Dr. Doom here (via Doug) again with an awards-voting post. We’re sticking with the Cy Young, but switching over to the National League.
Remember how there were two candidates in the AL who (at least on paper) stood head-and-shoulders above everyone else? Yeah… get ready for a repeat.
Clayton Kershaw is Clayton Kershaw. He continued to Kershaw. He Kershawed all over the season. Did he miss like 20% of his starts? Yes. Did he still somehow lead the league in wins – yes, with 18. Did he lead the league in ERA? Yes, with a 2.31. Did he lead the league in K:BB? Yeah, of course he did, 6.73. His 180 ERA+ was tops, too. Kershaw was the most efficient pitcher in the NL; was the lack of volume enough of a problem to hold him back?
If so, look at Max Scherzer who was nearly as good as Kershaw, just two wins behind the Dodger ace, Scherzer was also second in ERA (2.51), and a narrow second in ERA+ (177). Unlike Kershaw, Scherzer played his home games in a decidedly hitter-friendly park; but, that didn’t stop him from pacing the NL with a .902 WHIP and an ultra-stingy 5.7 H/9, the latter the 7th best number of all-time. Oh, and Scherzer’s 268 strikeouts were also tops in the league.
Here are some other top candidates:
- If there’s a third horse in this race, it’s Stephen Strasburg. Strasburg was third in ERA and ERA+, with marks of 2.52 and 176 respectively, and also third in FIP with 1.015. He allowed the fewest homers in the NL. Strasburg has not, perhaps, become the pitcher we were promised, but he’s become a force.
- Gio Gonzalez, the third Nats’ “ace”, was just about as good, with a 15-9 record and 2.96 ERA that might get it done in a lot of other years. Gonzalez’s weaknesses (a lower strikeout rate and weaker FIP numbers) perhaps don’t cut it as much in today’s analysis; there’s an argument there, too, but it’s a case that I’d be interested to read.
- Zack Greinke had an “up” season in 2017. When Greinke is on, he’s as good as anyone in baseball. He doesn’t outrank any of the top candidates in any significant category, but he’s a very solid down-ballot choice.
- Robbie Ray had his breakout year in 2017. 4th-best ERA (2.89) and ERA+ (166). He struck out the 3rd-most batters (218).
- And here’s my shameless plug as a Brewers fan for Jimmy Nelson, who was shockingly good. Third-best FIP (3.05); sixth-best K:BB (4.15). Only Strasburg allowed fewer homers.
- If you’re looking for others, there’s a fascinating Jeff Samardzija argument; Jacob DeGrom had a really interesting season, too; and Aaron Nola is young and plays for a terrible team.
I’d be interested to hear whoever else you’re interested in. Can’t wait to see what you all have to say!
DIRECTIONS: Please list 5 players on your Cy Young ballot in a NEW comment below (ballots with fewer than 5 candidates will be thrown out; I ask for a new comment because it’s easy to lose one if it’s in a reply, especially since we got rid of numbered comments). Ballots will be scored as per BBWAA scoring (7-4-3-2-1). Strategic voting is discouraged, though that’s unenforceable, so please just don’t do it, as the goal here is to (somewhat) mimic the BBWAA process. The post will be live for about a week; I will comment shortly after the post goes live to tell you when ballots are due. Please discuss and vote whenever you’d like, but there will be NO vote changes, so don’t vote until you’re sure you’re ready!
Two of the six H/9 seasons better than Scherzer’s belong to Nolan Ryan, the all-time career leader in H/9. What’s notable about those two seasons is that they were separated by 19 years, at age 25 and 44. Of those two campaigns, Ryan had better marks in WHIP, ERA+ and SO/9 at age 44, his third straight season leading the majors in H/9 and SO/9. Ryan topped 5 WAR as a 44 year-old, the oldest pitcher to do so, and the only hurler with three 5 WAR seasons aged 40+.
Nolan Ryan’s CAREER H/9 would rank #151 of all-time SINGLE SEASON H/9 numbers. It’s a pretty remarkable achievement. Of course, he was walking a lot of guys. The difference between Ryan’s 2795 career walks and #2 Steve Carlton’s 1833 is the same as the difference between Carlton and 181st place on the all-time list. If you add up the total walks issued by the THREE active players with the most walks allowed (CC Sabathia – 1009, Bartolo Colon – 923, Ubaldo Jimenez – 848), you’re at 2780 – still shy of Ryan.
Since the other election is still going on for another day, let’s keep this open all the way through next Sunday night – 12/3, 11:59:59 PM. Happy voting!
Strasburg bested Kershaw by almost 2 WAR (6.5 to 4.6), and did so in the same number of innings and with only one more start.
Nats are the first team since the 1891 Boston Beaneaters to have three pitchers with 6 WAR, 15 wins and a .600 W-L%.
The 1890’s – the good old days; as the horses died in the streets of Boston from cholera, that Beaneaters staff surrendered 658 runs – only 381 earned ! I really believe there are great players in every era, but it sure is difficult to compare them out of their era. James made the argument that Nichols was better than his contemporary Cy Young for the first 12 or so years of their careers and then Young just finished better than any body else ever did (possible exception of Nolan Ryan?). But, I have no idea how you can make a realistic comparison of Nichols to Greg Maddux or, for that matter, Ed Walsh and Jack Chesbro (deadball era) to even Lefty Grove (15 years later).
Scherzer
Strasburg
Kershaw
Ray
Jansen
Doug, you are correct – Aaron Nola plays for a terrible team
Wanted to throw something out for discussion. How much weight should we give to FIP in a single season? Isn’t the stat more useful in determining the possibility of future regression than it is in evaluating the net results for one year, unless we have two otherwise nearly equal pitchers? I realize this is somewhat Luddite, but we send pitchers to the mound to get results–which is to get outs that result in fewer runs and more (team) wins. Robby Ray has an ERA+ of 166, 4th, but he’s only 10th in FIP with a not eye-popping 3.72. Why would I care about FIP?
That’s a very fair question. The argument against is the one you’ve already made: pitchers are sent out to get results; those matter most. The argument FOR is this. Pitchers and batters only have three interactions that don’t involve fielders: home runs, walks (or HBP), and strikeouts. EVERY other play (well, except pitcher errors, I s’pose, but that’s a pretty limited number of plays) involves fielders. However good or bad fielders are greatly affects the pitcher’s results. Yes, FIP better corresponds to future events; perhaps the reason is that it actually has more to do with underlying talent and removing luck than actual ERA. Maybe that makes sense to you, and maybe it doesn’t. Either way, it’s a good and fair question to ask.
Thanks for the response. So, let’s roughly normalize for fielding by picking just one team (and one set of fielders)
Brewers had a tattered rotation (13 different starters). Three of them had more than 140 IP:
Zach Davis, 191 IP, 3.90 ERA, ERA+112, FIP 4:22
Jimmy Nelson, 175 IP, 3.49 ERA, ERA+126, FIP 3.05
Chase Anderson 141 IP, 2.74 ERA, ERA+160, FIP 3.58
What should I be making of this? Did Anderson just pitch in a lot of luck?
Re FIP:
Of the top 100 player seasons in FIP, only one occurred between 1918 and 1962, Hal Newhouser’s in 1946. Of the other 99, 11 occurred between 1962 and 1972, one in 1984, 2 in the 1990s, 3 in the 2010s, and the other 86 prior to 1919. Does this mean that there was a great dearth of top flight pitching in the 1920s through the 1950s, and from the early 1970s through the late 1990s? Or does it mean that there’s a correlation between low FIPs and eras of pitching dominance? If the answer to the latter question is yes, then it’s not surprising that starters who go five innings and K a bunch of batters are now seeing lowered FIPs. Despite some appearances to the contrary—notably the number of HRs hit—we’re now in just such an era. Lots of HRs got hit in the 1960s, too, come to think of it.
