Was Kershaw Really Better Than Halladay?
Clayton Kershaw received 27 of the 32 first place votes in the 2011 NL Cy Young Award balloting en route to his first major league award. He beat out all three of the Phillies horses — Roy Halladay finished second, with Cliff Lee third and Cole Hamels fifth — and Ian Kennedy to officially go down in the book as the best the National League had to offer this past season. But that isn’t necessarily true, and his case isn’t as shut and dry as all those first place votes make it seem.
It’s plausible to suggest that Roy Halladay and Cliff Lee were, at worst, Kershaw’s equals, and more likely than that his superiors. His winning the award is in no way a mockery, like it was when Bartolo Colon beat Johan in 2005, but it does feel like somewhat of a step back in the voting process. After Felix Hernandez won his award it sure seemed the voting body grasped that team context is important when evaluating players.
His wins total wasn’t up to par with traditional candidates, but voters understood that the Mariners offense was out of Felix’s control. He wasn’t penalized for perceived poor performance in a common performance indicator.
In the case of Kershaw v. Halladay v. Lee, a similar understanding wasn’t extended to strength of schedule and park effects. Kershaw deserves hearty congratulations, but his Cy Young Award win again illustrates the importance and utility of normalizing numbers. When adjusted numbers enter the fray, Lee emerges as a very viable candidate. Halladay steps forward as the best pitcher in the league, and by a long-shot. In our staff awards ballot, I did vote for Kershaw, but my stance has since changed. No matter how one chooses to slice it — unless they like slicing “it” incorrectly — Halladay was the best pitcher in the senior circuit this past season.
Kershaw pitched tremendously in 2011, but we’re at the point where it should go without saying that underlying factors influencing the numbers provide the appropriate context to properly analyze the numbers. Kershaw’s numbers were shinier from a raw perspective, and his pitching triple crown — he led the league in wins (21), ERA (2.28) and strikeouts (248) — put his candidacy over the top. Add in his gaudy innings total — 233.1, one out behind Halladay and two outs ahead of Lee — and it’s very easy to see why he won.
But everyone needs to look deeper than that. Kershaw led with a 2.28 ERA, but Halladay wasn’t far behind at 2.35, and Halladay didn’t have the benefit of facing the Giants and Padres a combined nine times. Lee finished the season with a 2.40 ERA, barely behind Halladay, and the same disclaimer applies. Facing two of the worst offenses for almost 25 percent of his starts has a material impact on Kershaw’s seasonal numbers.
Not to mention that all three of the Dodgers, Giants and Padres stadiums greatly favor the pitcher. Kershaw made 19 of his 32 starts in those three stadiums. Pitching in venues where run-scoring is tougher, and facing teams that frankly stink at scoring runs to begin with, goes a long way towards explaining Kershaw’s success this season.
Kershaw finished the season with a .269 BABIP, with Doc and Lee posting respective rates of .298 and .291. Dig a little deeper and we discover that Kershaw’s BABIP split was .249 at home and .288 on the road. For his career, he has a .271 home BABIP and a .287 mark on the road. While his skill-set may lend itself to weaker contact, it’s simply unknown right now if his hit prevention is more contingent on his repertoire or his home park. Believing that Kershaw was a true 2.28 ERA pitcher and the best pitcher in the league is to believe that he was fully responsible for the low BABIP.
In the end, Kershaw fell behind both Halladay and Lee in more telling areas. Halladay finished with 8.2 WAR, a full win and a half more than Kershaw. Both Doc and Lee posted lower xFIPs and SIERAs and higher K/BB ratios. And, perhaps most importantly, the luck-based metrics suggest that Doc and Lee were well-represented by their numbers. It’s understandable why voters would opt for Kershaw over Halladay and Lee, but the end result didn’t do either member of the latter pair justice.
Halladay was the best pitcher in the National League for the second-straight season, and Lee was right on par with Kershaw. It’s unfortunate that one of those Phillies pitchers lost because of unadjusted numbers and them “splitting” votes with one another.
He pitched very well, and nobody should forget that. When discussing who should win this award we’re arguing over who the top two or three were in the league. However, it amounts to more than splitting hairs because raw numbers dominated the vote. Kershaw pitched very well, but wasn’t as dominant relative to the league as 27 of 32 first place votes makes it seem.
0
Ha ha. After reading your article, Eric, I wondered whether you were from Philadelphia. I googled you and sure enough, I found that you were born in Philadelphia and suffer from a chronic case of Philadelphia syndrome, a common disease that causes you to think that real baseball teams and real baseball players can only be found in New York, Boston, and Philly. It must have been hard to get all that press all year and then end up with nothing.
So YOU were the one googling “Eric Seidman and assclown”?
The breakdowns are interesting, but this article does come off as one written by a Phillies apologist. Kershaw faced SD and SF 9 times? And, Halladay feasted on WAS & FLA seven times. So what?
3 easy reasons Kershaw won the Cy:
1. Kershaw had the Pitching Triple Crown. Frankly, I think it’s “progress” that 5 voters did NOT give him their first-place votes given this fact.
2. Roy won last year… unlike Gold Gloves, where previous wins tend to help build a reputation for the next one, previous Cy/MVP wins tend to have voters looking for other options unless a repeat winner clearly earned it.
3. Phils’ “greatest rotation ever” split the vote. (Of course, it doesn’t really matter when 1 guy gets 27 first-place votes.)
Also, although I obviously don’t put much stock into wins, it nevertheless is impressive to have a 20-game winner on such a shatty BB team. Meanwhile, the Phils were so loaded, they could’ve inserted anyone into the rotation (Vance Worley) and they’d be on a 20-win pace for a full season.
and the Mets….. don’t forget the Mets.
Rob, you’re not getting the concepts discussed in the article. Forget where the guy that wrote the article is from and address what is wrong with the concepts if you really disagree with it.
Rob – I couldn’t disagree with you more. With regard to the “Pitching Triple Crown,” let’s not get terribly excited. It has been done 6 times in 10 years. Moreover, we might want to think about what this Triple Crown really is: a measure of statistics over which the pitcher has limited control. This is why we have metrics in the first place, and your comments suggests to me that you don’t read or absorb the articles on this site. I would also point out that while the Phillies beat up on Florida and Washington some, their offenses and ballparks are significantly more pro-offense.
I just shrug at stuff like this – why come to this site if you are going to ignore the very metrics that make it tick?
Rob, the “so what” is that the Giants and Padres had the worst offenses in baseball. While the Nats and Marlins weren’t good teams, they were not last in offense.
