There was a tie vote in the Circle of Greats 1977 Balloting, so we will have a run-off election between the two tied players, Carlos Beltran and Roy Halladay. More after the jump.
| Seasons | G | PA | AB | R | H | 2B | 3B | HR | RBI | SB | CS | BB | SO | BA | OBP | SLG | OPS | OPS+ | TB | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Carlos Beltran | 1998-2017 | 2586 | 11031 | 9768 | 1582 | 2725 | 565 | 78 | 435 | 1587 | 312 | 49 | 1084 | 1795 | .279 | .350 | .486 | .837 | 119 | 4751 |
| per 162 games | 162 | 691 | 612 | 99 | 171 | 35 | 5 | 27 | 99 | 20 | 3 | 68 | 112 | .279 | .350 | .486 | .837 | 119 | 298 |
| Seasons | W | L | W-L% | ERA | G | GS | CG | SHO | IP | BB | SO | ERA+ | FIP | WHIP | H9 | HR9 | BB9 | SO9 | SO/W | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Roy Halladay | 1998-2013 | 203 | 105 | .659 | 3.38 | 416 | 390 | 67 | 20 | 2749.1 | 592 | 2117 | 131 | 3.39 | 1.178 | 8.7 | 0.8 | 1.9 | 6.9 | 3.58 |
| per 162 games | 17 | 9 | .659 | 3.38 | 35 | 33 | 6 | 2 | 232 | 50 | 179 | 131 | 3.39 | 1.178 | 8.7 | 0.8 | 1.9 | 6.9 | 3.58 |
Comparing a pitcher to an everyday player is an interesting challenge. Here’s a run down on some of the things that can be compared directly:
| Beltran | Halladay | |
| Career WAR (B-R/FanGraphs) | 70.1/67.9 | 65.4/65.4 |
| Career WAA (B-R) | 34.4 | 40.4 |
| Career WPA (B-R/FanGraphs) | 35.8/35.4 | 38.0/38.0 |
| RoY Award | Yes | No |
| MVP/CYA Awards | none | 2 times |
| Gold Glove Awards | 3 times | none |
| All-Star Selections | 9 times | 8 times |
| Pennant-winning teams | 2 times | none |
| World Series championships | 1 time | none |
| Hall of Fame selection | not yet eligible | Yes |
Here’s a look at the their Top 50 career rankings in the live ball era.
Beltran: G 39th, PA 30th, H 45th, R 36th, HR 44th, 2B 23rd, XBH 23rd, RBI 37th, SB 38th
Halladay: WAR 31st, WAA 19th, WPA 15th, ERA+ 12th, BB% 18th, W-L% 7th
So, the choice is yours. However you decide, your ballot in this runoff round, unlike the usual three-name ballot, should identify only the one candidate you prefer (you will also need to add at least a little bit of extra verbiage though, because the WordPress engine that supports the site won’t accept comments of only one or two words).
All votes must be in by 11:59PM EST on Sunday night, February 13th, with vote changes allowed until 11:59PM EST on Friday night, February 11th. If the result of this runoff is still a tie, the last vote cast will be discarded to determine the winner. So, vote early to ensure your vote counts! If you would like to keep track of the vote tally for the runoff, you can check this tally spreadsheet: COG 1977 Runoff Vote Tally.
Halladay seems like an easy choice to me. His top 50 ranks include 3 value metrics and all are 31st and higher, including 5 of 6 in top 20.
For Beltran, none of his top 50s make the top 20 and all are in counting stats (dare I say “compiler”, though a very good one, to be sure).
Agree with this, Halladay is the better choice, even without Beltran’s issue. I didn’t vote first time out, so I’ll stay out of the run off
My vote is for Carlos Beltran.
Thanks!
Black ink:
Roy, 48
Carlos, 1 (once played 162 games)
Which of these guys was in the discussion as the best player at his position in his era?
I vote for Halladay
(Forgiving his 2000 season)
VOOM,
“Which of these guys was in the discussion as the best player at his position in his era?”
My ‘notes’ from b-ref may be incorrect but, as far as I can tell Beltran ranked among CF’ers:
2001 AL 1st 6.5 WAR
2003 AL 1st 5.8 WAR
2004 MLB 2nd 6.8 WAR (Royals and Astros; Edmonds 1st w/7.2 WAR)
2006 NL 1st 8.2 WAR
2007 NL 1st 5.4 WAR
2008 NL 1st 7.0 WAR
2009 3.6 WAR in 81 games; microfracture surgery. Moved to RF
2011 4.5 WAR is 3rd among NL RF’ers
This coupled with the fact that Beltran absolutely hit the cover off the ball in the post-season…….I don’t believe it’s such an open and shut case between him and Halladay
Sorry, Paul; I didn’t see that you were making the same point! I’m using this as a distraction from a Zoom call, and I was composing slowly over time. Didn’t realize you had already said this!
Doom,
Yes, Beltran was a plus fielder, one of the greatest percentage base-stealers of all-time, switch-it with power, took a walk, and was, at one point well into his career, #1 in OPS amongst post-season hitters.
Was he a career 150 OPS+ hitter like Speaker, Cobb, DiMaggio, Griffey, and Snider? No, but that is a pretty short list.
I believe he was the superior of Andruw Jones and, when comparing peaks, a pretty similar case can be made:
Edmonds 1995-2005 1475G 56.1WAR
Beltran 2001-2011 1500G 55.8WAR
Certainly, Edmonds couldn’t run and steal bases like Beltran and that might give him the edge over Edmonds as well?
Good points. I hadn’t looked at it quite that carefully
Just out of curiosity: was there a better centerfielder over the first half of Beltran’s career than Beltran himself? Maybe Andruw Jones… but it’s one of the two – and I don’t think that was unknown at the time. I would say he was firmly in the discussion. Limiting it to his best decade (1999-2008), the only outfielder (regardless of particular position) who was better was Barry Bonds. The discussion would’ve been Andruw Jones and Carlos Beltran. Maybe Vlad Guerrero, in popular media? But WAR certainly doesn’t think he was in these guys’ league. So I think he was in that conversation as among the best at his position. Bottom line, though: I still think you’re right, and Halladay’s the right guy.
Andruw Jones completely collapsed in 2008: .158, 3, 14 in 75 games. 35 OPS+ -1.6 WAR. -2.3 WAA
My vote is for Beltran.
Halladay. Lest we forget Beltran’s leading role in the 2017 Astro cheating scandal: https://www.si.com/mlb/2020/02/11/carlos-beltran-astros-sign-stealing-scandal-role
One player was named in the report. That player was Carlos Beltran. MLB threw him under the bus to try to quiet it down and get everyone to move on. But Beltran and Cora were named as the ringleaders, and identified by other players as such.