“Does this mean that there was a great dearth of top flight pitching in the 1920s through the 1950s, and from the early 1970s through the late 1990s?” It’s hard to believe that could be true for such extended periods. I think it’s more likely style of play, what managerial expectations were. Look at one of our favorite whipping boys of the past, Whitey Ford. Ford had a 2.75 ERA and a 3.26 FIP. His K/9 was 5.6, a level that today would probably keep him fighting for a roster spot. Yet he finished in the top ten (AL) in strikeouts ten times, including a 4th in 1955 with 137 (137!). In the Dead Ball Era, you just weren’t going to get HRS. For later periods, old-fashioned mind-sets about going the distance had to merge with the reality that the ball could go over the fence. Except for physical freaks like Feller, you had to pace yourself.
nsb, I have to call you out for posting something likely to mislead others. You’ve implicitly claimed that there’s a fundamental flaw in FIP, in that “only” 14 of the top 100 seasons in history came post-1919. Well, FIP isn’t ERA+; it doesn’t normalize for era. Rather than replacing ERA+, it replaces ERA; therefore it absolutely correlates with periods of pitching dominance, exactly as ERA does. Did you know that only EIGHT of the top 100 ERA seasons come post-1919? So you have said that FIP indicates that most pitchers in history have not been as good as early in the game, when in fact it is MORE equitable than ERA, the statistic it mimics. Just some food for thought.
Doom: not trying to mislead, just trying to supply a fact I came across while looking up FIP and then to pass it along with two implied observations—that low FIP correlates strongly with eras of pitching dominance, and that we’re in one of those periods now.
As for their being a flaw, fundamental or otherwise, in FIP, I honestly believe that it is just a stat that gives a different look at pitching. Don’t think it’s nearly as important as some others. In 1968, though, there were around 10 FIPs in the top 200 all time. Don’t have it in me to go back and strain my eyes for an exact count.
Or, come to think, it might have been the top 300. That’s how deep I went into the list.
Could someone isolate the career FIP leaders for the live ball era, starting with Pete Alexander, who’s a notable carry-over?
Perhaps Doug would be willing? He has a PI subscription.
Here are the top-30 by my eyeballs (and btw, there were 18 post-1920 seasons in the top 100):
1. Pedro Martinez, 1999 (1.395)
2. Dwight Gooden, 1984 (1.685)
3. Bob Gibson, 1968 (1.775)
4. Clayton Kershaw, 2014 (1.811)
5. Sandy Koufax, 1963 (1.852)
6. Sandy Koufax, 1965 (1.927)
7. Tom Seaver, 1.931 (1971)
8. Hal Newhouser, 1946 (1.966)
9. Clayton Kershaw, 2015 (1.991)
10. Matt Harvey, 2013 (2.005)
11. Steve Carlton, 1972 (2.009)
12. Luis Tiant, 1968 (2.041)
13. Bob Moose, 1968 (2.063)
14. Sandy Koufax, 1966 (2.071)
15. Sandy Koufax, 1964 (2.076)
16-t. Randy Johnson, 1995 (2.081)
16-t. Sam McDowell, 1965 (2.081
18. Don Sutton, 2.082
19. Bob Veale, 1965 (2.107)
20. Bill Gullickson, 1981 (2.111)
21. Dave Righetti, 1981 (2.116)
22. Dwight Gooden, 1985 (2.127)
23. Randy Johnson, 2001 (1.128)
24. Sandy Koufax, 1962 (2.151)
25-t. Bob Feller, 1946 (2.157)
25-t. Mike Scott (2.157)
25-t. Don Sutton, 1971 (2.157)
28. Pedro Martinez, 2000 (2.171)
29. Roger Clemens, 1988 (2.174)
30. Roger Clemens, 1990, (2.177)
I didn’t use Pete Alexander, as his only relevant seasons were in the 1910s; if I included him, why not Walter Johnson? If them, why not Dutch Leonard, Rube Marquard, Eddie Plank, Eddie Cicotte, etc., etc. So I just used 1920 as a clean cut-off.
I’m just not understanding your point about pitching eras. I mean, FIP corresponds almost perfectly with ERA, so it would make sense that it’s low in pitching eras, just as ERA is. That doesn’t make it any more or less useful a tool than ERA, from my perspective.
Dr. Doom; I checked the list via the PI. You are spot on but you have your numbers reversed for number 7, Tom Seaver.
Eighteen is what I got, too, Doom. I’ve just lost the ability to subtract in my dotage, so it seems.
And here’s food for thought….Greg Maddux’s breathtaking 1994 ERA 1.56, ERA+ of 271, but FIP of 2.39, and his equally insane 1995 ERA 1.63, ERA+ 260, FIP 2.26 don’t make the list.
Something is slightly discordant.
Again, as someone who’s been researching this stuff for a long time, I don’t see what’s discordant at all about that. Maddux in ’94 and ’95 had PHENOMENAL numbers in terms of FIP. You’re making a huge mistake of forgetting context. It would be like saying something is amiss in ERA because Pedro Martinez’s 2000 season, possibly the greatest pitching season of all-time, is not in the top 100 seasons of all-time by ERA. OF COURSE it isn’t; part of what makes it so remarkable is that the league ERA was 4.91, and his ERA was 1.74. No, historically, a 1.74 does not put you among the best 100 seasons of all-time. But if you look at ERA+, Martinez’s season ranks second all-time (1st in the 60-ft mound era). Yet, by FIP- (the Fangraphs equivalent of ERA+, but for FIP), Maddux’s ’94 and ’95 rank as better than Carlton’s ’72, any of Koufax’s seasons, and Gibson’s ’68. The problem is that people are asking a context-related stat to NOT be a context-related stat. It would be like saying, “Mike Schmidt had no power; he never hit 50 HR!” It doesn’t work that way.
Hey, I said I didn’t know what I was talking about. My use of the word “discordant” was merely to display my sometimes-useful vocabulary. I’m going back to politics…I have another piece due next weekend, have no clear thoughts on it, and. based on this exchange, will not talk about baseball.
Mike, If I’d noticed Maddux’s figures in the context of this discussion, I would have written a post like yours and been schooled by Doom instead of you. Except I probably would have typed “dischordant.”
EPM, that got me laughing. I think “dischordant” relates to a cheap guitar. But I wouldn’t fret over it.
My son is asleep in the rim next to me, and that fret joke made me giggle hard enough that I worried I woke him. Don’t leave us because of my lack of couth, though. Your diction and commentary are welcome. And who needs more thoughts about politics, anyway? You’re better off in baseball. Sorry if I was harsh. I realize not everyone is as familiar with these things; I still need the reminder once in a while, though.
Doom, don’t sweat it. The fault is mine-I’m just not grasping the utility of FIP as a comparative measuring tool. The hardy band of survivors at HHS get the math better than I do, but it’s like going a concert–just because I can’t play the sax https://www.youtube.com/watch?v=acBixR_JRuM
doesn’t mean I don’t like the sound.
Seriously, I have eight days to come up with something brilliant (or at least passable) and you might have just given me an idea. Last time I wrote about coal miners. http://www.3quarksdaily.com/3quarksdaily/2017/11/why-did-the-coal-miner-refuse-to-cross-the-road.html#more
The link to the best FIP seasons is: https://bbref.com/pi/shareit/bDPZG
I’m trying to wrap my head around the idea of Bob Moose at #13. At first I thought it must just be one of those things like Norm Cash hitting .361 or Davey Johnson hitting 43 home runs (or however many it was) but then I scroll down and there he is again at #102 and #106.
Or Bob Bruce at #62 for his 1964 season. The 60’s were the zenith of my youthful infatuation with baseball. I not only had the vast majority of all of the baseball cards produced from 1964 to 1968 but knew almost all of the information on the back chapter and verse.
I have absolutely no memory of Bruce.
None whatsoever.
It seems odd to see him ahead of Cy Young seasons by Tom Seaver, Greg Maddux, Steve Carlton, Dean Chance, Fergie Jenkins and a whole bunch of others.
I have lots of conflicting feelings about Kershaw. (Whether he deserved the CYA is not one of them: he didn’t, though I can’t see how his pWAR drops as low as 4.6.) On the plus side, he not only had a lot of wins, he also had very few losses, a net difference of 4 on W-L compared with Scherzer, which is significant. Moreover, the Dodgers went 23-4 in Kershaw’s starts, vs. the Nats going 21-10 in Scherzer’s (despite which, I’ll be putting Scherzer at #1 when I get around to voting). And, of course, the black ink categories are, as usual with Kershaw, very impressive (of course it was touch and go whether he’d even qualify for black ink . . . still, he pitched only about 25 innings fewer than Scherzer, Greinke, and Gonzalez, and the same as Strasburg).