Err…you do know the Mets were 6th in runs scored in the NL last year, right? Ahead of the Phillies?
shouldn’t we normalize for streakiness? If you got Dan Uggla out the first half of the season, that’s not a big deal, if you go thim out in the second half, it’s a bigger deal.
You can’t just say “well let’s normalize performance” and then assume facing a team in April is the same thing as facing them in July. Baseball is such a streaky sport.
In Doc we trust. One of the best of this generation (pitched through the steroid era) and a class act.
p.s. not sure about the new password comment portal.
Not only one of the best in his generation, but the best. And it’s not all that close either. I’d count Pedro as before his time, really.
And yeah, the password system is proving difficult.
It’s lame alright. Just about as lame as the first comment here. Nothing biased about this. It’s what we were all thinking. Cut and dry over shut and dry, right ?
He pitched a couple excellent seasons at the end of the steroid era. He did not pitch through it. To me the end of the steroid era was 2007. Offense had been declining for a couple years but that was the year the extra juice finally started getting out of the players system. Offenses peaked around the turn of the century and declined a bit after that. Doc became Doc in 2002.
Please, people, it’s neither “shut and dry” nor “cut and dry”; it’s “cut and dried”.
Halladay was clearly the best pitcher in the NL in 2011. I’m not sure why that’s so difficult for some to admit when the evidence is staring them right in the face.
If Doc had broken up the Triple Crown in some way, I also think there’s little doubt he would have won the award. Writers still are rather fixated on those three traditional and largely meaningless stats.
Strikeouts are definitely not meaningless
Without context they essentially are. In this case, you have to ask home many IP and how many walks. When you find that Halladay was a lot better with walks, the raw strikeout total means less.
Kershaw allowed a hell of a lot less hits than Halladay. I know sabermetrically inclined people will say after the ball is thrown it’s not the pitcher who decides what happens, but when someone has done it their entire career, I don’t see how it can be discounted. A big reason why Kershaw was behind in WAR and FIP (as well as xFIP) is because hits allowed do not factor into those. Kershaw ended up with a lower WHIP than Halladay, despite Halladay walking like 1.1 per 9 innings or so.
Say what you will, hits allowed is meaningful.
No one is suggesting that hits be absolutely discounted (nor has anyone suggested it’s a travesty the Kershaw won) merely that they be taken in context: Pitchers don’t have zero control over BABIP (i.e. hits allowed) but BABIP is also influenced heavily by the defence behind the pitcher, the teams he’s facing, and the parks he’s pitching in.
When assessing the skills of pitchers, it’s probably incorrect to weight hits allowed as heavily as strikeouts, walks and home runs.
The run values for each type of hit, as well as a walk, are known, and the difference in walks between the two makes up more of that difference than you are implying.
Even if you look at a stat like ERA-, which still gives credit to Kershaw for his BABIP (the reason he allowed less hits than Doc), Halladay edges him by 1 point, likely due to park factors.
The NL East competition didn’t feature any great offense itself, but it also doesn’t feature parks as pitcher friendly as Petco and AT&T, and only one park lower ranking than Dodger Stadium, which is Citi Field, not Citizen’s Bank.
To say that one was clearly better, seems extreme. I think you could flip a coin for lots of different reasons. Both are quite different.
“While his skill-set may lend itself to weaker contact, it’s simply unknown right now if his hit prevention is more contingent on his repertoire or his home park.”
That’s a huge unknown for Kershaw. BABIP is a major influence within this argument, and to state that we know for sure what’s happening is kind of jumping the gun.
Even though the home/road BABIP split is most obvious this year, the previous 2 seasons, both sides were still quite below the norm.
When it comes to BB/K ratios also, if you’re walking as few a hitters as Halladay and Lee do, I think the value of that ratio can become inflated. I think there’s a greater difference in effect between a 9% and 6% walk rate as opposed to a 5% and 2% walk rate. Walking only 20 batters in a season is impressive, just as striking out an enormous amount of hitters is impressive, but are those extremes really that influential? I think extremely low walk rates might be a bit overrated just as extremely high K rates are.
Well, strikeouts are definitely not meaningless. It is the fallacy of the pitching triple crown: Wins, ERA, WHIP. The true triple crown is K/9, BB/9 and HR/9, the things pitchers exercise more control over during the course of a season.
Great article, this is exactly what struck me when I heard Kershaw received 27 of 32 votes.
I think Colon won the AL Cy Young in 2005, not 2006, though.
You’re right. Fixed. I was so busy thinking about whether our linker would link to Johan Santa instead of Santana that I lost track of the year.
Nice read Eric. Having lots of trouble with the password stuff. Anyway, would you rather measure pitchers on K and BB and GB/FB rates and park factors and strength of schedule? Where would Matt Garza have fallen in the voting had we used these? I’m just not sold on it yet. Do you think MLB will EVER change to the raw data measures FG and others discuss so much? It’s all great stuff and it helps me in my fantasy league. But give me ERA and wins any day.
Signed,
Old and ERA/Wins dependent
The password stuff isn’t difficult. IT SUCKS!!!
Ps. Halladay or Lee winning it would not have been bad either no matter where you’re from. Unless it was Garza. Then I would have liked it more. Ha!!!
How hard is it to copy/paste it?? It is definately easier than the captcha when half of the time they are written so that you have no shot of differentiating similar looking letters [is that a capital I or a lower case l??
Copy/Paste? Damn I am old. But I think I just learned a new trick. Thanks. Shit! What happened to the password stuff.
Yes, he was better. Markedly better? No Better? Yes. Take any pitcher you want to get one out and I’ll take Kershaw.
Halladay didn’t have the benefit of facing the Giants and Padres a combined nine times.
No, just the mets, nationals and marlins…
Kershaw was just better, accept it.
Wow, quite the rebuttal there.
2011 Runs Per Game
Mets 6/16 at 4.43
Nats 11/16 at 3.88
Marlins 12/16 at 3.86
Padres 15/16 at 3.66
Giants 16/16 at 3.52
OK, now which team had a better defense. Philly or LA?
Per UZR, Dodgers were +3, Phillies were -15.
That’s not fair, the Phillies knew what they had with the decrepit Raul Ibanez out there for them. Lol.
And the Dodgers clearly had better outfield defense what with Andre “The Glove” Ethier and Matt “Death to Flying Things ” Kemp patrolling the field.
You still need to look at every single one of their starts, both in terms of park and opponent before you can reach a definitive conclusion you twat.
