The Roger Clemens era

February 28, 2019
 

The Roger Clemens Era

 

 

            1988 AL—Roger Clemens (Both)

            Cy Young Award—Frank Viola

 

            1988 NL—Orel Hershiser (Both Systems and Cy Young Award)

 

            1989 AL—Bret Saberhagen (Both Systems and Cy Young Award)

            Saberhagen was only 25 years old in 1989, when he won his second Cy Young Award.  He would have won it in 1987, but he rode the pedal a little too hard the first half of that season and collapsed down the stretch.  He was 15-3 with a 2.47 ERA at the All Star break, went 3-7 over the second half.

            I always thought Saberhagen—who, by the way, was married last week—I always thought he was the most perfect pitcher that I ever saw.  What I mean by that was not that he was the greatest pitcher, although he was among the greatest, but that he graded out at the top of the scale in more areas than anyone else I’ve ever seen.   I saw Saberhagen hit 98 on the radar guns that were used then, which would be 102 on the guns that are used now.  He usually sat at 93-95, which would be 96-99 on the modern guns.  He had tremendous control, and he was not afraid to throw the ball over the heart of the plate if he had to.  He had a 60 curve ball and a better slider.   It wasn’t hard to tell which was which.  He had an exceptional ability to set up the hitter, moving the pitch up and down and in and out, ticking the corner low and away on one pitch and high and tight the next pitch.  He would knock one batter off the plate in every game, usually in the first inning but sometimes as late as the third, just to remind them that he owned the plate.  He NEVER forgot to do it; he never failed to do it.  You could take it to the bank.  He was an outstanding fielder with a tremendous pickoff move.  In his career he picked 38 men off of first base, allowed only 115 stolen bases, and more than 40% of opposing base stealers were thrown out.  (Dwight Gooden, in a career of about the same length in exactly the same years, allowed 452 stolen bases.  Mike Scott, in a slightly shorter career, allowed 298.)   

            Saberhagen pitched too many innings on an immature arm, pulled up with just 167 career wins.  Randy Johnson and a few others had a better fastball; Pedro had that remarkable changeup and a lot of other good stuff.  Maddux had control like Saberhagen’s and a pitch that would jump away from the hitter at the last moment.  But I never saw anybody better, as a complete package.

 

            1989 NL—Jose DeLeon (D-WAR) Vs. Orel Hershiser (R-WAR)

            Cy Young Award—Mark Davis (reliever)

            DeLeon was just 16-12; Hershiser just 15-15.   Ten pitchers were mentioned in the NL Cy Young voting; DeLeon was not among them despite leading the league in strikeouts (201).   No one in the National League really had a Cy Young season.  I have DeLeon just 0.4 WAR better than Hershiser, four runs better, whereas Baseball Reference has Hershiser 4.1 WAR (about 41 runs) better, a rather remarkable difference.  I don’t understand how anyone could possibly conclude that Hershiser was 40 runs better than DeLeon, but if you just insisted that he was a little bit better I would not argue the issue. 

              

            1990 AL—Roger Clemens (Both Systems)

            Cy Young Award—Bob Welch

            See article (below); the luckiest pitchers of all time. 

 

The Luckiest Pitcher/Seasons of All Time

            You will not be surprised, if you are 40 years old or older, to be told that Bob Welch in 1990 was the luckiest pitcher of all time.  He pitched pretty well; he probably should have been 15-11, more or less.   Good pitcher.  He finished 27-6.  Luckiest season ever.  He won games 7-4, 5-4, 7-5, etc., but mostly, it was just that he would pitch pretty well and get a win.   He’d give up three runs, win the game 4-3 or 5-3.    He wasn’t on the wrong end of a shutout all year, and had only three starts in which he was supported by 1 or 2 runs. 

            These are the ten luckiest pitcher/seasons in my data:

Year

First

Last

Actual W

Actual L

D-Win

D-Loss

D-WPct

D-Luck

1990

Bob

Welch

27

6

14.84

11.19

.570

17.356

1989

Storm

Davis

19

7

9.41

12.98

.420

15.576

1962

Jack

Sanford

24

7

15.11

12.69

.544

14.580

2009

Braden

Looper

14

7

8.29

15.74

.345

14.455

1977

Larry

Christenson

19

6

12.05

13.02

.481

13.973

1961

Whitey

Ford

25

4

18.66

11.62

.616

13.961

1948

Jack

Kramer

17

4

9.84

10.47

.484

13.631

1971

Chuck

Dobson

15

5

8.94

12.57

.416

13.628

1951

Preacher

Roe

21

3

14.41

9.99

.591

13.580

1973

Catfish

Hunter

21

5

15.02

12.23

.551

13.212

 

            Preacher Roe (1951) was 21-3 as a starter, and also won one game as a reliever. 