From the article: “Beltrán, who is the godfather of the whole program, ironically just swings at everything after taking a strike and probably does the worst with the info,” he wrote.
My vote is for Roy Halladay.
But before I make the case for Halladay, I want to try to understand the case for Beltran. I believe it comes down to a few points.
1.) Beltran was really good at a lot of things – he didn’t have a lot of blank ink, because he wasn’t excellent at one thing. Players with high-level skill, but not elite skill at any one thing – will be hurt by checking league leaders and such.
2.) The things Beltran was excellent at don’t translate to a big, traditional, statistical case. He was a plus fielder – but that’s not something that wins you black ink. He’s the best percentage base-stealer in history… but not a high-volume base-stealer, and that’s what people count.
3.) Beltran is possibly the greatest postseason hitter in history; one at-bat with the Mets shouldn’t color one’s opinion of him, even though many Mets fans will never talk about anything else.
Those are powerful cases for Beltran. I fully support him for the COG… just not yet. And that’s because we have Roy Halladay on the ballot.
First, a fun fact: on April 13, 2007, Roy Halladay threw the most recent 10-inning CG in MLB, defeating the Tigers 2-1. Who knows when we’ll ever see another. But that’s not a COG case in and of itself. Still, we get to remember him as a bit of a throwback.
Anyway, since Halladay and Beltran have a similar career total WAR (within the margin of error). But since I pointed out some things that cause Beltran to be underrated, I want to do the same for Halladay.
1.) The Dave Stieb factor: One of Halladay’s biggest barriers, I think, is that he was the top pitcher between eras of great pitching. Just as Stieb was the best pitcher in the middle era between the innings-eaters of the ’70s and the most effective pitchers of all-time in the ’90s-00s, Halladay fell between that latter group and the next stars: the Kershaw-Verlander-Scherzer group. Via Fangraphs (no Stathead subscription, and theirs is more searchable), in the decade 2002-11, Halladay was nearly 10 WAR ahead of the next pitcher (Sabathia).
2.) The Kevin Brown factor: One of the things we struggle with in the COG is the “less-obvious” candidates. Like, yeah – if you’re Randy Johnson or Tom Seaver, you’re getting in. But the fact of the matter is, we’re electing more than 25-30 pitchers here. And we really struggle to know what that group looks like. I think we are all biased toward those pitchers with both the mega-peak and the extreme longevity. But when you don’t have the “obvious” career shape, particularly for pitchers, I think we need to look at what our actual qualifications might be. Even the midpoint for the COG is ridiculous: Mike Mussina and Fergie Jenkins (by WAR, anyway). But that’s not who we should compare Halladay to: to me, the comparisons are Juan Marichal, Rube Waddell, Dennis Eckersley, John Smoltz, Kevin Brown, Dazzy Vance, Whitey Ford, Ed Walsh. I think Halladay fits very well among those groups.
3.) The Greg Maddux factor: Halladay was also part of what I personally believe to have been probably the greatest one-year pitching staff of all-time: the 2011 Phillies. Not only was he on that staff: he was its ace. Best pitcher on the best staff is… pretty darn cool. Not a COG qualifier by itself, but I think it really separates him from some other, similarly-qualified candidates.
4.) The Luis Tiant factor: Twin peaks – not the show – make a narrative on a player harder. (It may also be a literal disadvantage for his team(s)… but that’s another conversation.) Halladay was great in ’02-’03. From ’04-’07, he was a very good pitcher (two top-five Cy Youngs in that time, actually), but not a legendary one. If his career had a normal trajectory continuing after ’07, this wouldn’t even be a discussion. But what he did from ’08-’11 puts him in this conversation. Still, if those six seasons (’02-’03 and ’08-’11) had somehow been consecutive, we would have a very different narrative. Unfortunately, they weren’t. But had those seasons been consecutive, they would’ve created an extremely powerful short-term best narrative, that could make him more-easily elected.
5.) The Don Larsen factor: As the name of this one indicates, you’re not COG-material just because of one playoff game… but this dude threw a no-hitter in the playoffs! We shouldn’t forget that, particularly when we will/should undoubtedly give credit to Beltran for his playoff performance. Halladay had a 2.37 ERA and a 0.737 WHIP in the playoffs. That’s not far off from Pedro 2000 – only it happened in the playoffs. However…
6.) The Fergie Jenkins factor: Halladay had only 5 postseason starts in his career. Yes, the Phillies underperformed in ’10 and ’11. But I don’t think you can blame Halladay for that: his two postseason losses were a game 1 loss in ’10: 7 innings of 4-run ball. And his team had six games to recover! They just didn’t (even though they won his other start). And the ’11 loss is worse: 8-innings, 6 hits, 1-run… and a 1-0 loss to Chris Carpenter and the Cards. The main reason he doesn’t have great postseason stats is that (like the guy I named here) he didn’t get there very often. And that’s mostly because he pitched for the Jays when they were getting beat up by the Yanks and Sawx. They finished 3rd in the division every year he pitched there except for three, and only once did they even finish second. But even in that year (2006), their 87 wins were only 6th-best in the AL (which, back then at least, didn’t get you into the playoffs). Had they played a bit more manageable schedule – in, say, the West or particularly the Central – he’d probably have MORE postseason heroics to add to his exploits. You can’t downgrade him because he happened to play on a bad team, though.
In short, I think that, when you consider these factors in addition to Halladay’s excellent statistical case, he become a no-brainer for induction, even against as formidable a competitor as Beltran. My personal method says that Halladay’s peak plays out more like a guy with 83 WAR than 65. (It has Beltran and Andruw Jones both around 81. Not a huge difference, but we’re all splitting hairs at this point.
Doom, are you saying the 2011 Phillies had the best staff of all time?
You qualified it by saying the best one year staff. Not sure what the distinction is.
If you’re saying that Halladay, Lee, Hamels, Oswalt, Worley were the best staff of all time, that’s certainly worthy of an argument against.
However, I’m struggling to come up with a staff that was better, either in performance or pedigree.
1906 Cubs, perhaps.
FWIW, The 1906 Cubs are the only staff with an ERA+ greater than 150 in the 60’6″ pitching distance era. I can’t seem to figure out team pitching WAR on Stathead 🙁
Paul E:You can find team pitching WAR on Fangraphs.com.