The “full division” stats aren’t instructive if you ignore the specific mixture that relates to our pitchers in question. NL West teams score a lot more runs when they play their games at Chase Field and Coors Field. How many of Kershaw’s starts were in those environments?
Kershaw made 5 starts in SD/SF, 4 starts in COL/ARZ
Padres 15/16 at 3.66
Giants 16/16 at 3.52
=================
uhhh maybe part of the reason that the Giants/Pads had such a low run total was because they faced Kershaw so many times?
“Kershaw made 5 starts in SD/SF, 4 starts in COL/ARZ”
And 6 more starts against SD/SF in Chavez Ravine, where they are equally as challenged.
cs3, they faced him a total of 9 times in a total of 324 games. I don’t think he’s responsible for their low run totals.
Mets had a wRC+ of 103
They were very good offensively.
Also, how about you be honest with the numbers?
NL West, other than LA – 2629 runs, 4.05 runs/game
NL East, other than PHI – 2608 runs, 4.02 runs/game
Kershaw – 4 vs ARI, 3 vs COL, 3 vs SD, 6 vs SF
Halladay – 2 vs ATL, 3 vs FLA, 3 vs NYM, 4 vs WAS
So, where did that advantage go?
Well, nobody quoted full division #s? I clearly posted the RPGs of the teams in question above. You’re entitled to your opinion on the matter, but I fail to see any evidence in your line of inquiry that actually supports your notion.
Well, maybe you would have if they supported your argument….
Kershaw (2.03 ERA vs west) performed better than Halladay (2.40 ERA vs East) against their respective divisions, and kershaws division is the higher run scoring environment.
How is that NOT incredibly relevant?
In fact, Halladay a MUCH easier go of it last year than Kershaw did…
Kershaw made 18/33 starts against teams above .500
Halladay made 10/32 starts against teams above .500
Also… this password thing suuuuuuuucks.
Tom,
The record of the opponent is not particularly relevant in that SF counts as one of the teams above .500, yet they were the second worst scoring team in MLB.
Inrerms of the more difficult starts,
Kershaw made 15 Unique Start (opponent and home/road) and Halladay made 14. Obviously Kershaw pitches in more of a pitchers park but Home/Road seemed like a “decent” measure. All their other starts overlap. Just an example to what i am doing Doc started against @ Col once and Kershaw started @ Col twice so I will list @ Col once for Kershaw. The unique starts are as follows.
Kershaw:
@Ari (X2), @ Col, @ LAA, @ SDP, @ SFG (X3), Ari, Col, Det, LAA, SFG (X3)
Doc:
‘@ CHC, @ FLA, @LAD, @ PIT, @ TOR, @ WSN (X3), MIL, NYM, OAK, PIT, TEX, WSN
So Doc had Pittsburgh twice and Oakland but also had to face MIL, and TEX while Kershaw had Arizona and Col but 7 additional starts against the giants and padres.
Not sure the exat balance but eyeball test leans to more difficult for Halladay but I can’t run the numbers right now so someone lese can if they want or I’ll come back later and do it.
Really Eric? You don’t see anything about the full division runs per game that seems important? Because those numbers that Tom B posted seem to be contrary to the point your tried to make for most of this article. You can believe in your heart that Halladay was better, but your argument about the opponents faced was definitely lacking….
All that being said, I think Kershaw was a bit better but it was really close, and I do think Lee and Halladay split their votes.
The “full division” stats aren’t instructive if you ignore the specific mixture that relates to our pitchers in question. NL West teams score a lot more runs when they play their games at Chase Field and Coors Field. How many of Kershaw’s starts were in those environments?
(Sorry for re-posting, I replied to the wrong comment thread).
2 in Chase and 2 in Coors I believe?
all east coast teams are good… how could one forget?
Fun fact: Halladay beat Kershaw in literally every DIPS ERA estimator:
FIP
xFIP
SIERA
tERA
tRAr
ERA-
Other fun facts:
the NL West offenses are a complete joke.
CBP is a bandbox, and Dodger stadium is death valley.
I think you should check the park factors before you use the word fact.
I think you should:
http://espn.go.com/mlb/stats/parkfactor/_/year/2009
http://espn.go.com/mlb/stats/parkfactor/_/year/2010
http://espn.go.com/mlb/stats/parkfactor/_/year/2011
Bandbox is a bit of an exaggeration, but there is a huge difference between the two parks.
Agreed there is a difference. I should have been more specific actually. I hate the bandbox mentality people have about CBP. It really plays about neutral, and the bandbox thing gets blown out of proportion. That’s all. Actually looking back on my last comment, I just sound like a dick to be honest.
Lol it’s cool. I learned something.
You’re right, CBP is not the bandbox that I thought it was. Now, I’m not really sure how it got that stigma.
It is slightly pro-offense though.
It has to do with the home runs. I think CBP the last few years has been HR-friendly, but overall runs-neutral.
Coors and Chase Fields are in the NL West. One could argue just as persuasively that Kershaw is at such a disadvantage each year from having to pitch in what are far and away the NL’s two best hitter’s parks. That line of reasoning is a dead end.
Citizen’s Ball Park actually slightly favored pitchers in 2011. It’s basically a neutral park and always has played as such. Safeco Field actually was more homer friendly than CBP in 2011.
CBP is not a bandbox by any stretch of the imagination.
Eric, you are exactly correct. CBP plays very high in HR/AB, but runs neutral.
Fun fact: Jeff Niemann beat Jeremy Hellickson in all of those categories except ERA- this season. Who had the better year?
“except ERA-”
And boom goes the dynamite.
Jeff Niemann.
I don’t know if the quality of opposition argument holds much water here. When they were not facing Roy Halladay, Halladay’s opponents slashed .263/.330/.413. For Kershaw, the numbers are .263/.327/.416. (numbers from http://www.baseballprospectus.com/sortable/index.php?cid=1058305)
CBP inflates wOBA by 1-3%. Is that really a bandbox? It actually doesn’t hold a candle to Coors Field or even Chase Field, both of which are considerably better hitting environments than anything you’ll find in the NL East.
It would be nice to have those numbers park-adjusted.
Am I the only one who is surprised that voters who are (stupidly) averse to voting for dudes on mediocre/irrelevant teams just chose Matt Kemp and Clayton Kershaw, two players who not only played for a mediocre team but the SAME mediocre/irrelevant team, as their MVP and Cy Young Award winners?
If Matt Kemp and Clayton Kershaw played for a mediocre/irrelevant team that wasn’t a) a team in a huge market and b) a team with a very storied history, would they have gotten enough votes to win?
