"You cannot measure a player's skill or contribution to the tenths of points. The formulas are very useful until they become dogma. Then they are harmful. Dogma is always harmful"
I fully agree with your statement. Unfortunately it seems that:
1) everybody believe to have his own infallible formulas
2) even worse formulas are chosen and then used just to prove something they've already dediced to state
3) thus we have a so called "science" which on the contrary is banal alchemy"
KILL SABR
Q. If it were up to SSI, would Felix win the Cy Young?
A. Probably not. It's a pretty close race this year, and under those circumstances it's hard to give it to a .500 pitcher for a 100-loss team. That kind of pitcher should be a lot better than his comp to win it.
I might plump for Sabathia or Liriano.
..................
Bill James once called Cy Young voting "a tradition in the worst sense of the word, a tradition in the sense that we know we're doing it wrong, but we're going to keep doing it that way because that's how we've done it for a long time."
He was referring to the fact that Cy winners -- at the time -- were chosen mostly because of won-loss records.
I think that won-loss records should be considered much less than they are, but they certainly are not irrelevant. There is a whale of a lot of difference between winning a game that matters, and losing a game that doesn't matter. Buchholz and Price and Sabathia are performing in different, much harder, contexts.
.................
That's not to say that Felix couldn't do the same. But we're not talking about what might be or could be or should be: we're talking about what is. In 2010, the guys who are winning for Boston, New York, and Tampa, ARE IN FACT winning tougher games.
If Pedro Martinez led the league in ERA by 2 runs, but had a 10-14 record, sure, I'd give him a Cy for it. But if you're not pitching in games that matter, you've got to be a good bit better than the other guys.
Pedro never had a 10-14 record. Guys like that don't, not very often. The best pitcher in the league tends to have a very good W-L record.
We read somewhere that the worst W-L percentage in the history of the Cy was .600 -- like, 24-16 by Gaylord Perry or something. I approve of this as one, flexible, consideration.
...................
There will always be those who think that HOF awards and All-Star placing and Cy Young and MVP awards should be determined by computer. Any deviation from WAR is a deviation from "correct" thinking, right?
One time, online, a user challenged James about his top 100. "What gives you the right to arbitrarily change the records?" Bill's acid reply: "We do not have near-perfect measurements of baseball players. It is foolish to assume that we do." That'll do for me too.
You cannot measure a player's skill or contribution to the tenths of points. The formulas are very useful until they become dogma. Then they are harmful. Dogma is always harmful.
...................
Right now, the James predictor has the lead firmly to CC Sabathia, partly because he's just about as good as anybody else, partly because he's thrown so many innings. We might note that he is the one who is handling the 9,000 PSI in New York. Few can.
Fangraphs value has Liriano ahead of Felix in $ value, and Liriano is leading his team to the playoffs. Anyway, the bottom line is, it's a pretty close race this year, and under those circumstances it's hard to give it to a .500 pitcher for a 100-loss team.
But it would be perfectly fine to give it to Felix. His ERA is almost a run better than Sabathia's and if he gets it, he won't have to back up to accept it.
.....................
I'll tell you this. We have the best pitcher in baseball. That's a lot more important than having the pitcher who won the Cy Young.
.
Cheers,
Dr D
Comments
:daps: anon.
Actually, my motto is long live sabermetrics -- kill saberdog.
Respect for the complexity of the problem, gentlemen...
For our lurker amigos, I should spell out the fact that this particular post is just a driveby.
I have no idea who should be the Cy Young this year; to have a better zero on that I'd usually need to care a little more about major league baseball.
But, on this one, I'd caution against the dogma that W-L records constitute non-information. W-L records usually carry significant information.
I notice that of the 14 pitchers with 15 or more wins right now, all 14 of them are real good pitchers. So I have an idea that W's carry some kind of information or other.
................
Felix is great, of course.
Why does C.C. Sabathia deserve the Cy Young more than Felix does just because he's on a better team? That makes no gosh darned sense. Sabathia's had an EASIER time of it than Felix. WHen your team scores you fewer runs than anyone in baseball...ANYONE IN BASEBALL!!...by a wide...wide margin...if you go .500, that means you are PHENOMENALLY skilled at winning. I'm far more impressed with Felix' ability for clutch pitching than I am with C.C. Sabathia.
I don't understand your philosophy that you you'd vote for Sabathia just because he has more wins if the two pitchers are close together in value...which, BTW, they aren't. Felix is a run better per game than Sabathia. That's not tenths of a point in WAR...that's A RUN BETTER! PER GAME! But whatever.
There is no good second choice for me...Felix Hernandez should win the Cy Young...he won't, but he should.
It is true that the Cy Young Award is not purely to go to an MVP, most valuable pitcher. But value, in terms of what you mean to your teams ultimate success, should be a part of that.