What I meant by “one year staff” was that I’m not trying to compare it to the Maddux-Smoltz-Glavine Braves. Those are much stronger staffs, over a period of years. But I don’t believe they had any one year better than the ’11 Phillies. I would put that Phils staff up against any single year by any single staff in baseball history, but particularly post-integration. They don’t get the credit because of the perception that they underachieved – World champs in ’08, NL champs in ’09, lost NLCS in ’10, lost division series in ’11 is going backward, so it seems the staff didn’t help. But really, the offense was just getting worse as the pitching improved. They were down more than 60 runs scored from ’10 or ’08, and more than 100 from ’09… but they allowed only 529 runs! It was a problem of the pitching overcoming bad hitting, and the hitting just not being enough in the postseason. But yeah; I’ll take them over just about anyone.
Lemon, Wynn, Feller, Garcia? Palmer, McNally, Cuellar, Dobson?
Cy Young, George xxx Cuppy, and Clarkson? Kid Nichols, Clarkson, and anybody?
Mathewson , McGinity, and Wiltse? 1923 Cincinnati Reds?
Not sure those deadball staffs are particularly comparable – most especially the 19th century ones you mention. But we could look at the ’54 Indians and the ’71 O’s, assuming those are the one-year staffs you meant. Now, given that this is a bit of an apples-to-oranges, since starting rotations were managed really differently. But the ’71 O’s had four pitchers start 142 games; the ’54 Indians had five pitchers start 145; and the ’11 Phillies had five pitchers start 139. So while the rotations weren’t the same size, nor were the schedules, the final number of games is pretty close.
The ’54 Indians’ five starters with Innings Pitched & ERA+ were:
Early Wynn (270.2; 135)
Mike Garcia (258.2; 140)
Bob Lemon (258.1; 136)
Art Houtteman (188.0; 110)
Bob Feller (140.0; 120)
189.1 IP remaining for bullpen (see below)
The ’71 Orioles:
Mike Cuellar (292.1; 109)
Pat Dobson (282.1; 116)
Jim Palmer (282.0; 126)
Dave McNally (224.1; 117)
197 IP remaining for bullpen
The ’11 Phillies:
Roy Halladay (233.2; 163)
Cliff Lee (232.2; 160)
Cole Hamels (216.0; 137)
Roy Oswalt (139.0; 104)
Vance Worley (131.2; 127)
278 IP remaining for bullpen
Now, we’re going to assume three things: a historically normal league ERA of 3.60; 9 IP/TeamGame; a schedule = GS by these rotations. That is, we’re going to pretend that the ’54 Indians played a 145-game schedule, the ’71 O’s played a 142-game schedule, and the ’11 Phillies played a 139-game schedule. At 9 IP/team game, there are 189.1 uncovered innings by the ’54 Indians; 197.0 by the ’71 O’s, and 278 by the ’11 Phils. We will need this for later. And we’re going to pretend that each bullpen is staffed with pitchers who will fill unused innings with a 100 ERA+. That’s actually a little unfair to these teams; the O’s are closest to that, but the Phillies and Indians had better bullpens. It’s a worthwhile starting place, though, I think. Finally, I have no idea if 3.60 is a historically normal era; but the three teams played in leagues with ERAs of 3.72, 3.46, and 3.81, respectively; the average is 3.66, but I wanted an easier number to work with, so you get 3.60.
I’ll walk through the results: Early Wynn (since he’s first on the list). Wynn pitched 270 and 2/3 innings at a 135 ERA+. In a league with a 3.60 ERA, his ERA would’ve been 2.67 and he would’ve allowed 80.19 runs in those innings. Mike Garcia would allow 73.9, Lemon 75.98, Houtteman 68.36, and Feller 46.67. The imaginary ’54 Indians bullpen would’ve allowed 75.73 runs. Add it all up, and you get 420.85 runs – a 2.90 full staff ERA (bullpen included).
For the ’71 O’s, you get 107.28 runs from Cuellar, Dobson allows 97.36, Palmer 89.52, McNally 76.7, and 78.8 from the ‘pen. That’s a total of 449.65, or a team ERA of 3.17.
Finally, the ’11 Phillies give you only 57.34 runs from Halladay, 58.17 from Lee, 63.06 from Hamels, 53.46 from Oswalt, and 41.47 from Worley. Of course, they have far and away the fewest innings pitched, which means leaving the most runs for the bullpen. Their bullpen, in this scenario, allows 111.2 runs. But the grand total is still only 384.71 – an era of 2.77, which is the best of the group.
So yeah; I’ll stick with the ’11 Phillies. I’m not positive, but I’m fairly sure that using FIP would skew the results more in the favor of the Phillies.
One more, for fun: the 1997 Braves (Smoltz, Glavine, Neagle, Maddux) would have a 135-game schedule, or 1215 IP.
John Smoltz (256.0; 138)
Tom Glavine (240.0; 141)
Denny Neagle (233.1; 140)
Greg Maddux (232.2; 189)
Bullpen gets 253 innings.
Those totals are: Smoltz at 74.2, Glavine at 68.09, Neagle at 66.67, and Maddux at 49.24; then add in 101.2 for the bullpen and you get 359.4 in 1215 innings for an era of 2.66 – slightly better than the ’11 Phillies, but playing in front of an all-time defense. Again, I think it could make sense to lean on FIP here, and I feel confident the Phillies would come out ahead.
So yeah: my answer is that the ’11 Phillies are the better of those more “classic” staffs, even accounting era and increased bullpen innings. The ’97 Braves might have a better argument, but it’s almost certainly one of those two or the other.
Doom,
Thanks for the calculations. When I threw out these suggestions, I was only referring to the “talent” on the staffs in question. Like, three Hall of Famers on the ’54 Indians in Feller, Lemon, and Wynn….or, Nichols and Clarkson, who were pitching 80% of the innings for their staffs or Mathewson, McGinnity and whomever. I don’t necessarily believe that we should throw out the pre-integration/dead ball record book either but, that’s another story. I get the ERA comparisons, but what might WAR indicate for these same staffs?
As far as the Phila starting five, there’s one Hall of Famer and I seriously doubt the viability of Hamels’ (but, per WAR, he’s a lock?), Oswalt’s, or Lee’s HoF candidacy. Atlanta has probably the greatest non-steroided pitcher of the last 35 years, a consistent winner in Glavine and an absolute talent in Smoltz….but, for a single season, maybe WAR will prove out who is best? Like the 1973 Orioles have more fWAR or fielding runs than any other team (i think 🙁 ). Maybe there’s a staff that stands out by WAR…..and, I believe Bill James may have done a Win Shares comparison on staffs?
If it’s purely a question of the staff with the most raw talent, I would nominate the 1987 Twins: HOF Steve Carlton, HOF Bert Blyleven, 200-game-winner Joe Niekro, Cy Young-winner Frank Viola, 4-time all-star (and one-time saves leader) Jeff Reardon, plus Juan Berenguer and Allan Anderson. But that, of course, is not exactly what I was talking about, because Carlton and Niekro were worthless, and Blyleven past his prime, by the time it got to be ’87.