But here’s the stat that bothers me most about Kershaw. Everyone knows he gave up a lot of HRs this year. But the extent of the damage in terms of his pitching quality was beyond the pale. Compare these figures, which show for each candidate Doom mentioned the percentage of total season runs allowed that scored on HRs:
Nelson___31%
deGrom___34%
Strasburg___38%
Gonzalez___41%
Nola___43%
Samardzija___45%
Greinke___47%
Scherzer___47%
Ray___49%
Kershaw___69%
Moreover, for those who like to see home/away consistency or greater strength in “neutral” (away) parks, Kershaw looks like a poster boy: identical H/A W-L records, hits allowed, walks allowed: it’s eerie. But when it comes to proportion of total run scoring on HRs, for Kershaw at home the figure is 81%. In his four losses, 87.5% of all the runs he allowed came on HRs. This is not normal. Take away those HR pitches, and Kershaw is unbelievable: he allows only 0.77 other runs to score per 9 IP. Even if he had a ratio like Ray’s (next worse among contenders), that would yield a total run average of about 1.55 (or an ERA in the neighborhood of 1.40). By this logic, gopher balls alone inflated Kershaw’s ERA about a run.
There have been great pitchers who gave up lots of HRs (Roberts, Blyleven). But have we ever seen anything like Kershaw 2017: historically dominant when the ball stays in the park, but letting over twice as many total runs score on HRs as otherwise?
Kenley Jansen had one of the best relief seasons in modern times, according to WAR and WPA.
In a weak year, he is definitely getting a look. Probably will get a 5th place vote from me.
I have been thinking along similar lines, ThickieD.
Normally I don’t see closers as viable CYA candidates because of low IP totals, and because I think the average closer is doing something most solid starters could also do: pitch one effective inning under pressure with an arm that pitches few innings during the season. But the weakness of the CYA field — in which the leading candidates all pitched modest innings totals for starters — combined with Jansen’s near-perfection in his limited, assigned role, seems to me to open the door for an exception. (This thinking was behind my 10th-place vote for Jansen on the MVP ballot.)
Exactly my thought process.
If more guys had gone 230 IP (or even 200), Jansen wouldn’t make the cut – nowhere near – but this year, there is a dearth of really strong candidates, IMO.
Scherzer is #1, no question, and did have a really good year (I had him 10th or 11th on my MVP ballot, I think). After that, I like Strasburg and Greinke – due to low IP totals they weren’t even really in consideration for MVP, although I guess Strasburg would have been in my Top 15 or 20.
I’ll probably put Degrom at #4, unless I decide Kershaw’s somewhat better performance overall makes up for the 25 inning differential.
At that point, I can see slotting Jansen in at #5 if only to recognize his extraordinary, once-in-a-generation relief season.
389 guys have pitched at least 150 innings and had their HR allowed be at least 15 percent of hits allowed.
ERA under 5 … 305
ERA under 4 … 140
ERA under 3 … 12
2.31 … Kershaw (2017)
2.51 … Scherzer (2017)
2.61 … Johan Santana
2.68 … Billy Pierce
2.69 … Jim Bunning
2.79 … Scherzer
2.83 … Yu Darvish
2.89 … Robbie Ray (2017)
2.96 … Scherzer
2.98 … Jeff Robinson
2.98 … Curt Schilling
2.98 … Oliver Perez
Voomo, Your figures make it clear that 2017 was an exceptional year in this regard. However, besides Kershaw, the other two 2017 names on this list don’t necessarily show a deviation from past performance that signals a change in their records in this regard. Scherzer simply continued his trend from the last two years, and Ray’s change was that he brought his hits and ERA under control for the first time: 2017 establishes his initial profile as a quality pitcher.
But 2017 marked something dramatically new for Kershaw. The figure I’m using is different from yours (not HR as % of Hits, but Runs on HRs as a % of total Runs Allowed), but they’re related. Look at my figure (HRs as % of total runs) for Kershaw since he first established his profile as a quality pitcher in 2011:
2011__36%
2012__36%
2013__24%
2014__33%
2015__32%
2016__32%
2017__69%
Something changed for MLB in terms of HRs in 2017, though the change was on the order of +10% (which is certainly a lot). But in Kershaw’s case, the change was over 100% from his six-year average — the change was with Kershaw, not MLB.
(Interestingly, in Scherzer’s case, the change went in the other direction. Last year, he had a Kershaw-like figure of 62% of all runs allowed being scored on HRs. This year he cut that to 47%, just below his 2015 figure of 50%. But Scherzer does look like Kershaw a bit if you look back to his breakout year of 2013, when only 34% of his runs allowed scored on HRs.)
I’m finding this hard to sort out, and I do not know if I will vote.
…
Part of the difference between Kershaw and Max, other than the 4 extra starts, is that more was asked of Scherzer, in terms of pitch count. Do we know if this speaks to:
* their coaches knowing their capabilities
* the fact that the Dodgers had a better bullpen
* or that Dusty was old school and L.A. went by the modern 100-pitch book ?
Starts with more than 100 pitches:
22 … Max
12 … Clayton
Or, it could be that Kershaw was simply more efficient.
Aha.
They each threw 7+ innings 16 times.
Scherzer went over 100 pitches in every one of those starts.
Kershaw in only 7 of those 16.
He never game up more than 2 runs in any of those appearances, and the Dodgers won every game.
___________________
And the CY isn’t an MVP(itcher), is it?
It’s “best pitcher”.
But some folks see it as an MVP.
It is not clearly defined.
And seeing that both teams were runaway division winners with multiple “aces”, can either of these guys be considered MV?
Or would we have to look at a guy on a scrapping-for-a-playoff-spot team?
No Denver pitchers threw enough innings.
Jimmy Nelson was shut down after Sept 8th.
Greinke was good-not-great down the stretch.
Robby Ray was lights-out (save for one game) at the end, but missed 4 weeks in August.
No Cubbies stood out.
… So, is it a year worth looking at a reliever?
Davis and Holland certainly were valuable, but we couldn’t reasonably give them a vote ahead of Jansen, who was amazing.
…
However, there is a guy who was arguably better than Jansen, and played for a scrapper.
Corey Knebel
Unfortunately for his case, he gave up 5 of his 15 runs in the last week and a half, contributing to 2 losses
(they missed the 2nd WC by 2 games)
I don’t think it was Dusty being old school, as much as Dusty not having a bullpen for half the season. He certainly wasn’t old school when pulling Scherzer in a playoff game at 98 pitches and only one hit allowed, unfortunately (for Scherzer) to the last batter he faced.
By the way, Doom, when you mentioned a “fascinating Jeff Samardzija argument” I figured it was a lighthearted moment. Was it? If not, I’d be fascinated indeed.
Not a joke; again, this goes to how seriously you take fielding-independent numbers. But Samardzija walked the fewest batters in the NL (as a rate), while pitching the most innings. Fangraphs has him as the 8th most valuable pitcher in the NL, in spite of leading the league in losses. He was second in the league in SO:BB. The Giants were terrible, both defensively and offensively. There’s a strong argument, I think, that, with a different club, Samardzija ends up with a 2.80 ERA instead of 4.40, given the right team, and with a 15-9 record instead of 9-15. If that happens, he’s worth a down-ballot vote. But that would involve a lucky season, rather than an unlucky one, a good team rather than a bad one, otherwise stable performance, AND a belief that his underperformance of his more sabermetric numbers was a result of chance, rather than skill. I figured I’d throw it out there in case someone wanted to ask, and lo and behold, someone did! 🙂
Thanks for the explanation, Doc. I figured that if you did mean to build a case, BB/9 and Giant run support would have to be part of it. The low BB figure helps out Samardzija’s FIP a lot (his figure’s ok, but still only 8th in the NL, short of CYA territory), but still, he gave up 30 HR and 48 runs scored on HR. Kershaw, whose problems on that front were prodigious, allowed only 71% as many runs on HR as Samardzija in 84% as many IP. . . .
Enough: you’re trying to make the hard argument, and I’m not going to pursue the easy one further. It is indeed possible to imagine a guy with 3.61 FIP pulling off a 2.80 ERA (Kershaw 2017 shows how that sort of thing is done, though he’s doing it more than a half point lower), and in that universe, Samardzija, well . . . he coulda been a contender.