            The ten luckiest pitchers of all time, presented in a note given in regard to 1933, Paul Derringer, had (as a group) a deserved won-lost record of 142-95, but an actual won-lost record of 73-182.   The ten presented here have a deserved won-lost record of 127-122, but an actual won-lost record of 202-54. 

 

 

            1990 NL—Ramon Martinez (D-WAR) Vs. Ed Whitson (D-WAR)

            Cy Young Award—Doug Drabek

            We’ve got three pitchers in play here, so let’s chart them as we have done before:

Pitcher

W

L

IP

ERA

SO

BB

PF

ERA+

D-Wins

D-Losses

D-WAR

R-WAR

F-WAR

Total WAR

Martinez

20

6

234

2.92

223

67

91

126

17

9

8.0

3.9

4.9

16.8

Whitson

14

9

229

2.60

127

57

105

148

14

9

6.1

7.0

5.2

18.3

Drabek

22

6

231

2.76

131

56

108

131

15

10

6.6

4.2

4.2

15.0

 

            It looks like the reason that Baseball Reference prefers Whitson is pretty straightforward; he has a better ERA than Martinez, in a worse park for a pitcher.   My system may unfairly favor Martinez because he piled up strikeouts.   This was Ed Whitson after he bombed in New York, a 35-year-old in his last year in the rotation.  I don’t know that I agree with his selection as the best pitcher in the league, but it is not difficult to understand or explain, so I accept it. 

 

            1991 AL—Roger Clemens (Both Systems and the Cy Young Award)

 

            1991 NL—Tom Glavine (Both Systems and won the Cy Young Award)

 

            1992 AL—Roger Clemens (Both Systems)

            Cy Young Award—Dennis Eckersley                       

 

            1992 NL—Greg Maddux (Both Systems and won the Cy Young Award)

 

            1993 AL—Randy Johnson (D-WAR) Vs. Kevin Appier (R-WAR)

            Cy Young Award—Jack McDowell

            Three pitchers in play; I’ll chart them as I have done before:

Pitcher

W

L

IP

ERA

SO

BB

PF

ERA+

D-Wins

D-Losses

D-WAR

R-WAR

F-WAR

Total WAR

Big Unit

19

8

255

3.24

308

99

99

135

21

7

10.9

6.6

7.0

24.5

Appier

18

8

239

2.56

186

81

112

179

20

7

10.5

9.3

6.8

26.6

McDowell

22

10

257

3.37

158

69

99

125

15

10

6.3

4.4

5.2

15.9

 

            Appier clearly wins this one.   Appier’s ERA is not only better, but much better, and Appier was pitching in a park which not only appears in that season’s data to be tougher for a pitcher, but shows that way consistently over a period of years.  My system over-values strikeouts just a little bit, as does Fangraphs, which becomes relevant when Randy Johnson is contrasted with a pitcher having a similar season. 

            The Cy Young voters were still in the sway of won-lost records.

 

            1993 NL—Greg Maddux (D-WAR) Vs. Jose Rijo (R-WAR)

            Cy Young Award—Greg Maddux

            Maddux was 20-10 with a 2.36 ERA; Rijo was 14-9 with a 2.48 ERA.   Park Factors the same.  Rijo, fifth in the Cy Young voting, allowed only 5 un-earned runs while Maddux allowed 15, and Rijo struck out 227 hitters to Maddux’ 197.    Baseball Reference sees Rijo as 3.4 WAR (34 runs) better than Maddux, which is kind of astonishing, and obviously indefensible.   But if you stick to the argument that Rijo is SOMEWHAT better than Maddux, a couple of runs better, then I think you can make as good an argument for Rijo as you can for Maddux. 

 

 

            1994 AL—Roger Clemens (D-WAR) Vs. David Cone (R-WAR)

            Cy Young Award—David Cone

            1994 was a strike season; the teams played about 110 games each.  Cone went 16-5 with a 2.72 ERA.    Clemens was 9-7 with a 2.85 ERA, but had the league’s best ERA+ at 176—actually his best since 1990, and he was leading the league pretty much every season.   Nonetheless, because he "won" only 9 games, there was a perception at the time that he had slipped far backward, and people were writing not only that, but that he was fat and no longer committed.  That was totally wrong, and it set up the narrative that unfolded in the following seasons, with bitterness between Clemens and the Red Sox front office.