As for looking at WAR, I have some issues with doing that. Because the baseline is not zero, you can end up with really weird stuff, where one guy’s really outstanding performance can really skew things. It gets to how Bill James did those comparisons (though where, I can’t remember; I checked the NBJHBA last night, and couldn’t find the entry). He did something else, which was to take the top-four pitchers of a staff via Win Shares, then give you points for the top starter, double-points for the second starter, triple for the third, and quadruple for the fourth. It’s an innovative way to bring balance to a staff.
Personally, I like my method because it plays to the idea of valuing the starters for their actual contributions, weighted not by perceived value relative to a non-zero baseline, but relative to actual, expected runs, weighted by playing time, not arbitrarily by where they rank on their own staff, as per James. But bWAR says the ’97 Braves starting four were worth 22.6 WAR, and that the ’11 Phils starting five was worth 29.1 WAR. Subtracting one starter from the Phils (Oswalt’s 2.1) or adding one to the Braves (Kevin Millwood’s 0.2) is not going to change that conclusion.
I did also go back and check my method for each staff James ranked as the best of its decade (except the ’40s; he said the best staff was “Cardinals 1942-46,” which is hard because A.) they seriously messed around with starters and relievers in ways that make these calculations difficult, and B.) he was completely incorrect: the best rotation of the ’40s was pretty clearly the 1946 Tigers). Also, I’m not including 19th century teams as a matter of principle, and the fact that they really don’t compare based on pitcher workloads and mound distance. I mean, we’re not really asking if you’d rather have the ’11 Phillies or Old Hoss Radbourn pitching underhand from 40 feet; it’s not a reasonable comparison. Then I added in a few of my own choices from this millenium. Here are the rankings of the staffs I chose, (# of starters included), and estimated ERA:
1) 1906 Cubs (6), 2.23
2) 1913 Giants (4), 2.65
3) 1997 Braves (4), 2.66
4) 1926 A’s (6), 2.71
5) 1923 Reds (4), 2.727
6) 1931 A’s (7), 2.734
7) 2011 Phillies (5), 2.77
8) 2019 Astros (5), 2.80 (I included Greinke, even as a midseason-addition)
9) 1946 Tigers (6), 2.85
10) 1954 Indians (5), 2.90
11) 2015 Dodgers (6), 3.04
12) 2002 Diamondbacks (5), 3.07
13) 1966 Dodgers (4), 3.09
14) 2013 Tigers (5), 3.12
15) 1985 Royals (5), 3.15
16) 1971 Orioles (4), 3.17
Per this method, the 1906 Cubs obviously blow everyone out of the water. The gap between them and second place is larger than the gap between the Phillies in 7th and the last-place ’71 Orioles. But really, any of these teams in the 2-10 range is pretty equal: we’re talking about 0.15 runs separating that entire group.
Per WAR, the ’11 Phillies will rank higher if you just add the WAR numbers. If you use a FIP-based method, they will probably move up the rankings… but I don’t feel like recalculating for all of these teams, so I’m not going to do that. I’m not paid enough for that amount of labor.
Finally, as to method, if I wanted to make this better, I might make the following adjustments:
1) I would, certainly, include some FIP-based calculations. This was just a thing I came up with quickly and meant to do once, not an ongoing project. On the other hand, it feels to me like a really smart way to compare staffs, so maybe I’ll expand it out someday?
2) I would take a harder look at what sort of ERA+ I want to assign to the generic “bullpen” innings. The truth of the matter is, the earlier you look in baseball, the more likely you are to be assigning bullpen innings to pitchers worse than average starters. In the modern era, your bullpen era pitchers are likelier to have and ERA+ better than average starters. This is an enormous discrepancy, and explains why more teams push for more bullpen innings. It would radically alter the rankings if I included some adjustment for this, even if it were a very small adjustment. Modern bullpen usage is just so radically different, because there’s so much talent in the bullpen, and the patterns of use are different. We can debate whether these things are good or bad for baseball some other time, but they are present, whether we like them or not. The fact of the matter is, if the Cubs took Orval Overall out in the sixth, they were likely to get worse results out of the ‘pen for the remainder of the game; when the Phillies did the same with Vance Worley, they were probably going to be more likely to win. There should be an adjustment for that, and I just don’t have the desire to come up with it, nor have I done the research. (My guy, FWIW, would be something like assigning generic bullpen innings an ERA+ of 90 in 1900, adding an ERA+ of 0.15 each subsequent year; that means the Phillies would get an assumed bullpen score of 106.65, whereas the 1906 Cubs would get an assumed bullpen score of 90.9. That’s not a huge difference, but it would certainly help to counteract something we know to be an issue. That said, I haven’t studied the issue to know whether or not that’s an appropriate adjustment, whether the curve should be that smooth, etc., etc.)
3) I would probably compare methods: using raw WAR, a “weighted” WAR (a la Bill James), and using my own method, to see what patterns emerged.
What I feel most confident saying, after all this research is:
1) It is reasonable to conclude that the 2011 Phillies are the best staff of the last century, though they probably do not compare favorably to the ’06 Cubs;
2) The 1971 Orioles, in spite of the gaudy win totals, are historically extraordinarily overrated, and do not belong in the conversation of the greatest rotations of all-time;
3) Some of the very best pitching rotations of the last 100 years have pitched in the last 25 years. Most people probably do not believe that; most people probably believe that the best rotations pitched long ago. But actually, there have been excellent rotations quite recently, starting with those ’90s Braves, but continuing into even this last decade.
Thanks for indulging my mental meanderings!
Doom,
Thanks again for the further musings. I am not surprised by any of the staffs on your list except for, maybe, the 1926 Phila A’s. I don;t know whether to agree with you or not on the 2011 Phillies as “Hurlers Maximi” but, they certainly could have accomplished more in that ‘era’ as an organization with that kind of talent on the hill. But, I digress….The only ‘nick’ on the superlatives/positives on the current guys has to be the lack of CG’s. It has to be a lot easier to pitch every 5th day for 6-7 innings than every 4th day for 7-9 – even with a bunch of 160# middle infielders choking up and trying to move runners
The ’26 A’s were Bill James’ pick as the best rotation of the ’20s. Actually, having done this study, I would now say that the ’23 Reds deserve the nod, though I wouldn’t have known that before you mentioned them. I suspect that the 2nd A’s dynasty (the one that beat the Ruth-Gehrig Yankees) was a flashier pick, and it’s hard to argue against Lefty Grove. Eddie Rommel and Rube Walberg were also on both the ’31 and ’26 teams. So there was a lot of doubling up of names when I was crunching the numbers.