Further, are these same voters going to refuse to vote for Jose Bautista simply because his team was mediocre/irrelevant? What does that tell you?
NL MVP is next week, but I get your point.
Oh, I guess you’re right. He’s presumably going to win it, though, and when he wins it and Jose Bautista doesn’t, then this is inconsistent and just stupid.
Right, you mean inconsistent and stupid because Kemp led the NL in WAR and Bautista finished 1.1 wins behind Ellsbury. Gotcha.
Yeah, because all the BBWAA voters look at fWAR.
I realize they didn’t set the world on fire, but the Dodgers ended up over .500, so why exactly does everyone think they were actually shitty overall? Yes they were carried by a few players most of the year, but going by actual record, they did not end up bad. By the way people talk about the 2011 Dodgers they may as well have won 73 games.
Anyone of the threesome of Kershaw, Lee and Halladay work for me.
Guys, this password thing is FUCKING BROKEN. It fails half the time for no reason (I am copying and pasting so I can’t be entering it wrong).
The difference in competition is being greatly exaggerated. Sure Kershaw faced SD and SF 9 combined times. But Halladay faced the PIT 2 times, WSN 4 times, and OAK 1 time. Kershaw never faced these teams. The weighted average wRC+ of opposition for those 7 games is 88 (WORSE THAN THE PADRES). So quit your crying.
This argument is really one-sided. You mention that the Pads and Giants parks favor the pitcher. Though the Giants park is fairly neutral, I’ll assume that is true. Why no mention that the Mets, Marlins, Braves and Nats parks also favor the pitcher? Why no mention that the Dbacks and Rockies parks are the most hitter friendly in baseball?
You mention that the the Pads and Giants had horrible offenses and that’s true. Why no mention that the Braves, Marlins and Nats also had bad offenses? Take out these pitchers’ teams and the NL West far outscores the NL East (by 11%).
Park factors and strength of schedule both favor Kershaw.
FWIW, Kershaw pitched 67 2/3 combined innings vs. SD and SF and 44 combined innings vs. ARI and COL.
All pitchers face good teams and all pitchers face bad teams. What is the difference in expected runs given the differing competition? Probably not enough to bash the writers for choosing Kershaw
“Park factors and strength of schedule both favor Kershaw.”
[Citation needed]
Halladay had the better park-adjusted ERA and therefore a better Rally WAR according to the adjustments made by B-R Ref.
There is a reason we are subjective in this world, and there is a reason baseball is not all about the stats. It’s called enjoyment of being different.
I agree doc was better, and the park (and other normalization) factors are very important to look at.
but should the predictive stats (fip, xfip, etc) really count all that much? Yes it’s great to know what was luck based, and they tell us what to expect in the future, and what may be sustainable.
But the award is issued for what happend, not how it happened. [again, this is more in a general sense than for this specific case, since park & competition factors make doc's year look even better].
If a team lucked into winning a game it will count as a win. so if a guy lucks into great numbers, shouldn’t he win the award? I mena, i dont even know how exactly i feel about this issue, guess just sparking a discussion.
TEBOW!
xFIP is predictive. None of the other stats mentioned are.
I agree doc was better, and the park (and other normalization) factors are very important to look at.
but shuld the predictive stats (fip, xfip, etc) really count all that much? Yes it’s great to know what was luck based, and they tell us what to expect in the future, and what may be sustainable.
But the award is issued for what happend, not how it happened. [again, this is more in a general sense than for this specific case, since park & competition factors make doc's year look even better].
If a team lucked into winning a game it will count as a win. so if a guy lucks into great numbers, shouldn’t he win the award? I mena, i dont even know how exactly i feel about this issue, guess just sparking a discussion.
So sick of this argument. Every freaking day on this site.
FIP is shown to be predictive, but it is inherently not just a predictive stat (as xFIP is). FIP still measures exclusively things that actually happened on the baseball diamond. It just eliminates factors that depend on fielders.
FIP is not a perfect measurement of what a pitcher is. Everyone agrees some pitchers have some (difficult to measure) control over BABIP. While we don’t have a way to measure this right now, the one thing we do know from other peripherals is that it is really quite rare that a pitcher shows a statistical profile that indicates the possibility of BABIP suppression as a skill; hence, FIP is about the best measure we have right now, despite the fact we all know and agree it isn’t perfect.
“we do know from other peripherals is that it is really quite rare that a pitcher shows a statistical profile that indicates the possibility of BABIP suppression as a skill”
It’s also really quite rare that a 23 yr old puts up the kind of numbers Kershaw has. He’s an extremely rare kind of dude. His BABIP has been fairly steady over his 3 full years in the Majors, taking a small step forward this year (as did a few of his stats that we know he does control, like walk rate), so why are we assuming he does NOT have skill at suppressing BABIP? I guess I’m just not getting it.
@ Dave
Because BABIP takes 8 years to stabilize
http://www.baseballprospectus.com/article.php?articleid=14293
That makes it seem to me that normalizing BABIP in context of a single year award is completely irrelevant. Seems like we need to take what ACTUALLY HAPPENED somewhat at face value.
Post from phone. No password issues
I like the comment someone made yesterday after the article that said how great of a season Kershaw had. You can look at advanced stats and use them to say what should have happened or what might happen if people played the same next year. However, awards are given out based on what actually happened, not what some numbers say should have happened.
What did happen was that Kershaw had a 2.28 ERA and Halladay had a 2.35 ERA, yet Kershaw pitched ~50% of his innigs in a better home park for pitchers.
What did happen was Halladay did not win the Cy Young.
It’s amazing how many people are talking about the advanced metrics that Halladay was better than Kershaw in, and not mention how heavily influenced those metrics are by Halladay’s miniscule HR rate which was the lowest of his career and nearly 5% lower than his career average over a very large sample size. That, my friends, is the very definition of an outlier, and a very big reason why Halladay has better metric stats than Kershaw.
Also, I think the biggest deal that no one is mention here is that somehow Ian Kennedy got a first place vote!
dte, if the HR/FB rate bothers you, feel free to look at xFIP, where Halladay still bested Kershaw.
CBPs HR park factor is higher than Dodger Stadium’s. Halladay also has a GB% that is 7% points higher than Kershaw’s. Look at the batted ball profiles for each player. If you put Kershaw’s batted ball profile in CBP, a conservative estimate on the impact of the HR park factor alone puts Kershaw’s ERA over Halladay’s… again, not that ERA is the right measure, but that is what this argument seems to be revolving around. This is why Hallday’s ERA- is a point better than Kershaw’s, because ERA- adjusts for park.