In that context, Sabathia as a terrific stopper-type ace in the middle of a hot pennant race in a tough city is more valuable to his franchise. His performance may get the Yankees to the WS. Felix's may get the M's to avoid losing 102 games. Felix's is clearly the "better" year if you look at the "pure" pitching things he can control. But (and here's the best pitcher comparison I can think of) it's not a Steve Carlton-'72 type of year. Carlton played for the worst team in the league that year, won 50% of their games, started 41 times and finished them 30 times..and threw 8 SHO's. Oh..he threw 346 innings. his OPS+ was only slightly better than Felix's (182-175) but the other numbers suggest a truly deserving guy. Felix would need, I think, that kind of performance to win.
Most Valuable Player: In 1979, 39-yr old Willie Stargell (of the "We are Family" Pirates) co shared the MVP with Keith Hernandez. Check it out, Hernandez had the clearly better year. Dave Winfield, who finished 3rd in the voting had a better year than Stargell. Larry Parrish, 4th in the balloting, was arguably better. Heck, Dave Kingman and Mike Schmidt, whot finished 11th and 13th in the voting had better years! They were all "Better" ballplayers. But they were not "more valuable" in the context of their teams success and their part in it. The Bucs won the LCS and the World Series. they were a renowned "family-type" bunch (see Mariners '09). Papa Stargell had that team together...and put uup some pretty good numbers. his team won the prize and he was the still-producing glue that got them there. He deserved it. He was "most valuable" to the great success his team achieved. Winning surely counts in that award. It should be a consideration of some degree in this award, too.
The Cy Young award can not be purely numbers (SABR) based. If you had a pitcher who started and threw 1 1/3 innings in each game and never gave up a run, he would throw 210 innings with an ERA of 0.00! Would he be a better pitcher than a guy who won 20 games with an ERA of 2.00 and pitched deep into most of his starts for a pennant winner? Certainly not. Ask any baseball guy. Which starter would you rather have. the first guy or the guy who gets you into the 7th inning+ in 25 of his 3 starts with a very low ERA. SABR type numbers can't always rule.
Felix is the best pitcher in the game. Check out the numbers of his last two seasons and look how amazing consistent(and outstanding) they are. But if Sabathhia finishes strongly..he should nose out Felix for the award.
BTW, with all the kinesthetic talk around here, I'm surprised nobody has mentioned Felix's quasi-Luuis Tiant, turn your back delivery. Some insight on whether that reduces arm stress would be interesting.
moe
Felix went 19-5 last season and came in close second. He's been even better this season. Felix is carrying the entire franchise on his shoulders. He goes out there every fifth day knowing he has to be near perfect to give his teammates a chance to win. If he's the best pitcher in 2010, he deserves the award given to the best pitcher of 2010. If he isn't, give it to whomever is. But don't tell me Felix hasn't pitched under any more stress than any other pitcher, particularly those who have the luxury of plenty of backup from a real supporting cast. Felix has had to carry this team by himself. Heavy is the head that wears the crown. Assuming Felix continues to be spectacular through September, then a Cy Young is fitting. If future generations can't merely look at a won loss record to understand why, then let them look a little more closely at the stats. The reason will be made clear.
I think Liriano deserves it this year, although you could go with Felix or Lee too.
Just being on a better team doesn't mean anything. I agree with you about that.
Sabathia is playing in tougher games *and winning them.* In some respects, it's like being in AAA as opposed to being in AA.
You're a New Yorkah, right? How many guys have been dropped in that furnace and melted?
After all, that's supposed to be the definition of the award, the best pitcher in baseball.
I'm not married to my position, that hitting .300 in a pennant race is different from hitting .300 in Kansas City. But I put something of a thumb on the scale for performing when it matters.
..................
As far as your last couple sentences, Rick, the voters who consider W-L don't have an understanding gap. They know what WAR is. They simply judge W-L to be significant.
We don't need future generations to comprehend the difference between the two philosophies. They understand it now, but disagree with the computer-chosen winner concept.
Hey, the sabermetricians still argue among themselves as to what the criteria should be for winning the Cy Young award. There's consensus on Felix THIS year, but there's no consensus on what constitutes a "correct" choice in the abstract.
Most of us think of the Cy as "MVP for pitchers," though it's not.
If the 4th-best hitter in the league hit 10 homers -- against division rivals -- in September and single-handedly decided a pennant race, would he have been more VALUABLE than otherwise?
Even in the def of the MVP, it's clear that when and where you produced is a factor.
If Cy Young = best stats ... I agree that Felix has had a much better season than Sabathia.
We're not arguing about that. We're arguing about what the criteria for the Cy should be.
......................
But if Cy Young = best stats, then Liriano has more $ value than Felix, AND he's done it in a pennant race.
It's not like Felix clearly has "the best stats." Do you use WAR and $ value, or do you use xFIP, or do you use K/BB ratio, or do you use ERA, or what?