As for the desire for the Phils to have done more in the ’08-’11 period… my closest friend is a Phils fan, and he was crushed each year watching them take a step backward. Halladay, Lee, Hamels, Utley, Victorino, Howard, Rollins… it was a good era. Seems like they could’ve done more. In fact, in a two-division world, they might’ve done better. In a best-record-gets-the-pennant world, they still would’ve gotten two pennant out of it – but it would’ve been the other two years. They had the NL’s best record in ’10 and ’11, but not in ’08 nor ’09. Funny how things work. Also: the 2011 Phillies were the only National League team to win 100+ games in the period 2006-2014. (Only the ’08 Angels and ’09 Yankees did it in the AL in the same period.) I suspect they belong in a group with the ’56-’63 Braves, the ’12-’19 Nationals, and the ’91-’05 Braves as teams that did get one, but should’ve had more based on talent. Still – better to be in that group than the ’95-’01 Mariners who never even managed one…
I, too, would like to see more aggressive use of starters. Alas, I fear the decision to bring the DH to the NL will not be any help in that regard. Still, I often wonder what would happen if some team decided to go with an 11-man pitching staff and 14 hitters, and just asked more of their pitchers, then played the platoons more on offense. I suspect there’s an inefficiency there waiting to be exploited. But you’d need a real outside-the-box thinker as a GM who was willing to roll the dice on it.
You may recall Billy Martin utilizing Langford, Norris, Keough, Kingman,and McCatty till their arms fell off in the early ’80s? We’re only talking substituting one position player for one pitcher, right?…. I guess whoever that manager is, his GM better be on the same page. That 1980 A’s team imporved 29 games over their 54-108 performance in 1979 amd eventually made the playoffs in 1981
Doom,
1926 A’s, entire staff, had an ERA+ of 139. The 2011 Phightins’ 127…which, on the surface seems fairly significant. Will have to sharpen a few pencils to get the starters sorted out
A couple things: first of all, that A’s team used only 11 pitchers all season… and all 11 had an ERA+ of 100 or better. I wonder if that’s unique in baseball history. I suspect it is. (FIP would see Stan Baumgartner as sub-average… but only him, and he only pitched 22.1 innings all year.)
For my purposes, in the case of the ’26 A’s, I used the first four players B-Ref designated as their “starters,” then included the two “relief pitchers” who started 18+ games, but ignored the last player designated as a “starter,” since he only started 10 games and averaged less than 6 innings/start. I should’ve listed in my ideas for future study that I would have to come up with a stricter definition of “starter” if I were going to do this again. For example, had I done the Phillies a bit later in the study, I might’ve included Kyle Kendrick as a member of the rotation. It’s a close call. If you do include Kendrick, it raises the Phillies to a 2.85 rotation ERA (by taking 114.1 innings away from the assigned league-average bullpen). Including Slim Harriss (that 7th A’s starter) seems fair in light of including Kyle Kendrick raises them to 2.77… now ahead of the Phillies. So these are hard things, for which creating hard-and-fast rules would be difficult.
But putting the Phillies above the A’s is not crazy. The A’s best starter, Lefty Grove, had an ERA+ of 165 in 258 innings. Cliff Lee and Roy Halladay had an ERA+ of 162… in 466.1 innings. So you start with the A’s nearly 200 innings behind. Their next most-used starter was Eddie Rommel at 216 innings; Cole Hamels pitched 219. Rommel had a 134 ERA+; Hamels, conveniently, had a 137. So at this point, we’re at tied production with the Phillies 200 innings ahead. Now, if you think that the remainder of the staff of the A’s was better, that’s fine. But it’s awfully hard to make up for a 200-inning gap of the same quality. It could be that they get there. But particularly when you’re just talking starters, the Phillies essentially had two 1926 Lefty Groves. Gonna be tough for a team with just one to catch them!
Again, FWIW, B-Ref has the A’s pitchers worth 30.3 WAR and the Phils’ pitchers (again, full staff, not just starters) worth 36.5. All the pitchers for both teams worth 2.0 WAR or more (A’s and Phillies):
1) Halladay, 8.8
2) Lee, 8.5
3) Grove, 7.5
4) Hamels, 6.4
5) Rommel, 4.9
6) Ehmke, 4.1
7) Pate, 3.5
8) Quinn and Worley, 3.3
10) Walberg, 3.1
11) Oswalt and Madson, 2.1
Doom,
Looks like Fangraphs calls the ’70 Cubs “starters” good for 26 WAR; the 2011 Phillies are at 27
Doom/Voom,
That 2011 Phila staff was, believe it or not, a case of one-year wonders. While compiling the most WAR (per Fangraphs – thank you, Richard Chester !) in the live ball era amongst starters,they never led the ML in starter WAR in any surrounding seasons. As far as I can tell, these are the best “starter” seasons of the live-ball era:
2011 PHI 27.0
1970 CHI 26.0
1971 CWS 25.9
1997 ATL 25.4
1967 MIN 24.7
1996 ATL 24.6
1946 DET 24.5
1998 ATL 24.4
Two shallow observations:
1) If you pro-rate for the strike seasons of 1994 & ’95, the Braves staff did not have a season of less than 22.8 WAR for starting pitchers in any season from 1994 though 1999 and, probably qualifies on that basis as the greatest staff of the live-ball era.
2) Bill James, out of what practically appears to be “nowhere” and an absolute case of “No Way, Dude”, called the 1970 Cubs staff “great” in a historical context in, I believe, the BJHBA. And for everybody who blows off Win Shares as total steer waste, how reliable is WAR if it concurs?
To your “shallow” observations:
1) I can’t imagine any argument that ANY staff was better than the Maddux-Glavine-Smoltz crew, if you take into account a series of years. There’s nothing like them in the history of baseball. 100 years from now, they’ll probably still be anomalous (and they’ll look like one of those old-timey things that would be impossible in “today’s modern game.”)
2) I am at work, unfortunately, and don’t have my books handy. Was that in the original BJHBA, or in the NBJHBA? If it was in the original, that’s a bit of a surprise, seeing as Win Shares hadn’t yet been developed. That said, James was aware of park factors, and it doesn’t take a genius to see that the top three guys pitched nearly 900 innings of sub-3.50 ERA baseball, all in one of the most extreme hitter’s parks of all-time (118 one-year park factor!) with an offense that scored at basically a league-average rate (given that park factor). They were a great staff that year. Also, while only Jenkins made the Hall from that group of starters, it’s not too far-fetched to think that another starter may’ve joined him. (I’m always going to take the opportunity to plug an old post of mine!)