            But Cone or Clemens in 1994. . .well, I kind of think you would have to take Cone.   Baseball Reference has Clemens with 6.0 WAR, Cone with 6.9.   That’s not unreasonable.   I have Clemens with a smaller advantage going the other direction (7.9 to 7.3), but certainly Cone has at least as good an argument to be considered the number one pitcher in the league as does Clemens. 

 

            1994 NL—Greg Maddux (Both Systems and won the Cy Young Award)

 

            1995 AL—Randy Johnson (Both Systems and won the Cy Young Award)

 

            1995 NL—Greg Maddux (Both Systems and won the Cy Young Award)

 

            1996 AL—Roger Clemens (D-WAR) Vs. Pat Hentgen (R-WAR)

            Cy Young Award:  Pat Hentgen

            Roger Clemens from 1984 to 1992 had won-lost records which were fair representations of how well he had pitched.   In 1987, as I figure it, he had exactly the won-lost record that he deserved—20-9.   That’s an odd thing that doesn’t happen often, but over the years, his won-lost records were consistent with how he pitched:

 

DESERVED

 

ACTUAL

Year

W

L

 

W

L

1984

8

7

 

9

4

1985

7

4

 

7

5

1986

21

6

 

24

4

1987

20

9

 

20

9

1988

21

8

 

18

12

1989

18

9

 

17

11

1990

19

6

 

21

6

1991

21

8

 

18

10

1992

19

7

 

18

11

 

 

 

 

 

 

Total

154

64

 

152

72

 

            Then he went through a four-year period, 1993 to 1996, in which his won-lost luck was just terrible:

 

 

DESERVED

 

ACTUAL

Year

W

L

 

W

L

1993

13

9

 

11

14

1994

15

5

 

9

7

1995

10

7

 

10

5

1996

19

8

 

10

13

 

 

 

 

 

 

Total

57

29

 

40

39

 

            To have a four-year stretch, bad luck in three out of four years, is not in a statistical sense a meaningful event.  About 20% of pitchers pitch in relatively tough luck in any season.  If you assume that’s the case, about 3% of pitchers would pitch in tough luck in three out of four seasons.  His 1996 season was "luck historic", on the list of the ten unluckiest seasons in baseball history.  He was 10-13; he should have been about 19-8. 

            Later in his career (2001 to 2004) Clemens had a four-year stretch in which his won-lost luck was good every year.  He went 20-3 when he should have been about 16-9, then 13-6 when he should have been 12-10, then 17-9 when he should have been 15-10, then 18-4 when he should have been 18-8.   He wound up winning a couple of Cy Young Awards for which he was probably not the most deserving candidate, because he was fortunate (in those years) in the won-lost columns.

            I figure that Hentgen in ’96 also should have been about 19-8, the same record as Clemens.  Hentgen finished 20-10.  I have Clemens ahead by a nose, but I wouldn’t argue that point either way.   

            We treat meaningless events as meaningful events, and we treat meaningful events as meaningless events, and this often has unfortunate consequences for all of us.   It did for Clemens, and it did for the Red Sox.   This is one of the things I say to myself, when I am trying to understand why I do what I do.   I am trying to learn distinguish better between what is real and meaningful, and what is not. 

 

            1996 NL—John Smoltz (D-WAR) Vs. Kevin Brown (R-WAR)

            Cy Young Award:  John Smoltz, with Kevin Brown second in the voting. 

            Kevin Brown had a 1.89 ERA for Florida, and Smoltz was 24-8 for Atlanta.   I have Smoltz ahead by 6 runs (10.6 WAR to 10.0); Baseball Reference has Brown ahead by 5 runs (7.9 to 7.4).   Obviously, you can argue it either way. 

 

 

            1997 AL—Roger Clemens (Both Systems and won the Cy Young Award)

 

            1997 NL—Pedro Martinez (Both Systems and won the Cy Young Award)

 

            1998 AL—Roger Clemens (Both Systems and won the Cy Young Award)

 

            1998 NL—Greg Maddux (D-WAR) Vs. Kevin Brown (R-WAR)

            Cy Young Award—Tom Glavine

            I’ll chart the data for the three candidates:

Pitcher

W

L

IP

ERA

SO

BB

PF

ERA+

D-Wins

D-Losses

D-WAR

R-WAR

F-WAR

Total WAR

Maddux

18

9

251

2.22

204

45

100

187

21

7

11.1

6.6

7.5

25.2

Brown

18

7

257

2.38

257

49

73

164

19

8

9.4

8.6

9.6

27.6

Glavine

20

6

229

2.47

157

74

100

168

17

9

8.1

6.1

4.8

19.0

 

            Maddux and Glavine pitched for Atlanta, of course, and Kevin Brown for San Diego.  It would seem to me that since Brown has a huge advantage in the park in which he pitched and no advantage in ERA that he would be the least favored of the three.   But since Fangraphs and Baseball Reference both agree that he was the best pitcher in the league, I will accept their conclusion. 