Yeah, he underperformed his FIP by 0.81 runs. If he goes 0.81 in the other direction with a better team and a more favorable ballpark for homers, he’s in contention – not to win the award, mind you, but for a down-ballot slot.
Leading league in IP and BB/9.
3 – Robin Roberts – 1952, 1953, 1954
2 – Walter Johnson – 1913, 1915
1 – Jeff Samardzija (2017), Roy Halladay (2010), Greg Maddux (1995), Fergie Jenkins (1971), Jim Kaat (1966), Lew Burdette (1961), Pete Donohue (1926), Eddie Cicotte (1919), Pete Alexander (1917), Christy Mathewson (1908), Cy Young (1903)
Jenkins 1971 NL CYA winner
Maddux 1995 NL CYA winner
Halladay 2010 NL CYA winner
Kaat 1966 – Koufax took all 20 1st place votes (1 ML award)
Burdette 1961 – Ford had 9 1st place votes, Spahn 6, Lary 2 (1 ML award)
If there had been league CYA’s in 1966, hard to imagine that Kaat wouldn’t have won it with his 300+ IP and 25 wins for the defending champs, five more wins than anyone else.
Walter Johnson also won 25 in the two seasons shown here (and a bunch more). Only other Senator/Twin to win 25 – General Crowder in 1932.
…. a little late with this but, how about Alvin Crowder is 3rd all-time (behind Grove ,136, and Hubbell, 133) in the live ball era (1920-present) for wins (124) from ages 29 through 34 ? The General is tied with Bob Lemon
I’d like to hear that, too.
Because his numbers are lovely (well, other than W/L, ERA, and WAR).
…
Let’s look at Game Logs.
20 Starts in which the Giants scored 3 runs or fewer.
11 with 2 or fewer
…
7 extremely good starts where he got a no-decision.
5 Quality Starts with a loss.
So, put him on a team with an offense, give him all the luck in the world, and a W in every gave with at least 6 IP and no more than 3 runs allowed, and his W/L would have been: 20-12
At the risk of being tedious, I’d like to go at FIP again. Here is the info I was hoping I could con someone else with more knowhow into providing, so as not to make minor errors. I may have made a couple. My eyes, seriously, are not what they might be.
Career FIP in the live ball era:
The 1000 inning cutoff at B-R is far too generous, plus it includes relievers, while I’m only interested in pitchers who were predominantly starters, but OK, I’ll try to deal with it.. I come up with 72 live ball pitchers out of the top 250 in the career list, from Kershaw at 2.599 to Derringer at 3.265. Of those 72, 19 were predominantly relievers, most of whom never made it to 1200 IPs, so out they go, leaving 53. Six of the remainder are active, Kershaw, Strasburg, Kluber, Sale, Baumgarner, and Sherzer, so I’m putting them on hold, leaving 47. Of those 47 remaining, around 20, or 42.5%, pitched either predominantly in the period from the early 1960s to the early 1970s, or, in the case of Don Sutton, appear in the top 250 because of the low FIPs they recorded in that span during a longer career.
Here is an approximate breakdown by predominant innings pitched per timespan:
1920s—1 (Dazzy Vance)
1930s—1 (Dizzy Dean)
1930s- mid ’40s—1 (Paul Derringer)
1940s—4 (Hal Newhouser, Mort Cooper, Max Lanier, Harry Brecheen)
1950s—2 (Mike Garcia, Billy O’Dell)
1950s-mid ’60s 1 (Whitey Ford)
1960s—16 (Jim Bunning, Sandy Koufax, Bob Veale, Bob Gibson, Bob Moose, Dean Chance, Don Drysdale, Juan Marichal, Sam Mcdowell, Don Wilson, Jim Maloney, Turk Farrell, Dave Giusti, Gary Peters, Fritz Peterson, Larry Dierker)
1960s-70s—3 (Bill Singer, Gary Nolan, Mickey Lolich)
1960s-1980s— 6 (Don Sutton, Gaylord Perry, Steve Carlton, Ron Reed, Jerry Koosman, Tom Seaver)
1970s—2 (John Matlack, J.R. Richard)
1970s-1980s—3 (Bert Blyleven, Steve Rogers, Rick Reuschel)
1970s-1990s—1 (Nolan Ryan)
1980s—2000s—3 (Roger Clemens, Greg Maddux, Randy Johnson)
1990s—2000s—3 (John Smoltz, Pedro Martinez, Curt Schilling)
2010-present—6 (Kershaw, Strasburg, Kluber, Sale, Baumgarner, Sherzer)
Live ball era starters in the top 100: 6
In the next 100: 21
In the next 50: 26
Comments:
Three of the 1940s pitchers were Cards. With Newhouser they could play bridge. WWII factor in play?
There are no long career pitchers on the list into the 1960s, then comes a bunch.
3) As I’ve argued in previous remarks, the evidence is pretty clear that low FIP correlates with eras of pitching predominance, and that the rise of the 2010s crop of young masters is a sign that we’re in such a period.
Not tedious at all, nsb, This is great information and we’re fortunate you decided to tabulate it for us. I’m not certain, however, that I’d agree with your conclusion that we are in an era of pitching dominance. Or perhaps I should say that if we are, the meaning of “pitching dominance” is not the same in the 2010s as in earlier eras.
We know what a year of pitching dominance looks like: compare this past season to 1968 (the compilation numbers are average per team):
__________RA/G_____HR______BB______K________FIP
2017_____4.65______204_____528___1337_____4.36
1968_____3.42______100_____458____957_____2.98
I think, in fact, we are in an era of hitter dominance, and this would be supported by high batting and slugging averages, XBH counts, etc. However, where I think you are right is that we’re in an era with dominant pitchers — an unusually large group of elite starters who deviate from the average player by degrees that are also unusually large (sort of like mini-Pedro-2000s).
It seems to me it is the nature of the hitter’s era we’re in — superhigh HRs; super-high Ks (compare the Ks in 1968) — that creates the conditions for this, but if I happen to be right, I can’t say that I actually understand the specific mechanics of the dynamic behind this.
nsb, to your point that “the evidence is pretty clear that low FIP correlates with eras of pitching predominance,” yes; it necessarily does that. FIP is calculated using a constant (a constant that’s recalculated each season) so that FIP and ERA are always the same at the league level. If the ERA is low, the FIP is low; if the ERA is high, the FIP is high. It’s not correlation; it’s a dependent relationship. It’s akin to saying, “I’ve noticed that home runs correlate to eras of good power hitting.” Well, yes; but that’s because the one thing defines the other.
As to whether we’re in a pitching dominant era, the most offense-heavy era in history gave us Clemens, Martinez, Johnson, Maddux, Smoltz, and Glavine. Those are six phenomenal pitchers whose stats run against the grain. Outstanding performance tends to be outstanding, even in unfavorable circumstances. I think we’re just in an unusual offensive era, and some pitchers seem to have found their niche in it. But who knows? We’ll just have to see where things are going in the future.
Something I should have done earlier but didn’t—i.e., taking a look at recent league FIP averages—tends to substantiate what you and epm are saying about the nature of the current era, but I wonder a little all the same if, as you also say, the current era is “unusual.” “Unprecedented,” is the term I’d use, and hard to pin down.Unprecedented strikeouts; unprecedented HR production up and down lineups; pitch counts determining outcomes; bullpens full of one inning or even one PA specialists. I haven’t verified this through research—it’s beyond me—but there seem to be more blowout games or possibly more high-scoring games (I’m thinking double digits) at the same time that there games in which one team absolutely shuts down the other.
Can scoring still be high in a pitcher’s era?
I wonder whether you haven’t identified in passing one of the key conditions that would allow for dominant pitching in the current hitter’s era, nsb: pitch counts.
Basically, less is being asked of all dominant pitchers than was the case in the past. Look at the highest IP totals among those who received Cy Young votes at decade intervals:
2017__214
2007__241
1997__264
1987__281
1977__319
1967__302
1957__271 (only two candidates)
Just as closers now rise to unheard of ERA and WHIP figures, with their mission reduced to the intensity of a single inning, starters now can count on pacing their “fuel use” for a 100/110-pitch mission, rather than anything up to 140/150.