....................
Part of the subtext here, is that I don't care for sabertistas' focus on the hypothetically best pitcher (if you assume HR/F and BABIP and S% should have been league average).
Though that isn't a complaint against Felix' 2010.
In some cases I think its even more impressive for player to perform so well on a terrible team.
Its much harder to maintin focus and motivation on a losing team.
He's on one of those 27-for-30 quality start type runs, like last year -- he brings his A game every night, 100% regardless of what's going on around him.
He's the Clemens of the decade, totally cemented in that now.
Definitely arguments to be made for Liriano, Lee, Sabathia & Felix (and I guess Price). In Felix' favor I think innings pitched and the fact that the M's offense hasn't been just bad but genuinely historically bad - worst of the DH era bad should count for something.
Is there a reason to consider Felix as the best "numbers" CY candidate? Sure. But, it depends on which numbers one looks at and which are ignored.
Felix has a great K rate, (but Morrow, Lester, Weaver, Liriano and Colby Lewis all have better).
Felix has a great K/BB, (Lee, Weaver, Shields, Marcum, Liriano, and Baker are all better).
He's got a great ERA+, (only Bucholz is better).
He's second only to Lee in IP/GS.
Only Lee, Cahill and Weaver had a better WHIP.
*BUT* -- Felix also leads all the potentials (by a WIIIIIDE margin) in unearned runs allowed, (17). The typical number for most of the guys is between 6-9.
The thing is -- if one starts looking at stats - and suggesting W/L doesn't matter, why is Jered Weaver not in the discussion? Better K rate - better K/BB rate - top 3 in QS% - 2nd in Ks. He's also 11-11. Yet, nobody is suggesting him. Why?
In the end, it's because the tendency for most "number" guys is to go and utilize the stats that help support the desired conclusion. (I've been guilty of this myself, and will likely be so in the future, too). But, absent a formula to use *BEFORE* the season begins, there is going to be a subjective element to the selection, regardless.
By FIP and xFIP, both Liriano and Lee come in ahead of Felix. (Grienke won in 2009 while leading in FIP, btw). Felix wins by WPA, (Pirata will be thrilled!).
Which is the "right" stat to use?
I agree with Doc's basic sentiment, though. While the CY was often a joke by relying almost solely on Wins for so many years - it is a mistake to dismiss them COMPLETELY. Should Nolan Ryan have won the Cy Young in '87, when he lead the league in ERA and Ks, but went 8-16? Or would giving the CY award to an 8-16 pitcher kind of insult the basic value of all pitchers. (Pitching is sooooo unimportant in terms of winning and losing, the best pitcher in the league can lose 2/3 of his decisions).
There's been talk about all the innings Felix has pitched. Would he have pitched those if the Ms had Rivera in the pen? Is there a point at which we start rewarding SPs for having crappy instead of good bullpens?
Honestly, the guy *I* currently lean toward, (in a very crowded race), is C.J. Wilson. Why? Context. (which is the heart of Doc's argument, I think). Felix is a great pitcher, throwing in a great pitcher's park, with a solid defense behind him in games that mean nothing. He's got about every advantage a pitcher could have to rack up great PERSONAL numbers. He's even managed to slosh 17 runs into the unearned column, twice what the other potentials have done. And when Lee was with Seattle, he out-pitched Felix ... easily.
Lee, in Texas, has been another story. Suddenly those warning shot outs are flying over the fence. In 88 innings with Texas, Lee has surrendered 10 HRs, (he gave up 5 in 103 IPs with Seattle). C.J. Wilson has been throwing in the Texas context all season - giving up 8 HRs in 180 innings. Felix has allowed 15 in Seattle.
Of course, C.J has worse K and BB numbers. But, he's *GOT* to pitch differently in Arlington than one pitches in Seattle. Two games in Arlington this season, Felix has a 6.23 ERA with 2 HRs allowed in 13 innings. For his career, Felix has a 4.44 ERA in Arlington with a 7.6 K/9 and 2.09 K/BB, (both numbers almost a full point lower than his career figures).
ERA is a poor stat. We get this. Yet, you stick a plus on the end of it, and the pretense is that the stat is tons better. Well, it's a LITTLE better - but it's just a minor polishing of a stat this is still flawed. The park adjustments aren't perfect - and they don't fix most of the underlying problems with ERA as a stat to begin with. But, ERA tells you "something". Just like QS% tells you "something". Just like wins and losses tell you "something". But, even FIP and WAR don't take into everything there is about a pitcher. SHOULD a pitcher be trying to strike out the side when he's up by 7 runs? Or, should he just be pitching to contact (and avoiding HRs)?
Yes, lack of run support hurts the W/L record. But, lots of run support likely does (and should) skew some of the peripheral numbers, too. You do *NOT* pitch around Branyan when you're up by 7 runs in the 3rd inning. You don't intentionally walk Ichiro when up by 10. Pitching is inherently linked to context.