Do we know who voted for Ian Kennedy? If you’re wanting to be a smart aleck, why not just vote for, I don’t know, Joe Saunders? I’d love to see/hear the Kennedy voter’s argument as I love things that are unintentionally hilarious.
First, I would have voted Kershaw. Seriously, this could have gone the other way and I wouldn’t have been a bit upset. The only thing that surprised me was that Kershaw received 27 of 32 first-place votes. Still, this is FAR from scandalous.
I am a Cards fan so I don’t have a pony in this race, but I can’t imagine anyone saying Doc was CLEARLY better when he was MAYBE A WEE LITTLE BIT BETTER BUT IT IS VERY CLOSE AND LET’S BE HONEST, KERSHAW ISN’T A BAD CHOICE. In fact, I think he’s the right choice.
Yes another Halladay vs Kershaw post! Boomslang!
I certainly can hear the argument for Halladay as the better choice, and I think he did have a SLIGHTLY better year. The tone of the article is a bit much though, with lines such as “No matter how one chooses to slice it — unless they like slicing “it” incorrectly — Halladay was the best pitcher in the senior circuit this past season.” Don’t you think that’s a bit much?
And Halladay does clearly come out ahead in the fangraphs WAR system, but most others I look at it’s pretty damn close (and Lee usually a step below Halladay and Kershaw).
Am I happy Kershaw won? Yes (I’m a Dodgers fan after all).
And I think that Halladay was SLIGHTLY better. But the tone of arguments using ONE primary measure/uber-stat and concluding based on that seems… pretty damn silly.
I think the most striking issue is that the vote was such a landslide.
You’d think there would be a few more BBWAA members who pay heed to contextual analysis. In the end, it is what it is. Not the end of the world.
I think it was such a landslide because Kershaw was the new blood, much like Votto to Pujols last season when it could have easily been close to a 50/50 race.
Hmmm…attempt 1 to post = fail
What will also be interesting is how many MVP votes each gets.
I predict that Halladay will get more, which is nonsensical if the voters have determined that Kershaw is the better pitcher, since Kershaw definitely provides more offense with his bat.
That worked. What is the ‘password’ issue?
I wouldn’t consider it nonsensical because: a) Standings don’t matter in the CY Young race, but many voters consider them in the MVP race, and b) Different people vote for different awards.
A lot of talk about opponent quality without much data.
What was the wRC+ of the batters who faced Kershaw? Halladay? Any other pitcher? What did each pitcher hold the opponents to?
Even better, compare expected batted ball speeds, K rates, BB rates from hitters to the actual results.
This discussion is a great example of why the next wave of stats need to include opponent quality.
The voters strongly consider results, and Kershaw had better results this year. Performance and results don’t always match. Halladay maybe was better this year, but it was close enough that Kershaw winning with better results isn’t a travesty.
A lot of talk about opponent quality without much data.
What was the wRC+ of the batters who faced Kershaw? Halladay? Any other pitcher? What did each pitcher hold the opponents to?
Even better, compare expected batted ball speeds, K rates, BB rates from hitters to the actual results.
This discussion is a great example of why the next wave of stats need to include opponent quality in them.
The voters strongly consider results, and Kershaw had better results this year. Performance and results don’t always match. Halladay maybe was better this year, but it was close enough that Kershaw winning with better results isn’t a travesty.
Like others have said, the strength of schedule is being overbolwn
THere were 6 NL teams that scored 625runs or less. Halladay faced them 13 times. Kershaw faced them 13 times
There were 5 NL teams that scored 720 runs or more. Halladay faced them only 8 times, Kershaw 12 times
Kershaw won the triple crown. I could see discrediting the wins if he played on a 100+ win team (like Halladay did), but he won all those games for a blah team. He also led the league in WHIP . So he essentially won the quadruple crown
IP is more important as a 4th stat than WHIP.
Let’s call IP a wash
Actually, HR/9, K/9 and BB/9 are all more important than any of those 4 stats.
Do you guys get what Eric did here? This is a metrics site, like it or not. You get a metrics analysis. And, to be honest, no matter where you come from, Halladay wins the metrics argument. Does it mean he should win the Cy? No…Kershaw had a great year with great results. It does strike me, however, that Halladay probably should have gotten a few more 1st place votes.
HR/9, K/9 and BB/9 are good stats, but rate stats are meaningless without knowing the duration of the rate.
And Kershaw faced a lot more of them at Chavez Ravine, whereas Halladay did it in CBP. That’s why Halladay saved more runs above replacement than Kershaw this year, either by ERA method (Rally WAR) or by FIP method (fangraphs WAR).
Since there is so much talk about value and money around this site I’m surprised no one mentioned that Halladay made $20 million for his work last year and Kershaw only made $500K?? Talk about value!!!
The CY isn’t about “value”, it is about the “best” pitcher, period.
from Olney today
Kershaw, Roy Halladay and Cliff Lee all had remarkably similar stats and Kennedy tied Kershaw for the league lead in wins. But against teams .500 or better, Kershaw was 12-3 with an ERA under 2, a WHIP under 1, and a strikeout-to-walk ratio over 5.
A team’s record doesn’t always correlate to the offense. The Giants were above .500 but had a horrible offense.
TRue but he did have to outduel LIncecum in many of those starts. I Know that does not change the offensive environments either one faced (Halladay or Kershaw) yet impressive nonetheless.
Kershaw is also a lefty while Halladay is a righty…since we’re analyzing everything.
Their handedness has to have some impact right?
Yes Kershaw is at a distinct disadvantage because he got to face fewer batters of his same handedness. That’s outside a pitchers control; no. So while it may not add to the real life value of either pitcher, this could suggest that Kershaw had a more difficult road to hoe than Halladay. LOL, Now my head is really spinning.
I don’t think it really matters what division you are in, it matters how good of batters you face. I mean, even the best teams have some bad hitters so you can get lucky and face them one a day that their best hitters have the day off. So looking at the teams they face is pretty close to irrelevant.
That being said, here is the slash lines for the actual batters that Kershaw, Halladay, and Lee faced:
Kershaw: .263/.327/.416
Halladay: .263/.330/.413
Lee: .266/.332/.423
I got these stats off of BP. Indecently, their WARP has Kershaw ahead of Halladay 7.1 to 5.2. Lee is at 6.2.
Eric, you;re pretty familiar with BP’s stuff…why is their version of WAR come up with such a different ranking then Fangraphs’?
I believe this is because they use runs against, while FanGraphs uses FIP.