            Brown was a singular pitcher.   I think he was the only pitcher of my lifetime who threw hard and got ground balls and had a long career.   The other guys I have seen who threw hard and got ground balls all burned out after a couple of good years (Chien-Ming Wang, Brandon Webb, Roger Erickson.)  Tommy John threw pretty hard when he came up; people remember him as a 40-year old, plus he came out of the Cleveland system at the same time as Sam McDowell and Luis Tiant, who threw REALLY hard, but TJ had a very good fastball when he was 25.   Spahn did too, I guess; he was a high fastball pitcher when he was young.

 

 

 
 

COMMENTS (7 Comments, most recent shown first)

CharlesSaeger
Blauser was pretty well spotted. He only appeared in about half the innings for the season overall (778.1). At his most frequently-used position, shortstop, he played 435 innings, but good-fielding Rafael Belliard got in 990. Nixon was similarly used, which seems weird since he's the good fielder of the four while Blauser was the crummy fielder of the infielders. BB-ref.com says the Braves were next to last in CG by all fielders, odd for a league champion. Cox was a damn good manager.

Going back over my own evaluation from when someone first pointed out this bizarre situation at BB-ref.com, I see that I had both the outfielders and infielders at +1 run each, with -6 for the catchers, so my memory was spotty. I didn't break out the catchers individually for some reason, but among the non-catchers, Blauser at -6 runs is indeed the worst, while Pendleton at +5 was the best.

Among the non-DER elements, double play rate was below average, opponents caught stealing rate was below average, reached on error rate was below average, advancement on balls hit to the outfield was below average, errors that didn't put a runner on base was average.
6:04 PM Mar 2nd
 
shthar
Lesse; Bream, Treadway, Belliard & Pendleton around the infield.

David Justice, Ron Gant and Lonnie 'Skates' Smith in the outfield.

Otis Nixon got in a 124 games. But so did Jeff Blauser.


1:36 AM Mar 1st
 
CharlesSaeger
@chuck: My evaluation of the traditional statistics plus a few more say that the team overall was average afield and that the outfielders were the ones dragging it down.
9:15 PM Feb 28th
 
bjames
Regarding Glavine in 1991. . . I have almost the same WAR for him that Baseball Reference does, 8.5 vs. 8.4. I see that the defensive support number for him is odd, but for some reason it works out about the same.
6:28 PM Feb 28th
 
chuck
CharlesSaeger, I looked at this a little bit in the 1962 Cy Young thread today, post #17. Still no explanation for Baseball-Reference's numbers, though.

I see from the team splits that Atlanta allowed a much better average on balls in play allowed on ground balls (.182) than the NL without them (.218). That was a savings of 73 hits just on grounders.
On balls in the air, though- liners or flys in play- the Braves had a ball-in-play average of .343 compared to the NL without them of .335. That was 18 hits worse than league average. Sure doesn't look like that tallies up to a horrendous defense, though.

Even if it just was a little above average defense, that would still mean a very sizable correction for the defensive support estimated for Glavine and bring him back to the pack of other pitchers in WAR that year.
3:10 PM Feb 28th
 
shthar
So at this point which War is winning?

Dwar or Rwar?


11:08 AM Feb 28th
 
CharlesSaeger
Clue flood of folks claiming that Clemens was "washed-up" in 1996 in 5 ... 4 ... 3 ...

@bjames: I mentioned this in the comments to yesterday's article, but how does Glavine's 1991 shape up in your system versus BB-ref.com's? I ask since BB-ref.com has a downright idiotic evaluation of the Braves' fielding: leading the league in DER in a hitters' park but being 100 runs below average. Granted, their other fielding markers (double plays, baserunner advancement on hits to the outfield, steals, wild pitches, passed balls, fielding percentage) aren't good (I figured they were average after adding those), but I'd imagine there's a gap between your system and its owing to that.
8:53 AM Feb 28th
 
 
©2019 Be Jolly, Inc. All Rights Reserved.|Web site design and development by Americaneagle.com|Terms & Conditions|Privacy Policy