Even so, many seem fragile, winding up with a two-week or one-month rest because of injury: perhaps an injury that was brought on by the way seven intense innings wears on the entire body in a way different from the old nine-inning, pace yourself, standard. So perhaps rather than saying that less is being asked of dominant pitchers, it would be better simply to say that with the mission change in starting pitching, a broader range of quality differences in basic stuff and control is becoming more evident.
Since you added a late post on the last string, Doom (linking to Tom Tango on Aaron Judge), I thought I’d mention here that I’ve added a comment suggesting some common ground between the issue there to this discussion on FIP.
nsb has claimed that he is risking tediousness in his pursuit of FIP data, but I think this post is going to set a standard of tedium that it will be hard for him to challenge.
On the last string, I wrote at length about issue of chance/luck, and then decided to tack on a thousand words more. Anyone who made their way through that won’t be surprised to find that I’m very wary of over-reliance on FIP. FIP is an attempt to assess pitching quality independent of “chance,” by which is meant specific batted-ball placement that no one player determines and defensive skill that the pitcher cannot control (that is, fielding). Since I don’t see batted-ball placement as chance, and do see it as very much a product of pitching skill (together with batting skill), I don’t believe we get a clearer picture of pitchers if we exclude it. And I’m very skeptical that variation in fielding quality is of such magnitude that we learn more about a pitcher by excluding from consideration any play that involves it.
Having said this, I like FIP for what it can do: I think it’s a good reflection of a pitcher’s stuff and control, and that’s precisely why it has predictive value. What it doesn’t tell us is how well the pitcher pitched in real game situations, because it excludes such a large proportion of the real games.
I also think it can lead to conclusions that are not necessarily valid. I’m going to do something really dangerous here and pick on Dr. Doom. In his famous “Samardzija Argument” (ut supra — I hope you’re impressed: I had to look that up to be sure), Doom said that it was possible to believe that Samardzija’s 4.42 ERA could have been a 2.80 ERA had he played for a good defensive team, since his FIP was 3.61, equidistant between those two extremes. I think this is the way FIP is often treated, with the idea that as a Three True Outcome measure, it provides a true measure of isolated skill, while everything else is polluted by chance: in Samardzija’s case the bad luck that put him on a bad defensive team.
(Perhaps I should add that Doom himself actually flagged the problem of that premise by specifying that his suggestion about Samardzija made sense only if you accepted it — I’m not actually picking picking on Doom, I’m picking on Doom’s hypothetical “Samardzija Arguer,” who is not Doom but a fictional creature of his creation.)
If you don’t buy the “everything non-TTO is dominated by chance” point of view, then excluding so vast a range of data concerning pitcher skill performance seems wrong if the question is, “How well did he pitch,” rather than, “How good is his stuff and control?” The problem has to do with our tendency to ask for answers to complex questions in a single number. FIP is great because it gives a nice, one-number metric to answer one question, but we can’t elevate it to greater importance just because we don’t have a reliable one-number metric to answer the other, more inclusive one.
To illustrate with Samardzija again, Doom actually points to two aspects of his record that could be distorted by the Giants’ weakness: the second is the team’s lack of run support for Samardzija, which, Doom speculates, suggests that Samardzija’s pitching could equally well have yielded a 15-9 W-L result, rather than the 9-15 record he was stuck with, if he’d played for a good run-producing team (which means, I suppose, that he pitched well enough to expect a 12-12, .500 record). We generally handle this with two metrics: W-L record and ave. run support, which B-R conveniently provides. That two-metric measure is really not subject to reduction to a single figure, but fortunately, evolution provides fans of all teams with the ability to hold two pieces of information in mind when making an assessment — that seems good enough to me if no single-metric figure exists. In the case of the question, “How much did good or bad fielding affect Samardzija’s pitching,” we can look at two additional figures: team Rtot and FA (or, a rough equivalent, errors per 9 IP) in Samardzija-pitched innings. That provides all the context we need to assess the impact of team defense — team errors would give a measure allowing us to assess the degree to which Samardzija prevented unearned runs, in addition to ERA’s device of simply excluding them. (But more simply, we could just look at ERA in light of Rtot.)
As for the outcomes of balls-in-play, since I regard an essential part of the pitcher’s job to be denying the batter the chance to make strong contact or to control ball placement, I take success and failure on those features of balls-in-play to be essential to assessing pitching quality. Of course there are instances where the pitcher succeeds, but the batter’s squiggler proves too slow or too tough to handle by the fielder (good or bad as he may be), but we’re now getting closer to flagging those since play-by-play now distinguishes IF and OF hits, records “soft” pop-ups, etc., and Statcast will undoubtedly be aggregating data on velocity off the bat, etc., for individual pitchers, since it already collects that data — these are the essential details that will reveal whether batter intention or pitcher intention proved dominant in the match-up of skills.
So I hope that, ultimately, FIP, which achieves its metric simplicity through massive exclusion of data relevant to pitching quality, rationalized by pointing to an irrational concept of “chance” (= Fate), goes the way of the dodo and is replaced by metrics — perhaps more than one, but still within the range of the human mind to grasp in a single thought — that both isolate pitching-skill quality and incorporate the entire context. (Point of clarification: I mean no disrespect to dodos.)
Yet another long post, but of a different sort. I’ve put a lot of detail into this because it may be that someone more statworthy on HHS will see where the limits of my thinking are, and open my eyes to how to work with stats better (Doom, I’m actually looking at you). In trying to figure out how to vote, I’ve encountered a difficult problem (for me). I would like to place Kenley Jansen fifth, for reasons discussed on this string in response to ThickieDon’s indication that he was thinking along the same lines. That means I have to drop one of the top five starters on my list: Scherzer, Kershaw, Strasburg, Greinke, and Gonzalez. But how do I sort those five so that I’m confident I’m dropping the one who would be #5 if not for Jansen.
Normally, my first resort would be to use WAR. And, indeed, WAR gives a very clear answer: drop Kershaw. The other four are all >6.0, while Kershaw is not only far below, at 4.6, but he’s not even next in line according to WAR, that would be Robbie Ray at 5.0.
My problem is that Kershaw doesn’t look like his WAR should be so much lower than the others’. He was, after all, league leader in ERA and ERA+, not to mention wins and K/BB — and, actually, if you add HBP to WHIP, he turns out to be better even than Scherzer, the league leader. That’s a lot of quality — if it weren’t for the damage he allowed from gopher balls being so astronomical, I’d tend to rate him #1, not #5, 6, or 7, despite his pitching fewer innings than the IP leaders (actually, not so many fewer).
So what is it that makes Kershaw’s WAR drop so low? Looking at his B-R stats, it seems to be a combination of factors that include strength of schedule and park factors, but, most dramatically, quality of defense. I’ll just focus on that last one, because of the scale involved.
For the three Nationals pitchers, the RA9def (quality of defense, with average = 0) ranges from -0.10 to -0.13, and in each case, the figure is a drop-off of about 0.08 from 2016, showing a moderate decline in team defense. For Greinke the figure is positive 0.08, a pretty sharp rise from 2016, when it was a full 0.2 lower. So Greinke essentially gets slightly penalized for pitching with strong defense behind him relative to the Nats guys (which is how the figures should work). But in Kershaw’s case, RA6def is 0.41, absolutely spectacular defense, up even more from 2016’s strong number than in Greinke’s case (0.22).
Where did this figure come from? It couldn’t just be Rtot, since the Dodgers’ Rtot (+19) was good, but nothing spectacular: league average was -8; the Dodgers ranked second, slightly above the Cards (+15), but way behind the Cubs (+36) — Greinke’s Snakes were at -4, creating a little bit of a puzzle, in light of the positive RA9def figure in Greinke’s case. Actually, where the Dodger defense stands out is in “Rdrs” (defensive run saved above average), an umbrella figure calculated by Baseball Info Solutions. For Rdrs, the Dodgers rank #1 by far: runner up to the Dodgers’ 68 is the Cards at 36; the Nats are at -41. This seems to me the clear source of the huge difference in RA9def that seems to have such impact on WAR differences between Kershaw and the other leading candidates.