Clayton Kershaw benefited from facing lesser competition?
News to me.
http://www.baseballprospectus.com/article.php?articleid=15019
why is there so many articles on this site about how halladay is better than kershaw, instead of articles about how ridiculously good kershaw is at such a young age?
but again, this is the same site that compared roy halladay to sandy koufax…
and as someone stated in here, it’s also one minded to only use your site stats (fWAR, FIP, xFIP) and not use others like WARP, bWAR, OPS against, WPA, win shares, which are all categories Kershaw leads in.
Kershaw only leads in bWAR if you include the hitting portion, which seems largely irrelevant for an award based on pitching. In pitching bWAR, Doc posted 7.4 to Kershaw’s 7.0.
How much that adds to the equation is anyone’s guess, however.
In addition to Phrozen’s great reply, I would also like to add that it is obvious that FanGraphs chooses to use their method of calculating WAR because it is their method…
They didn’t choose to use FIP just to be different or for the hell of it, they chose it because they felt it was a better way to measure the isolated performance of the pitcher himself without all the noise related to the fielding (in)ability behind them. FG have posted articles explaining this in the past, you are encouraged to search for them and read them.
If you disagree with the choice they made to use FIP, that’s a totally fair and intellectually honest point. Just realize that they aren’t ignoring brWAR in an intellectually dishonest way — they are choosing to use fWAR in their evaluation for all the reasons that they chose to calculate pitching fWAR they way they did.
I think Lincecum lost like three 1-0 games to Kershaw last year.
I mean there’s bad offenses, then there’s historically bad offenses, then there’s a dead ball era, which is where the 2011 Giants reside firmly.
Someone was asking about wRC+ in terms of strength of competition. I took a quick look and took the average seasonal wRC+ for each batter that Kershaw & Halladay faced. (This probably isn’t the best way to do this, but it gives you some idea):
Halladay – 93
Kershaw – 91
So, seems like Halladay on average faced slightly better batters based on the 2011 stats.
Interesting.
I posted something similar on the NL CY Young thread yesterday.
“Total number of hitters faced
Halladay: 961
Lee: 936
Kershaw: 946
Average wRC+ of hitters faced
Halladay: 93.73
Lee: 93.46
Kershaw: 91.66
For comparison sake, here is the AL equivalent.
Total batters faced
Verlander: 997
Sabathia: 1007
Weaver: 959
Average wRC+ of batters
Verlander: 96.69
Sabathia: 99.77
Weaver: 96.96″
Seems my average wRC+ is a little off your data, so I’ll have to do some double checking.
Why do you think average wRC+ isn’t the “best way to do this”?
Average wRC+ of hitters faced
Halladay: 93.73
Lee: 93.46
Kershaw: 91.66
This establishes quality of opponent, but what was the wRC+ that hitters posted against each of these pitchers?
I’m not sure that I’d be able to calculate the wRC+ that hitters posted against each pitcher, but I also don’t think it would tell us anything that ERA- and FIP- doesn’t already tell us.
ERA- and FIP- are the pitcher equivalents of wRC+.
http://www.fangraphs.com/library/index.php/pitching/era-fip-xfip/
Halladay’s FIP- was 56 and ERA- was 61, meaning he was 44% and 39% better than league average respectively.
Kershaw’s FIP- was 66 and ERA- was 62, meaning he was 34% and 38% better than league average respectively.
The NL league average wRC+ was 94. (I thought this should be set to 100. Is there a particular base year that is set to 100?)
Halladay’s average wRC+ batter faced of 93.7 means that his ERA- and FIP- don’t need adjustment.
Kershaw’s average wRC+ batter faced of 91.7 means that his ERA- and FIP- give him a little more credit than he is due. I’m not sure precisely how much adjustment is needed. Given the average batter he faced was around 2% worse than average, it might make sense to increase his ERA- and FIP- by 2%.
Halladay was better by these measures before any adjustment and Kershaw’s easier schedule means that the gap should really be a little bigger than the unadjusted ERA- and FIP- suggest.
Forgive my ignorance, but is wRC+ park adjusted?
Yes
http://www.fangraphs.com/library/index.php/offense/wrc/
I have wOBA if you prefer though.
And Halladay did it in an overall tougher run-scoring environment.
Phils, read the Baseball Prospectus article linked by Chad Moriyama above. It takes park effects and strength of schedule into account. Kershaw comes out slightly ahead.
Does that adjust for “splits” handedness of the hitters?
Halladay was the best pitcher this year, though Kershaw was awesome and had some classic things in his favor, i.e. high strikeouts and big success in head-to-head matchups with Lincecum. I understand perfectly well that Kershaw beat the crappy SF hitters, not Lincecum, but those kinds of things trigger the nostalgic edge in voters.
I wonder how in depth you would need to go in order to truly justify highlighting differences in teams that the pitchers faced. In other words, how many times did Kershaw play SF with the wind blowing OUT in left in SF and with Beltran in the lineup, versus in LA withOUT Beltran. I know Beltran wasn’t that amazing this year, but apply that to any player you want. Just something that crossed my mind.
Great stuff on this site, btw.
I meant blowing out in RIGHT in SF.
The big three in the NL were all very close this year in pitching. I do think Halladay and Lee probably put up slightly better numbers than Kershaw if you put them on the Dodgers and let them pitch against SD and SF that many times.
Eric, if I recall correctly you were one of the few Fangraphs’ writers to select Kershaw as your pick for Cy Young winner at the end of September. I even have some faint recollection of you discussing with D. Cameron why the type of stats you discuss in this article weren’t your determining factors in selecting the Cy Young winner. Why the change of heart? Or am I misremembering?
Here we go:
“Dave: So, I took a sneak peak at the FanGraphs author awards ballot, and you’re kind of a traitor. You can make a strong case for Roy Halladay or Cliff Lee, but instead, you pick Clayton Kershaw, even though he has a WAR of 6.8 compared to Halladay’s 8.0. You’re from Philadelphia, you write for FanGraphs, and you pick the pitcher with a lower WAR who doesn’t play for the Phillies? Don’t you know that you’re supposed to be a slave to the stats, and our most recognizable stat says Halladay has been better? You’ve got some explaining to do.
Eric: I’m a loner, Dottie, a rebel. At least when it comes to the Cy Young Award it seems. But I don’t think it’s crazy to support Kershaw for the NL’s best pitcher of the year award even though Halladay has a 1 WAR advantage. Bear in mind that by voting for Kershaw I’m not dissing Doc or his tremendous work this season. It wasn’t like Kershaw clearly stood out above the rest. I wrestled with the decision but ultimately decided that it would be nice to see him take home some hardware. The difference between he and Halladay is actually an interesting proxy for discussing why WAR isn’t the be-all, end-all when it comes to evaluating pitchers. “
I mentioned my flip-flop in the post Eisenator.