But I want to talk a little about Win Shares. Win Shares are great – they can be very instructive. And in a lot of cases, like this one, they work. Particularly when you’re comparing pitchers of one era against pitchers of another, they do the trick. But when you’re comparing pitchers against position players, Win Shares loses some of its value.
For example, the size of the “pie” devoted to pitchers and the size to hitters is predetermined. So there’s only so much to go around, rather than by revolving it around where the actual runs are. This may or may not be important; I suspect that it is. Per bWAR, in 2019, two of the top nine players in the NL were pitchers; via Win Shares, zero pitchers appeared in the top-20. In 2018, via bWAR, the top three players in the NL were pitchers; via Win Shares, Max Scherzer was #16… and that’s it. Same year, AL: via bWAR, numbers 7, 8, and 9 were pitchers; Win Shares has no pitchers in the top-20. 2017 brings a similar story. According to bWAR, Corey Kluber was tied as the 2nd-most-productive American Leaguer, and Justin Verlander ranked 9th; Win Shares says Kluber was the Junior Circuit’s 20th-most-valuable player. In the NL, bWAR has four pitchers in the top-10, while Win Shares has none. I have a very hard time buying that only one of the best 40 players in baseball in 2017 was a pitcher. This pattern continues going back for years, but I’m not going to keep going because it’s both tedious and time-consuming.
Furthermore, as I’ve detailed at length before, the biggest weakness in Win Shares is its fielding analysis, in particular that players cannot be a fielding liability. In the real world, you play Ralph Kiner, even though he produces negative value on the field. You do this because his offensive production outweighs his defensive liability. So you can look at seasons like 1949-1951, when Kiner was hitting a total of 143 homers. His Rfield is pegged at -17 in those years, meaning his fielding cost his teams seven runs relative to an average outfielder, but he had 177 Rbat (offensive runs) to more than make up for it in those years. That makes sense. Does it really make sense to anyone that Kiner was adding defensive value in those year? It does to Win Shares: he receives 2.4, 2.1, and 2.2 defensive Win Shares in those seasons. This is because there’s literally no such thing as a defensive liability in Win Shares. The problem exists on offense, too: Chris Davis in 2018: 128 games played, .168/.243/.296 line. That’s good for a -32 Rbat… or, according to Win Shares, he was a positive contributor, giving his team 0.3 Win Shares those 522 times he came to the plate. With offense, though, the numbers and opportunities get SO big, it doesn’t really worry me too much. But on defense, it’s just preposterous to think that the worst you could be is a zero. It artificially inflates bad fielders and deflates good ones: since you start at the team level, you check the “outfield.” Well, if you have a Hall of Fame guy in centerfield, who’s a +20 run defender, and you play him next to two -5 guys, Win Shares will find him to be the best still… but it’ll tell you that, essentially, you distributed those +10 runs something like 1/8/1, instead of the -5/+20/-5 WAR would give you.
What’s more, there’s the issue of defining Win Shares from wins rather than runs. Basically, Win Shares takes the 2021 Seattle Mariners and says they have 270 Win Shares to devote to their players. WAR, obviously, does things very differently. But if it’s philosophy were mapped on, it would essentially give them 228 Win Shares to spread around. (For the record, if you’ve never looked at the 2021 Seattle Mariners page, I strongly encourage you to look at the page and what’s next to “run differential.” Baseball-Reference is truly an amazing website.) This is because WAR is based on runs not wins. I think that, when you’re getting down to the individual player, it makes much more sense to look at the runs-level, because an individual has so little impact on how well-distributed those runs are. Your mileage may vary on that point; it’s not that Win Shares is inherently “wrong” to do that.
You can get to other stuff, too: B-R smooths park factors over three years, whereas I’m pretty sure Win Shares uses a one-year factor. I’m actually not sure how I feel about that one. I think Fangraphs uses a one-year factor in their WAR calculations… but someone can correct me if I’m wrong about that.
Point is, when you’re using a blunt tool, you’ll often find that it works fine for blunt answers. “Who’s the best pitching staff?” can be answered about as well by Win Shares as WAR. Well, “Who’s the best hitter?” can be answered nearly as well by OPS as it can by WAR. The thing is, you can adjust for era, park, etc., and get a better answer. But honestly, the OPS leaderboard and an Rbat leaderboard are going to have more in common than they conflict – and that’s fine. Still, all things considered, I think WAR has considerable advantages over Win Shares, in spite of the fact that Win Shares gets a lot of things – most things, actually – right. But it WAS the best fielding system prior to WAR (undeniably so); it revolutionized that analysis. Personally, I like James’ original commitment to only using whole numbers and multiplying wins by 3. There are very good reasons to do that (even though his own website uses fractional Win Shares, as I’ve quoted above). And it was the first, widely-available total player rating system I’m aware of. Plus, initially, at least, it was totally public domain: he published the whole book. I’m not saying you’d want to work through it… but if you did, you could calculate any Win Shares for any player ever. WAR is a bit more behind-the-scenes, with spreadsheets doing the calculating and no one walking you through it, step-by-step.
Ultimately, the answer to Win Shares is simple: publish Loss Shares. James has said for years that he has a calculation. But he said within the last year or so that he’s hesitant to publish them, because if it’s not perfect, once it’s published it’s published and you can’t put it back in the box. I think that’s all pretty weak sauce, actually, and that Bill James has published many, many, many things that weren’t perfect before, so this is an extremely odd hill to die on with Loss Shares. And it would more-or-less totally fix any problems one might have with Win Shares (except for the wins-versus-runs thing; but then it’s just a quirk that makes Win And Loss Shares different from WAR). Then we could really get into which system makes better choices. But ’til that day, we deal with something that’s 80% done and listen to some folks act as if they have the whole story.
And thus ends my Win Shares soap box. I can’t imagine anyone read it to the end.
While waiting for an appointment, I just did. You were in fine form with your soap boxing, Doom. 🙂
Yes, I’ll check and verify the NBJHBA for that ‘factoid” on the Cubs starters. As far as one year versus three year averaging on park factors, 1970 was an aberration with a real break-out in run scoring – kind of like 1977 (“Rawlings Rocket baseball) and 1987. With three-year averaging, this staff’s accomplishment will unjustifiably fade.
I have a real hard time with WAR from a fielding standpoint. Does someone watch baseball all night and decide, “Beltre made a real nice play there. I’ll bet only he and Scott Rolen can make that play” Then, said viewer subjectively decides that “this play he just made saved a run because it was to Beltre’s left but, if it was to his right, that’s got to be good for two runs”? Or, if a CF’er constantly plays deep, he “robs” someone of a homerun once every month (and gets credited for saving three runs) while singles fall in front of him routinely all month long?