Now, Rdrs is, for me, a black box: I don’t know how it’s put together. B-R helpfully explains that it includes Rpm, Rbdp, Rbof, and Rbcatch in one tidy figure . . . edifying but not comprehensible. I suppose I could stay up till 3am to become familiar with all these terms and a loyal reader of the Baseball Info Solutions (which prominently lists Bill James as an Associate, so I’m sure it’s a real thing, if Sean Foreman’s implicit endorsement weren’t enough). However, at my age, I think that would produce a lot more fatigue than learning (not much learning left in these cells at the best of times). So I’m left thinking that Kershaw’s low WAR and the implication that he’s the guy I should bump from my list to install Jansen is entirely dependent upon an opaque stat that does not line up with similar stats (like Rtot) that I can at least pretend to understand.
The upshot is that I’m left confused, but gaining deeper sympathy with nsb’s complaint on the last string that, given the convolutions within convolutions that the keepers of WAR sustain but do not translate clearly to non-specialists, “How can we, who are denied the priestly mysteries, accept uncritically the orthodoxy they espouse?” My sense is that the keepers of B-R and FanGraphs would say they don’t mean WAR to be orthodoxy, and in that spirit I think I’m going to suspend it in the Kershaw case and look for someone to bump whom I don’t intuitively feel deserves to be #2 or 3.
Sorry, epm. Didn’t see this post earlier.
There are a few things working against Kershaw. The defense (spectacular), his RA9 is not that great (I mean, it’s great, but it’s his worst since 2012), and his ballparks were immensely favorable to him. I re-ran the numbers from baseball-ref to my limited understanding, and I came up with 28 RAA (bb-ref has 29, so I feel pretty good about that). The math looks like it checks out.
The other thing is, the Dodgers DID have spectacular defense; I don’t know the best way to quantify that, but it doesn’t seem out of line to me that Kershaw, for maybe the first time, wasn’t quite as good as the numbers look. It’s fair to add some healthy skepticism. Though, I would point out, any deductions you want to give Kershaw would also APPLY to Jansen, so perhaps that should be another consideration, too. Just some food for thought.
Thanks, Doom. I appreciate your going to all trouble. I recognize that the figures for things like Rtot and Rdrs show that the Dodgers’ defense was terrific. What I can’t figure out is what those figures are reflecting.
I start with the famously weak traditional fielding stats because they’re easy to understand, even though we know they can be very misleading. The Dodgers played slightly more innings than average: how do they look in traditional stats? They have an ordinary number for PO, but far fewer assists than any other team: 1367 (league average: 1558; next fewest, Washington, with 1434). The Dodgers were a little above middle of the pack on errors, with 88 (league average: 92), but, of course, with fewer total chances that resulted in a league-average fielding percentage. But what about DPs: the Dodgers executed 131 of them? Third fewest, as it turns out: league average was 147. So, really, nothing so far. Traditional fielding stats don’t provide the least hint of why the Dodgers are evaluated so high.
At this point, I wanted to compare the non-traditional stats of range factors vs. league norms for NL teams, but those figures aren’t aggregated, so I took a look at those figures for a few individual Dodgers who had outstanding Rtot numbers: Forsythe, Seager, Puig. Nothing there: overall, each has RF/9 and RF/G levels at or lower than league norms.
But there is one non-traditional stat in which the Dodgers stand out: “Defensive Effeciency” (DefEff). Here they lead the league with a percentage figure of .703 for balls in play converted to outs: league average is .686. Clearly, given the low number of total chances, this figure is connected to the fact that the Dodger pitchers kept balls out of play by striking out a league leading number of batters: 1549, slightly over 200 more than league average (which is a lot!). Dodger pitchers also faced far fewer batters than any other team: 5925 (league average 6190; no other team’s within 140 of Bro- — . . . of LA).
So here’s what I’m seeing. As a team, the Dodgers encountered significant fewer balls in play than other teams, but they converted BIP they did see into outs at a higher rate than any other team. The strikingly small number of BIP was due entirely to the pitching staff, which was so effective at messing up hitters that they disposed of about 5% more of them by Ks than the runner-up team (Arizona), and 15% more than a league average team. When batters did manage to put balls in play, Dodger fielders were able to convert the to outs 1.7% of the time more than a league average team (which is a significant amount).
This last figure appears to me to be the key to understanding why the Dodger defense is rated so highly. But it appears to me that this judgment is based on the assumption that when there were BIP, the outcome of whether they resulted in hits or outs depended 100% on the skill of the fielders. Pitchers are treated as innocent bystanders — the BIP went wherever they chanced to go, and if the fielders were so skilled that they converted them to outs at a high rate, that’s just the pitchers’ good luck.
Now, I’ve written too much already on the fact that I believe that the location and speed of a BIP is, pebbles, wind, and so forth aside, not a matter of chance: it is the outcome of a hitter exercising skill to make good contact and place the ball out of reach and a pitcher exercising skill to use ball speed and movement to frustrate the batter’s goals, resulting, if there is a BIP, in poor contact that reduces the speed of the batted ball and minimizes the role of batter intent in ball placement. Fielders encounter BIP as an outcome of that contest and they take it from there, their greater or lesser skills then making a contribution to the outcome of the play. But the entire history of the BIP, from the start of the play to its end, bears the mark of the contest of skills between pitcher and batter, as well as the skills of the fielders.
So here you have a Dodger pitching staff that has demonstrated its exceptionally strong skills by striking out batters at a rate well beyond any other staff, and B-R analyzes the fact that the other main product of the pitcher/batter contest, BIP, has an equally superior outcome because . . . those pitchers were just darned lucky to have such exceptionally skilled fielders behind them! And because they were such lucky dudes, benefiting from the skills of others while they just stood on the mound as innocent bystanders, the advanced stats of those pitchers are adjusted down so that pitching staffs who give up BIP that fielder’s don’t handle very well are not disadvantaged. The idea that there might be a connection between outstanding pitching skills and outstanding fielding results is not to be entertained, because we all know that between the time the ball reaches the bat and the time it reaches the fielder, everything is up to Chance.
Well, the likelihood that my analysis is on target is probably not very high, because there remain other advanced fielding stats that I really don’t understand. But if the analysis is on target after all, and this is how the defensive component is factored into pitcher WAR, then I don’t buy it.
You are basically 100% on target. I read the “preview” version of your post and was all ready to work out the numbers for defensive efficiency to point out that the traditional numbers DO show the fielders to be outstanding… but then I saw that you noticed that already.
I understand your point about pitchers being able to have some control over balls in play. That’s not how you put it, but that’s the gist of your argument – it’s their fault if the balls don’t go near the fielders, so they should be punished appropriately for it. There are a couple of key counter-arguments.
1. You don’t choose who’s fielding behind you. It DOES matter, and the pitcher has no control over it. If Cecil Fielder is in your outfield, I’m gonna take a guess that you’re going to have a lot of balls dropping for hits, no matter WHO you are.
2. While your argument would seem to have some merit on the face of it, the BABIP conundrum is still at play. You can SAY as much as you’d like that the pitchers should/do have some control or fault, but the numbers do not bear that out, even a little bit. The classic example, I believe, is Pedro Martinez in 1999 and 2000. These are two of the greatest pitching seasons ever – strikeout rates of 13.2 and 11.8, respectively, fewest hits allowed, fewest homers allowed, astronomical ERA+ numbers, eras 3 and 2 runs better than the league… basically unheard of stuff. Yet, in spite of being as good as anyone in history at missing bats, he allowed a .325 BABIP in one of those seasons, and a .227 in the other.
Basically, BABIP has no predictive value; it’s also not a predictABLE value. Sometimes, the best pitcher in the league has a dreadful BABIP, sometimes the worst pitcher has a spectacular one. It’s totally randomly distributed. The one thing that seems to predict it? The quality of the defense behind the pitcher. That’s why DIPS theory (the theory on which FIP is based) doesn’t take any of that into account. So if that leads us to believe that the result of balls in play is totally reliant on defense (there IS an exception for knuckleballers, by the way, because there IS a significant statistical correlation in regard to their batted balls turning into outs), that means, the school of thought goes, that you should be able to utterly disregard what happens to batted balls, or at least adjust for it.
The other problem you have with WAR is that the defense needs credit SOMEHOW. If you put it all on the pitcher (a choice one COULD make; the choice, in fact, that WPA and ERA and a host of other stats make, for example), the defensive WAR for every team would have to be zero. Nobody would be any better than anyone else. I think we can agree that this solution is absolutely wrong; wronger than anything any WAR system does. Fangraphs makes the decision to ignore all such plays. Baseball-reference makes the “compromise” position that we only adjust for defense, but still DO charge the pitcher for what happens to batted balls.
