouting/knowledge of hitters, good slide step to keep runners in place, being lefty, etc.) to offset the fact that more balls-in-play (things being equal) = more hits and runs. If you don't mitigate it, the TENDENCY toward randomness in distribution of balls-in-play will catch up to you fast (just ask Esteban Loaiza).

In today's game, hitters often flail so wildly that they will miss both well-placed pitches and not-so-well-placed ones.

This is wild speculation. It assumes that 1) pitchers are less patient and skilled today than in the glory days of yesteryear (I don't have data handy, but I'd wager this is not the case); 2) strikes are primarily a matter of placement (when it's just as much about speed differential). There have always been freeswingers who miss good and bad pitches to hit...again, I'd wager a dollar to the dime that there are way FEWER such "bad" hitters in the majors today than there were, say, 10, 20, 30, 50 years ago - MANY fewer.

If a guy doesn't have about 5 Ks per 9, it probably indicates that his pitches are ending up where the hitter wants them more often than where the pitcher wants them.

Precisely (if "where" means at the location and speed the hitter wants)!