Or, Keith Hernandez, of course a great fielder but, he’s a great fielder at 1B. Like, who do you take here: Hernandez or Willie McCovey?….From WAR’s perspective, they’re a wash based on WAR/PA. Seems bizarre to me
A couple things:
First of all, the one-year-versus-three-year park factor thing has nothing to do with an overall bump in scoring. Wrigley played as more of a hitters’ park than usual relative to the other stadiums in question. Whether the league scored more runs or not seems irrelevant to that question. It just played with more offense. But if you smooth it out over three years, you’re supposed to help determine whether what was happening was fortune or skill. Maybe Wrigley just was bouncier that year – perhaps the winds of the lake were more favorable because that’s what the weather was really like. Or maybe it was just random variation that made it appear to be so. So, again, I would say that it’s just a choice you have to make somewhere along the line. But the overall run environment of the league doesn’t matter; what matters is the individual context of the team, relative to the league.
As for the defense, I mean, it’s pretty explicitly spelled out, if you want to read The Fielding Bible Vol. III by John Dewan. (Note: John Dewan works for Bill James; I’m fairly sure, though not positive, that the current Win Shares use Dewan’s DRS, same as baseball-reference does. Of course, Bill James would actually have to tell us what he’s doing if we ever wanted to know.) There is a video component, yes; but really, it’s mostly just measuring where a player was, and how likely he is to get to a certain ball based on that positioning. And whether he hits a cutoff man or not. B-Ref says it’s this:
So it’s basically a data-driven system, not some totally subjective gut-reaction.
I’m also unclear why you think they’re only measuring certain plays and not others. You state that, somehow, a homer-saving catch would be measured, but that singles dropping in front of a guy wouldn’t be. But why would you make that assertion? The first thing listed is basically range factor, including a component that measures the velocity of the ball (so it’s worse to fail to get to a bloop single than it is to get to a rocket off the bat that no one had a chance of catching). Those things are all taken into account – every defensive play, by every fielder.
As for the players you mention (McCovey and Hernandez), their defense is based on TotalZone, which is very similar to what Bill James was doing with Win Shares, which is analyzing the defensive statistics of a team, and then drawing conclusions about the individual players based on those numbers. It’s been a while since I’ve really researched how they calculate it, and whether it’s sufficiently different from Win Shares.
The biggest problem with McCovey was probably those 3 years they tried to play him in left… most of his “bad defense” is from those seasons and after he turned 34. But I don’t think it’s crazy to think that a creaky-kneed Willie McCovey was substantially worse than average defensively. And Hernandez’s numbers aren’t totally outrageous for a guy who’s considered the best defender of all-time at his position. Ozzie Smith has 11 seasons as valuable (or more) in terms of saving defensive runs, relative to position. Cal Ripken has 7. Nomar Garciaparra and JJ Hardy have seasons better than Hernandez’s best… and that’s without the positional adjustment. So the “yeah but it’s just first base” thing is not a great piece of evidence, in my mind, because Hernandez is not receiving totally outrageous numbers.
But look at, for example, the ’79 Cardinals – a good defensive team. Their defense was saving runs. But… who was, if not Hernandez? Just looking at the names, Hernandez is certainly, by reputation, the best defender on the roster (again, relative to position). And that’s how he scores. I just don’t see it as being out of line. The main reason their WAR numbers are “out of whack” is that McCovey simply didn’t play enough. His lost so many games as a young player that his potential best WAR years are just lost… then he played till he was 42, instead of getting out like Hernandez did. So the numbers look weird on a rate basis. Had McCovey played more as a young man, the numerator would get so big that the slightly-bigger denominator wouldn’t be such a problem on a rate-basis. McCovey was probably cheated out of more playing time than just about any other Major League player (in the integrated era), so it’s going to be a tough comparison on him. Plus he was hurt all the time. But also: Keith Hernandez was a very good ballplayer, and when you factor it all in, I don’t think it’s a crazy conclusion that he was about as productive (not good, but productive) a Major Leaguer as McCovey.
Doom,
So based on the thorough evaluation described above, how is it possible they’re evaluatiing players from 50 years ago in the same fashion and to the same extent/standards as they evaluate guys today? Mike Schmidt, Graig Nettles, Brooks Robinson versus Beltre, Rolen, and Arenado….is it the same evaluation? Or are they just using standard stuff like range factor for the old guys?
They’re not using the same evaluation methods. DRS, which is the system they use now, only dates back to 2003. They can’t use it for any seasons before then. For the prior seasons, it’s a method called TotalZone. There’s a full explanation here:
https://www.baseball-reference.com/about/total_zone.shtml
But the short of it is, you have to estimate the hits to each location (using pitcher handedness, information on who fielded the ball (because that often IS known for every play, etc.) and charge them appropriately. You do know how many outs a particular player made; based on groundball/flyball tendencies, handedness of pitchers, and an individual offensive player’s profile, you can estimate how many hits he should have to a given player. Then you compare the two numbers. It’s pretty simple, actually, and pretty ingenious for eras in which we have no positioning data, batted ball speeds, etc. Baseball-Reference makes the case (and a strong one, I’d say) that the best thing to do is to use the best available defensive measure, even when that measure is not the same across time. If you truly want to make the apples-to-apples comparison between players, TotalZone is available on Baseball-Reference for any player. (It’s available under the “Finders & Advanced Stats” pull-down, and it’s the last option available: Fielding “Advanced Stats.” Here’s Derek Jeter’s page:
https://www.baseball-reference.com/players/j/jeterde01-field.shtml
You have to scroll down to “Sabermetric Fielding.”