Very strong response, Doom. I find your Pedro example as strong as anecdotal evidence can be in challenging my approach. I’m going to have to stretch to offer even a weak response. Here it comes.
The fact that BAbip is not predictable is certainly true, but that doesn’t mean it’s random, only that we don’t and probably never will understand it’s specific causes (at least on my theory): only when batter or pitcher skill is overwhelmingly dominant will the outcome be easy to assess, because the primary agency is so aligned with the intention of one person. (For such an exception, think of the announcer saying, “He fooled him with a slider there and got him to pop it up! The first baseman makes the easy catch” — an example of a BIP outcome that the fielder’s skill had little to do with because it’s almost entirely the product of the pitcher/batter skill match. A complementary example would be the long fly that the center fielder reaches over the fence to bring back into the park, an out for which the pitcher deserves virtually zero credit and the fielder about 100% [with the batter justly cursing his bad luck that a less circus-worthy fielder wasn’t playing center that day, as the crowd goes wild]. Statistically, however, the two fielders receive equal credit.)
Now, we don’t know the reasons behind Pedro’s change from ’99 to ’00, but there is one factor it seems likely that we can rule out. Since there were minimal changes in the personnel making up the defense behind Pedro over those two seasons, there is no more reason to assign the improvement to changes in the skilled performance of the fielders than to any dramatic change in Pedro’s pitching. This doesn’t in any way mean that fielding quality doesn’t matter — I hope I didn’t imply I was making that absurd claim! It seems to me an axiom of terminology that better fielders will improve a pitcher’s BAbip over worse fielders. However, since you can’t run history twice, there is no way to demonstrate that any particular season BAbip result is due to better or worse fielders, except to the degree that you can isolate and evaluate individual plays in a qualitative, visual assessment frequently enough to establish that particular fielders conformed to our model of what “good fielder” means. Even RF/9 can’t give you a direct measure of fielding quality, because although we assume that lots of chances means greater mobility and skill, that may not be the case in individual instances, such as the first baseman who fields a lot of pop-ups (hmmm . . . I don’t mean to say that long-term, consistent fielding stats for a player can’t provide strong evidence of the player’s base quality, just that the evidence could and sometimes may be misleading).
So I’d have to claim — knowing that it’s not as persuasive an assertion as your Pedro contrast was in challenging my post — that we should not say that Pedro ’00 (like Kershaw ’17) was “the beneficiary of a great defense,” but rather that he was the beneficiary of a great BAbip, to which his fielders surely contributed to an unknown degree — a degree we can’t know from stats, but that we might actually get close to knowing if we watched every play on video and subjectively assigned “degrees of difficulty.” (Obviously, this will be something Statcast, with its degree-of-difficulty replay function, may handle for us in the future through aggregations, assuming that the governing computer program aligns with our subjective judgments and is not, for the sake of quantifiability, substituting non-intuitive measures of quality that don’t actually realize what we mean by “good fielding.”)
— A closing tangent: Not sure what it means, but there’s an interesting, Kershaw2016/17-like effect in Pedro’s pitching from ’98 to ’01, where high/low BAbip correlates strongly and counterintuitively with low/high HR/9 (this inverse correlation holds more mildly for ’02 and ’03 as well):
______BAbip___HR/9
’98___.272____1.00
’99___.325____0.38
’00___.227____0.71
’01___.310____0.38
Since the BIP/HR divide is actually completely arbitrary in terms of the isolated pitcher/batter match-up (it’s not as if either one has the skill to measure the dividing line between the warning track and bleachers as he throws or swings), this pattern somewhat buffers the contrast between Pedro ’99 and Pedro ’00, measured by BAbip alone.
Just a couple of kibiztzes—not a word, but what the heck.
1. The Washington pitching staff benefitted from a .698 DefEar, or whatever it is, roughly 99.3 % of the .703 enjoyed by LA. So—I’m not sure it isn’t that old bogeyman Park Effect that’s really raising its ugly head here.
2) Let’s try ‘Chance’ instead of ‘Luck’. Joseph Conrad wrote a novel on the subject—how people come together and interact through apparent chance as dictated by various forces. Chance is what it means—something that provides an opportunity. Luck implies randomness. They’re called chances for a reason.
Time to vote, but first some observations:
The candidates in this year’s senior circuit CY running—as someone else, maybe more than someone, has already noted—aren’t particularly inspiring of speechless awe, despite some gaudy stats.The three pWAR leaders all come from the same team, Sherzer, Strasburg, and Gonzalez. Trouble there is, they played in the weakest division by far in either league, finishing 20 games ahead of the second place Marlins who ended up 77-85. In fact, the Nats only played 42 games against teams on the plus side of .500, this with a modest 23-19 advantage, leaving the other 120 games against losers of all shapes and sizes. Here they finished with a good but hardly spectacular winning record of .617.
The result of this imbalance, at any rate, gives Sherzer 8 starts out of 31 vs the tougher opposition, Strasburg 5 out of 28, Gonzalez 8 out of 32. They did all pitch well against winners, but in only 21 starts vs 70 vs under-.500 teams. Weakness of schedule, frankly, gives me, if no one else, the impression that their records are soft and squishy.
Kershaw. Unlike epm, I have no apprehensions about his relatively unspectacular WAR, because I only look at WAR as one measure of many.
Some of the things I do look at and find exceptional even for a great pitcher on a very good team: Kershaw started 27 games going 18-4. The Dodgers, in those games went 23-4, suggesting that even when Kershaw was pulled for some reason, he left the team in a good position. In contrast to the Nats pitchers discussed above, Kershaw made 13 of his 27 starts against winning teams, going 7-2, the Dodgers going 11-2. Another point that I find remarkable: Kershaw’s W-L in games when he was supported with 2 runs or less was 4-2. I’ve checked a fair number of pitchers over the years on this subject, and I don’t remember seeing a .667 winning percentage before. Bob Gibson went 9-8 in 1968. Koufax went 7-5 in 1965. I might have missed others who did better, but that doesn’t make Kershaw’s dominance in 2017 less remarkable. Too many HRS? The home runs allowed argument concerning Kershaw somehow fails to bother me, especially since 10 of those dingers came in three starts. When he was off he grooved a few, but how often was he off?
Next: the forgotten man, Kershaw’s teammate Alex Wood. Wood didn’t begin the season in the Dodger rotation, but when he was elevated to it he excelled. From April 26 through July 15 he started 12 times and allowed a run or less in ten of those starts. He won ten times, the Dodger 11. He cooled off toward the end of the season, true, but so, for instance, did Sale for the Red Sox, and that didn’t keep Sale from garnering lots of support for the AL CY. With only 152.1 IP, Wood didn’t make the official ERA list, but his 2.72 would have ranked fourth with a few more innings.
Of the rest mentioned by Dr. Doom or someone else, I personally don’t find much to recommend one over any of the others.
My Vote:
1) Kershaw
2) Scherzer
3) Strasburg
4) Ray—better than Greinke, just fewer innings
5) Wood
Time to get the ball rolling on voting & since I went last on the AL vote (I think), I’ll go first this time (unless I overlooked someone):
1) Scherzer I’ll admit that his being an ex-Tiger & a seemingly good guy was a factor but it was the additional 25 IP’s that put him over the top.
2) Kershaw Close call over Strasburg, no one deciding factor, but I was just a tiny bit more impressed by his numbers
3) Strasburg
4) Greinke An easier call over Gonzålez than I expected, mostly because of WHIP and SO/BB.
5) Gonzålez Not entirely sold on the FIP argument at least in part because it’s a reversal of his past 5 years history even tho he had a better WHIP than in all but 1 of them. But he did walk a bunch, plus see #4
Can’t bring myself to vote for a guy with fewer than 70 IP’s altho if the current trend in starters continues pretty soon ain’t gonna be nobody with 200+ IP’s any more
Couldn’t bring myself to vote for Robbie Ray even tho he’s got a great beard because I’m still pissed about the Tigers trading him away.
Damned near voted for Samardzija (actually figured out the FIP & SO/BB argument on my own pretty quickly, which surprised me) just because he led the league in IP’s and that’s another trend that’s really pissing me off.