Sorry, I missed that when reading it. Would you mind possibly going into a bit more detail about why you changed your mind?
Consistency is the hobgoblin of little minds, but it is a significant shift from your fairly well articulated position just a few months ago. Particularly you made the case Kershaw may be responsible for his low BABIP, even rebutting Cameron when he argued the H/A split you use in this argue.
It’s good to see the stat elitists come out in mass at Fangraphs.
“You questioned our claims to epistemological objectivity!! Mraaaggh!”
This was very good study, but it’s hard to notice when all the egos are blocking the view. Lighten up, fellas. It is a stats site so of course this is a great thread interrupted by some posters who are simply emotional, not armed with facts. But in the long run this sort of thing is not worth giving in emitting the stench of elitist hubris.
You mean, like the stench of this post?
Poor Tom B. Do you need a hug?
No sir, doing just fine.
Just wanted to point out that your post reaks of the same elitist attitude you are referring to.
This article could have been good, but there are far too many mediocre statements.
“It’s unfortunate that one of those Phillies pitchers lost because of unadjusted numbers and them “splitting” votes with one another.”
Splitting votes? Perhaps you should take another look at the results of the voting. Lee did not take meaningful votes away from Halladay. Don’t talk out of your ass.
“After Felix Hernandez won his award it sure seemed the voting body grasped that team context is important when evaluating players.”
I thought they had grapsed that wins were not the end all be all, that’s all I thought of Hernandez winning. Kershaw winning does not change that, it’s not true that he won on his win advantage alone, though it surely helped some.
“Not to mention that all three of the Dodgers, Giants and Padres stadiums greatly favor the pitcher. Kershaw made 19 of his 32 starts in those three stadiums. Pitching in venues where run-scoring is tougher, and facing teams that frankly stink at scoring runs to begin with, goes a long way towards explaining Kershaw’s success this season.”
This makes it sound like 19 of his 32 starts were not impressive, only 9 of those were against “teams that frankly stink at scoring” in the Giants/Padres, the rest were a mixture. Besides, this fails to point out that Kershaw also pitched in Coors field and Chase field allot. You can’t just cherry pick the most pitcher friendly of Kershaw’s starts and act like it makes his season less impressive. I could do the same with Halladay and pick out his starts against the Pirates/Athletics/Dodgers/etc. If you are so hell bent on using this as a main part of the pro Halladay argument, do a fing good job of it. Why not actually compare all of both pitchers starts, their entire strength of opposition and the average park factors of every park they pitched in throughout the year.
The 9 starts against the Giants/Padres argument could have been convincing if it were not so lazy.
By quality of opponents faced:
Kershaw: .263/.327/.416
Halladay: .263/.330/.413
Lee: .266/.332/.423
If anyone should be getting a little bump based on quality of opponents, it is Lee, not Halladay. Kershaw also pitched twice each at either Arizona and Colorado (hitters parks, they ain’t), while Halladay pitched once at Colorado.
I just don’t think the quality of opponents argument carries much water. You could pick any one of the three for Cy this year and get no argument from me.
yet Halladay is 34!
and he is still improving his game….
1) Dodgers Stadium isn’t the pitcher haven it was say 10 years ago
2) AT&T is a neutral park
3) CBP is a neutral park and has been for a few years now
Halladay and Kershaw are tied for ERA+ and Halladay has a 1 point lead in ERA-. I think you could’ve made this article on how close the race was instead of propping up Halladay (and Lee) over Kershaw.
While I’ve seen a number of responses (both here and on Keith Law’s ESPN article that puts forward the same fallacies) shooting down stats that show Kershaw facing anywhere from significantly better offenses (by opponent winning percentage), to very slightly worse teams (by wRC+). I have yet to see anything supporting the idea that Halladay faced significantly better offenses other than the intellectually lazy “SF and SD suck” and “CBP is sooo much more offense friendly.” I also haven’t seen any responses to the link Chad Moriyama posted that shows Clayton Kershaw our performing Roy Halladay after adjusting for opponent and park effects. Show me 1 park adjusted, lineup adjust analysis showing that the totality of Halladay’s competition as significantly better than Halladay, please.
Mr. Seidman, what say you?
I should note that my main complaint here is only about the strength of schedule arguments. Using any other advanced metrics to make a case for Halladay is fine with me. I wouldn’t have been angry if Halladay had beat out Kershaw, I just don’t subscribe to some of the false reasoning I’ve seen.
Have we figured up how many 1/100′s better Halladay’s 2011 was as compared to Kershaw’s?
Certainly it has to be in 3/100 range, illustrating his deserving of the award.
I’m surprised they didn’t just hand it to Doc for being the best pitcher on the best staff on the best team.
I’m being somewhat cheeky, but a solid arguement can be made for both guys.
Someone else made an interesting point about CK being a lefty which is interesting b/c most batters are RH with favorable splits against LHP. So, there may be even more to consider for environment other than opp wOBA, wRC+, etc. Have we looked at those numbers for the relevant teams in regards to vL and vR?
“Someone else made an interesting point about CK being a lefty which is interesting b/c most batters are RH with favorable splits against LHP.”
First of all, the NL had a .311 wOBA vs LHP and an identical .311 wOBA vs RHP this year.
Second of all, a CYA should not be based on uncommon attributes. If a soft-tosser manages a 65 ERA- next year when the league average split against soft-tossers is 110, that doesn’t make his 65 ERA- better than anyone else’s. Having a tougher hill to climb is the soft-tosser’s burden. It doesn’t represent an added value. Parks and competition are a different story. Those are environmental factors that can be measured against replacement level.
I’d say Tim Lincecum changes the run environment. Otherwise, being Tim Lincecum is no big deal. Granted he doesn’t affect the Giants O, but a matchup of CK and TL drastically changes the combined run environment.
Hello, nerds! I’d also like to point out that Roy and Cliff are both having “password issues” yet I had no problems posting my comment. So there!
This argument is dumb. All 3 had excellent years. Saying one faced worse competition more often is a lame argument. Kershaw had a great year and put up great numbers against the teams he was required to compete against. They are all big league hitters and Kershaw excelled. How about we write an article about a 23 year old pitcher who is a class act and will be a stud for years to come?