For Jeter, TotalZone hates his fielding – as does literally every system. But from 2003-2014, Total Zone says he’s only 148 runs below average, while DRS says it was 165 below average. That’s only a difference of 1.5 R/Yr, which seems to me within any reasonable margin of error. But one could if one were so inclined, do a real check on some players and see if they would be appreciably different measured by one system or the other. Point is, it is still available; it’s just hard to find! And it’s not used in WAR calculations. But again, it’s pretty much always within the same range. Here are some players, chosen at random:
Derek Jeter (2003-2016): -165 DRS, -148 TotalZone
Albert Pujols (2003-2021): 141 DRS, 106 TotalZone
Adrian Beltre (2003-2018): 201 DRS, 152 TotalZone
Matt Chapman (2017-2021): 79 DRS, 39 TotalZone
Mike Trout (2011-2021): -2 DRS, 42 TotalZone
Yuni Betancourt (2005-2013): -69 DRS, -20 TotalZone
Carlos Gomez (2007-2019): 60 DRS, 14 TotalZone
Miguel Cabrera (2003-2021): -38 DRS, -5 TotalZone
Harrison Bader (2017-2021): 39 DRS, 22 TotalZone
So, before I did this, I wouldn’t have known, but may be true (or maybe this random list of players I thought of just happened to have this be true): TotalZone is the more conservative defensive rating. I didn’t know that; I would’ve assumed that they were all about equal. But in nearly every single case here, the player in question is closer to 0 very TotalZone than by DRS. (Mike Trout is the exception. Isn’t he always?) Point is, while some of those gaps look large, they’re really not. The largest gap on a per-year basis belongs to Matt Chapman, at about 8 R/Yr. That’s not even 1.0 WAR, though. And you’re really supposed to consider anything within 1.0 WAR to be within the margin of error. So I don’t worry about it too much. But that fact does make one wonder if Dave Winfield or Gary Sheffield’s notoriously bad defensive WAR numbers might’ve been even worse under DRS. Or maybe DRS is way too volatile, and we should think more seriously about using something a bit more “stable” like TotalZone. Interesting thought, either way. But thanks for the trip down the rabbit hole.
Yes, it appears that if I guy has a reputation as a good fielder (or bad) DRS will emphasis/exaggerate that in its findings. But, a margin of error of 1 WAR seems extreme when your average regular is good for only 2-3 WAR. Also, I would think that most players’ (good and bad) fielding skills diminish more quickly than batting.
The NBJHBA comments regarding the 1970 Cubs appear in the evaluation of Jenkins as the 23rd greatest pitcher of all time. He states they were the greatest one year staff of the 1970’s. But, according to Fangraphs WAR, they’re the second best of the entire Live-Ball era.
I believe Bill James may have cited the ’70 or ’71 Cubs with Jenkins, Hands, Holtzman, and Pappas as “great” per WS? Don’t believe their WAR exceeds 23..
Paul,
Out of curiosity, I did also try the ’23 Reds. The slotted in between the Braves and Phillies at a 2.72 ERA (using their four primary starters on a 136-game schedule). I have really mixed feelings about that, though. All those starters were used in relief at least 4 times each (22 times total). When it’s just an appearance or two, it doesn’t bother me as much. But when it gets to be this big of a number, I’m not sure how I should balance it. My guess would be that, in real life, you didn’t Dolf Luque’s relief appearances were not coming in Eppa Rixey’s games, and vice versa. Yet, that’s what this calculation kind of naturally assumes. So I’m a little more skeptical of this one, but it’s a good pull of a staff to check.
The 2002 Diamondbacks (Randy Johnson, Curt Schilling, Miguel Batista, Rick Helling, and Brian Anderson) also score quite well, at 3.07. They were one of the most durable staffs ever, starting 153 games as a unit.
Another really fun one is the 2013 Detroit Tigers. I have them at 3.11 – in 156 games for just 5 starters! Basically no one missed a start all year, and they had Justin Verlander, Max Scherzer, Rick Porcello (in a down year), and Anibal Sanchez. The thing that makes them really intriguing, though, is that they managed really great ERA numbers in front of one of the worst defensive infields you’ll ever see: Prince Fielder, Omar Infante, Jhonny Peralta, and Miguel Cabrera. Every single pitcher among them had a better FIP than ERA. Put them in front of the ’71 O’s defense, and now you’ve got an interesting conversation! I’m not going to go re-do all of these for FIP… but it might worth be trying to systematize this someday, rather than just hunting and pecking every couple hours when I have a brainstorm of a team to check… like when I thought, Y’know, there were a couple years there where Verlander and Scherzer pitched together…
Here’s an overlooked great staff:
2017 Diamondbacks
163 / 4.7 / Robbie Ray
147 / 5.7 / Greinke
139 / 4.2 / Godley
135 / 2.6 / Taijuan Walker
116 / 2.6 / Corbin
Overlooked, indeed! I scored them at a 2.94 rotation ERA… pretty close to the 1969 Cardinals of Gibson and Carlton, who score a 2.90. It’s a very impressive season.
1939 Yankees are also an interesting case.
Staff ERA+ of 132
Eight pitchers with at least 11 starts.
Six of whom had at least 2.3 WAR
That’s the kind of team where you really start to question who/what to count as part of the starting “rotation.” I mean, they score very well by this metric (2.74) using all eight players who started more than three games… but do we really consider eight guys to be a starting rotation?
Additionally, as I said, I regret not using some FIP component. Per ERA+, there are three guys on this staff (Ruffing, Hadley, Hildebrand, Sundra, and Russo) with an ERA+ of 140 or greater. Yet, per FIP, not a one of them has an FIP+ of even 130. Playing in front of a great defense would do that.
In a normalized environment of a 3.60 ERA, the ’39 Yankees would be assumed to give up 409 earned runs (as a staff; I included the relief pitchers because the spreadsheet was easier if I just copy-pasted); it goes to 477 via FIP. In other words, it’s about 16% higher using just the three true outcomes. I’m a little skeptical of a staff whose divergence is that large. But again, this could be made into a bigger study. Not that I plan to; but it’s an interesting area of exploration. And it’s really fun having people just bring up different staffs through time!
Those 2013 Tigers actually had 6 pitchers with a start, with Jose Alvarez making 6 starts and 8 relief appearances in his debut season (Tigers traded him to the Angels for Andrew Romine just before the 2014 campaign).
Only 5 teams in the modern era have used only 5 starters, the 1901 Sens and Braves and 1904 Americans, followed by the 1966 Dodgers and 2003 Mariners. That Dodger team had a 2.70 ERA in the Starter split, despite an off-season (96 ERA+) from Drysdale.
There are 20 teams with 3+ pitchers having 20+ starts and 140 ERA+ (that’s 19 teams with three, and the 1907 Cubs with five).
– 11 teams in 42 seasons (1901-42)
– Zero teams in 42 seasons (1943-84)
– 9 teams in 37 seasons (1985-2021)
I will go with Halladay.
I’m voting for Halladay.
I’m voting for Halladay.
Halladay is the clear winner in the run-off election.
Thanks everyone for your votes and the lively discussion.
I hadn’t really thought of this, but Roy Halladay was not only still alive, but actively pitching when the Circle of Great began. first post was December 9, 2012; Halladay pitched another season after that. Mariano Rivera is the only other COG-member pitcher from the COG era; Derek Jeter, Alex Rodriguez, and Ichiro Suzuki are the only position players who were active when our voting began, and who have been inducted into the COG. Halladay is, of course, the only one among them who is deceased. It’s a shame to have lost him so young.