Paul voted in a comment, I believe.
I was so captivated by horses dying of cholera on HHS that I missed Paul’s vote too.
Next election I talk about the Blizzard of 1888.
Yes. Then we get to the early stages of the expansion era. Talk about progress, the only thing “Found On Road Dead” was my father’s ’67 Galaxy. From there, road travel improved for us in the form of an 8 cylinder, 283 cubic inch Chevy that had a valve tap that sounded like squirrels arguing in an animated cartoon
Here’s my vote:
1) Max
2) Kershaw
3) Strasburg
4) Gonzalez
5) Jansen
I’ve already explained the Jansen vote, both in my MVP vote and in reply to ThickieD’s suggestion. The issue for me was choosing a #5 starter to bump so Jansen had a seat, and the candidates were Gonzalez and Greinke, whom I think were of almost identical value. It really wasn’t a matter of designating one as better than Jansen and the other not: I’ve included Jansen as a statement about trying to give outstanding excellence in a designated role its due.
Gonzalez’s weakness was his high BB count — it’s what sunk him below Greinke for Hartvig. After poring over various “advanced stats” comparing the two, I wound up going for Gio because, over more batters faced, he gave up 38 fewer TB than Greinke, which seemed to me largely to compensate for the extra 29 BB (discounting IBB). Batters simply fared more poorly against Gonzalez than against Greinke in almost every category, and this was reinforced by a significant gap in runs allowed per 9 IP: 3.1 vs. 3.6, despite Greinke having what B-R rated a much superior defense behind him.
(I appreciate nsb’s adventurousness in plunking for Ray and, particularly, for Wood, but I can’t quite get there.)
Here’s my vote:
1. Scherzer
2. Strasberg
3. Kershaw
4. Gonzalez
5. Ray
Reminder and vote.
Reminder first. It’s your last day to vote, everyone! Don’t forget!
Here’s my vote:
1) Max Scherzer
2) Clayton Kershaw
3) Stephen Strasburg
4) Zack Greinke
5) Kenley Jansen
Razor-thin for the three short of Scherzer, in my opinion. Scherzer is the top by enough of a margin that I don’t have any problem picking him. Kershaw’s standard numbers are better than Strasburg’s; Strasburg probably had better advanced numbers, and better numbers when you give advanced and traditional equal weight; yet, I had to put Kershaw up there. His standard numbers are too good.
I threw Kenley Jansen my fifth-place vote, simply because I’m not sure how to value leverage. How important were his innings? If they were twice as important as the average starters, he should probably be in the top-two, maybe even number one. It’s just a matter of how you do these things. I put him fifth because I don’t think that’s right, but I’m not sure which direction to put him – lower or higher. Fifth is the best compromise I can come up with.
1. Corey Knebel
2. Kenley Jansen
3. Felipe Rivero
4. Max Scherzer
5. Clayton Kershaw
1. Scherzer
2. Strasburg
3. Kershaw
4. Ray
5. Gonzalez
Here are your results for the HHS 2017 NL Cy Young, with scores next to names and first place votes in parentheses. Tiebreaker procedure is explained below:
1) Max Scherzer, 48 (6)
2) Clayton Kershaw, 29 (1)
3) Stephen Strasburg, 24
4) Kenley Jansen, 7
5) Robbie Ray, 7
6) Corey Knebel, 7 (1)
7) Gio Gonzalez, 6
8) Zack Greinke, 4
9) Felipe Romero, 3
10) Alex Wood, 1
Tiebreakers go as follows: total ballots first, then highest ballot position. Had there been a tie thereafter, we would’ve removed the last ballot which created the tie.
Paul E was the only person who named all five top vote-getters on his ballot. Seven of the eight ballots had either Ray or Jansen, six of the seven had only one or the other. Beyond that, I think it’s safe to say that Voomo had the most, uh, non-traditional ballot. 🙂 I’d love to hear some of his rationale for his picks. If you’re around Voomo, I’ll check the thread for a few days in case you want to let us know more.
Anyway, Scherzer coasted to victory. It wasn’t quite as thorough as the victory Corey Kluber had in the AL vote, but it was pretty thoroughly dominant. There’s not too much commentary to add to this one, given the amount of discussion left in this thread. Anyway, that’s three posts down, one to go. Hope to see you all around soon for the final vote, the AL MVP!
I’m going to go out on a limb here and predict that Miguel Cabrera will get exactly the same amount of support from us as he did from the BBWAA. Beyond that I haven’t a clue.
Sure. I plugged Scherzer ahead of Kershaw because this is an award named after Cy Young.
So, all other things being nearly equal, I went with the guy who didn’t miss any starts.
Knebel got zero other votes, both in this forum, and in the ‘official’ vote.
I think that is an oversight.
Were this not an open-ballot, I probably wouldn’t vote him number one, but I wanted to bring attention to his effort.
He did blow two saves on Sep 20th and 22nd, which might have cost the Beer-Guys a playoff spot. Oops.
Before that crappy last week and half he gave up 10 runs all year, and were a big reason why Wisconsin was in the hunt.
This award is not just to recognize a Denton True dopple, it can also be seen as a Pitching MVP. And here’s an up-and-coming team, and a young pitcher who has never succeeded in that role. I do think that ‘closers’ are overrated, but regardless of role, he was the most effective reliever in the league.
And he did not have an ‘easy’ closer job. His Game-Entering Leverage Index was 1.93
(the other Save leaders:)
1.88 … Rodney
1.79 … Holland
1.69 … Jansen
1.62 … Wade Davis
…
And yes, a vote for Felipe Rivero seems like a goof.
Really, I just had no strong impulse to vote for anybody as the clear cut winner, so I did not want to influence the results at all.
But again, Rivero was amazing, and I wanted to draw attention to him
(certainly he is unknown, Dr. Doom didn’t even spell his name correctly).
He earned the closer job by giving up 2 ER in the first 2.5 months, and his season-long numbers in not giving up hits, HRs, or walks was elite.
Congratulations to Aaron Boone, whom we all know from this game-winning home run:
https://www.baseball-reference.com/boxes/CIN/CIN199809270.shtml
You know, the game where two sets of brothers played on the same infield.
Wow ! Larkin’s only ML game !
Do you really believe Aaron Boone will do as well as Girardi did this past season? I mean, I’m not a Girardi fan (or NYY) and it would have been cool to see him get a lifetime contract and have to wear that #28 for the next 15 years, but I honestly don’t get it…..
How much money are they saving on that move?
I’m with Paul. I don’t recall seeing a managerial change that made less sense to me. I’m also like Paul in being no Yankee fan, but I do think highly of Girardi.
Sorry, when I stated “I’m not a Girardi fan”, I meant I wasn’t a guy who was going to cheer his accomplishments. But, yes, he certainly did accomplish much with the NYY in 2017. I was, actually, shocked that they played so well – even if their W-L record doesn’t much their Pythagorus’ estimation.
Supposedly, he was a little tough on the young guys? Maybe they’ll appreciate him when they’re watching the AL playoffs on TV next October?
My suspicion is that Cashman, who supposedly never liked Girardi, felt that “If we can finish 10 games under our Pythagorean with you, we can do that without you.” I suspect that he believes that have the talent to win, irrespective of management.
I’ll miss Boone as a broadcaster: insightful and easy to listen to. By my count, Boone is the 7th Yankee manager, excluding player managers, with no prior professional managing experience; Lou Piniella is the only one of the previous six (the others are Bucky Dent, Gene Michael, Dick Howser, Yogi Berra and Bob Shawkey) to last more than one season in the Bronx (you could perhaps add Bill Donovan, who followed two seasons as player/manager with just one as manager only).
Dent and Piniella were two of six Yankee managers in a 5 year period (1988-92), a group that also included Buck Showalter and Stump Merrill among those making their major league managing debuts. Howser and Michael were similarly part of another 6 managers in 5 years period in 1980-84.
That Larkin/Boone brothers game was played the same day as Roy Halladay’s second major league start and first complete game and win; it was an almost no-hitter broken up by Bobby Higginson’s pinch-home run with two out in the 9th.
For those who are wondering where the AL MVP post is, I did send it to Doug, and I’m sure he’ll get it up as soon as he’s able. I’m looking forward to the discussion!