Yeah if you check out the numbers, Kershaw dominated weak offenses, average offenses, and elite offenses.
As one who doesn’t give a damn who wins any awards, I must say that this article and these comments were nevertheless very interesting.
In order to “do this right” wouldn’t you have to go through and look at …
[1] Every batter faced.
[2] Park pitched in
[3] wOBA v. L & vR.
Teams don;t have the same lineup for every game, so each pitcher may not face the same “team” that we’re using numbers for. Platoons, injuries, rest days, etc.
The whole discussion is absurd in the regard that both guys were absolutely outstanding and either one could have taken the award without complaint. At this point, it’s more of a process of analysis discussion.
We’ve looked at parks, divisions, etc. Let’s forget that for the moment and look at the actual batters faced, how they performed against Kershaw/Halladay vs. rest of league, how drastic the splits were, etc.
Someone mentioned that Kershaw pitched better at home. Don;t most batters hit better at home, even if their home park slightly favors pitchers? The exception would be extreme parks. Kemp performed just about the same at home & away. Shouldn;t he have been far worse at home if we’re using Dodger Stadium as a reason why Kershaw was better at home?
I’d be interested in seeing the average stat line of batters Kershaw and Halladay faced but just with their splits against L or R, respectively. How Kershaw’s opponents fared against RHP’s really shouldn’t have any factoring in, should it?
2011 NL
—————-
RHB v. RHP: .246/.305/.377/.682
RHB v. LHP: .262/.335/.410/.745
LHB v. RHP: .261/.331/.409/.740
LHB v. LHP: .224/.289/.343/.632
LHB hit RHP about as well as LHB hit RHP. The issue is that there are many more RHB than LHB. How does that balance out with LHBs being so much worse against LHPs than RHB are against RHPs?
Of course we’d have to look at the actual stats of the lefties and righties each guy faced, and then regress it to league average by some number that I don;t know. But it would be interesting.
Seems to me though that the actual batters faced are more important than the park factors, team stats, etc when looking at things with this much detail.
On one hand, people say the MVP should come from a winning team, and on the other hand, people say certain guys shouldn’t get MVP or Cy Young Awards because their stats were affected by the team they played on. It’s kind of a contradiction. I think we should just go by the stats and not worry about what factors contributed to them. And that WAR stat is off. It said the Yankees had the best team WAR for pitchers, and Oakland was second, yet the Yankees ranked third in fewest runs allowed and Oakland ranked seventh.
Chad’s link links to an article on AL strength of opponent.
…With a chart showing the top 20 in all of MLB, with Kershaw sitting 3rd and Halladay 6th.
Well, I have read all the comments and now I have to agree that Halladay probably deserved the award. My advanced sabermetrics shows that Los Angeles has lower humidity than Philadelphia and therefore Kershaw did not have as much sweat in his eyes, giving him an unfair advantage. Also, Kershaw is a much better fielder, giving him a lower ERA. Without that fielding advantage, Halladay’s ERA would have won. Kershaw picks off more runners. This is an unfair advantage that has nothing to do with throwing strikes. Yes, Halladay should have won. Tears for poor Roy and all those Philly fans are staining my Dodger souvenir pillow.
Eric,
I think by using SF and Padres to pad your argument automatically leads the reader to question – why were these teams singled out? (other than their horrible offense) I am new to sabermetrics, but I am a believer. Having said that, a more sound argument that would thwart many of strength of opponent voices that I read above would be something like this:
**Average runs scored per game of opposition**
I think this would alleviate many naysayers who say you are cherry picking specific stats to pad your argument. Like I said, I am new to sabermetrics, so I don’t know any easy way to assemble these stats. Judging by how horrible SF and the Padres were, it is likely it would support your argument without seeming selective. Just a thought.
For all those die-hard loyal sabermetrics fans out there ready to pounce – I understand there is more to this argument, but if strength of schedule is brought up it deserves to be properly vetted.
Good article,
Eric
Shucks, guys. I know there are a lot of good pitchers out there, and I’m probably not a better pitcher than Roy Halladay on a scale of 1-100. Maybe I’m a 95, that’s pretty good. Roy would probably be a 99. But gosh darnit I pitched well this season, and I had more strikeouts and lower ERA than Roy. Maybe I’m not a better pitcher, but I had at least as good of a year as Roy, so please, can you stat geeks give me a little credit?
Because baseball is so streaky, why don’t we look at how each player in a lineup had performed in the 25 ab before and the 25 ab after the game. Estimate the run total that lineup should have produced, adjust it for the park, then see if the pitcher gave up/held them to more/less than they should have scored.
Baseball is too streaky, and lineups change with injuries, trades, and again, streakiness, all the time. So to say “well so and so pitched against this crappy team 4 times” is kinda dumb. Just look at the marlins month-by-month w/l. 2 awful months but mostly a decent-good team. Or player-wise, Lance Berkman’s first 2 months vs the rest of the year. Or that streak where the Red Sox scored like 30 runs/game or something insane like that.
If we’re going to adjust to find ‘true talent”, don’t half-ass it and go all the way in depth with it.
I would love for a Fangraphs writer to really examine the Halladay HR/9 point mentioned above by dte421? Was there some luck there at all? I don’t know but would like to see it properly addressed in an article such as this.
By fWAR, Halladay was ~20% better than Kershaw (which is a massive gap 6.8 vs 8.2).
But it looks like Halladay (albeit he is an excellent GB% pitcher) was somewhat lucky on his HR/9 due to a low HR/FB%, especially as his home park should increase HR?
And it seems like Kershaw has ‘some’ BABIP skill that is discounted totally from WAR.
does this aaccount for the 20% fWAR gap? Well rWAR has them much closer (but still Halladay top), so probably not, Halladay was probably better, but the gap is not 20% as espoused by fWAR.
So maybe the no.2 guy won over the 1 guy, but as the the no. 2 guy was the youngster, and the 1 guy had already 2CYA (as well as the TripleCrown) so i find the result totally defensible.
To be hoenst this is one of those arguments that put people off advanced stats, I find it hard to believe that most basball fans would accept that Halladay was 20% better than the next best (or joint next best) pitcher in the NL this year.
I would have voted for Kershaw based on the triple crown, 21 wins for a lousy team, very similar statistics to Halladay, and the reality that they play a very similarly competitive schedule, despite what the author (a Phillies fan) of this article suggests. While many of the baseball writers of America may be voting on past metrics, I think the writers of fangraphs are sometimes full of themselves, smoking too many sabermetrics to even see straight. This is a good example of that. I love the sight but this is a very short sighted article.