Greinke has a K/BB almost identical to Felix'. He's got run support exactly equal to Felix. But, his ERA is 3.90? Why? Because he's got the worst defenses in baseball behind him.
Should Felix be in the discussion? Absolutely. But, Felix has the #1 easiest context one could hope to put up great individual numbers. CC has great offensive support - but plays in a HR-dangerous park, in a city notorious for pressuring its sports stars - in the midst of a pennant race. Wilson plays in the likely #1 hitters park in the AL - in a pennant race - for a team that hasn't done anything in a decade - where the guy who was even better than Felix (while in Seattle) has largely struggled since arriving.
I don't know of any stat that is adept at putting ALL that context around the raw stats.
I can see a half dozen potentially deserving CY candidates. The guy I view as having accomplished the most while facing the greatest hurdles (so far) is Wilson. Maybe that isn't the "best" way of looking at things. But, I definitely agree with the Doc that "context" has to be a part of the discussion. Yes, FIP and WPA are and K/BB are great to look at. But, even the stats that are concerned with context don't capture ALL the variables of context. Wins is simply another stat to consider which captures a part of context that might be lost elsewhere.
That's the problem with your post, Sandy...you didn't look at the statistics that COMBINE all of the things on your list.
Felix is better than Lee because he pitched more at nearly the same rate
Felix is better than Bucholz for the same reason plus a better DIPS profile.
Felix is better than Sabathia or Liriano because he had a significantly better DIPS profile.
Felix is better than everyone in baseball...because he's doing all of this while under the constant threat of losing generated by being on a horrid team and he's eating HUGE numbers of innings without showing any signs of slowing down.
Liriano leads the league in tERA, FIP, and xFIP. Liriano has been the best pitcher in baseball, but Felix and Lee are very close.
Liriano: 172.1 IP
Felix: 225.2 IP
It's not even KINDA close.
That was my point. Sandy was focused exclusively on the rate statistics and not at all on the bulk statistics. It is actually significantly easier to have a low xFIP with 170 IP than it is to do the same with 225 IP...the more you pitch...the harder it is to be at max effort on each pitch. That's why relievers have a higher K rate, better K/BB and lower BABIp than starters.
Ya, but his periphs are a LOT better than Felix. He leads in WAR even with the IPs gap. Its not a huge gap. Liriano has pitched less innings, but been much more dominant non-ERA wise.
Cliff Lee has actually been better too and is in between the two IPs wise. If he finishes strong I could see an argument for him too.
I think its between those three (at least for me) and the last few starts could end up deciding it. I can't really see an argument for anyone else since those three have been so far ahead of the pack this season.
Because Felix led in Sean's WAR, Tango's WAR, pRC, WS and B-Ref's WAR four days ago when last I looked.
Liriano's peripherals aren't THAT much better than Felix' when you include comparative in-play results. DNRA for the top pitchers:
Francisco Liriano: 2.81
Felix Hernandez: 2.95
Cliff Lee: 3.22
Everyone Else: 3.86 or higher
Meaning using my own pitching metric (DNRA...with which the long-timers here would be familiar...I can re-introduce it if there is interest), Felix is far...far and away the best pitcher in the AL by value.
DNRA would be another important metric in the equation (the gap is a bit larger in tERA and FIP), but I don't see it as Felix being far and away ahead. Felix's ERA is a bit fortunate this year considering unearned runs, park, and defense.
Its really going to be determined in the last few starts. These guys are neck and neck.
...DNRA treats errors as expected outcomes of BIP too...Felix had more errors behind him than normal...I don't think that makes his ERA fortunate. I think it means the Mariners make too many mistakes whe he pitches. When ANYONE pitches, for that matter. They have a good defense, but they do make a lot of errors for being so defensively talented.
And of course, DNRA factors out the park and defense.
Ya, but the overall metrics (including DNRA) don't differentiate Liriano, Felix, and Lee much value-wise. Bref, statcorner, fangraphs, all have them within a couple runs of each other.
The race for "best pitcher in '10" is really tight this year. The race for "Cy Young" I don't really know.
..that Felix was worth about 16 runs more than Liriano (if you simply take the DNRA, invert it to offensive scale using Pythagoras assuming the average R/G league context of 4.64, and figure out how many runs above average Felix is by multiplying by IP...and do the same for Liriano).
That's not just a few runs...that's many.
Felix has a case definetly (DNRA would have him ahead by a good amount, BRef by a little, the others Liriano by a bit). I just think its very close right now and could go either way between these 3 pitchers in the next few weeks.
One thing is for sure. CC Sabathia doesn't deserve it IMO.
I'm not entirely sure what the Cy Young is measuring anyhow. The best pitcher on the best team? Or the best pitcher in baseball?