To trace this argument back to its earliest recess in my memory, I remember in the fall of 1969, arguing with Tony Bandle in the kitchen at Stephenson Hall.    We were arguing about the relative merits of Juan Marichal versus Bob Gibson, and Tony was from St. Louis, so you can guess where he stood on this.  Marichal, I pointed out, had won more games than Gibson in 1962, 1963, 1964, 1965, 1966, 1967, 1968 and 1969.   Over the eight years Marichal had won 24 games more than Gibson, and had lost eleven fewer.

            “Yeah,” said Tony, “but I’ve heard that Marichal picks up a lot of wins by beating up on the Mets for all those years and the Astros, the bad teams.  Gibson’s record is better against the good teams.”

            “Bullshit,” I argued eloquently.

            “Bullshit yourself,” Tony replied with annoyance.  Tony was a very level-headed guy, thank God.   He was as big as I am, which is to say he was a big guy, but he was much, much more athletic than I was, and he could have pounded me into gyro meat.  I remember one time I threw a coke can at him and hit him in the forehead, opening up a circle of blood between his hairline and his eyebrows.   I apologized, of course, and Tony shrugged it off.   Another time I launched a campaign to have somebody else replace him as the Hall’s athletic chairman; our friendship survived that, too.   But now he was annoyed.  This was Bob Gibson we were talking about.

            The question we are researching here, 41 years later, is the extent to which a pitcher exploits weak competition, versus the extent to which he steps up against the strongest teams in the league.    This question came up recently in “Hey, Bill”:

 

Hey, Bill, In your Statistics section on this site, you ask/wonder, what a pitcher's record is against "weak sisters" vs "quality" teams. . . How deep do you go with this? ...a quality team could be a QT one year and a WS another. . .what parameters describe "weak" team and "quality" team? I once worked out Don Drysdale's record against contenders in years when the Dodgers contended or won. Strictly arbitrarily, I defined contender as finishing the season 6 games or less out of first place. Drysdale was under .500 vs those teams, and in a pretty good sample because it often covered several teams in one season. (He was 49-48 or something like that, and on that basis I questioned his HOF credentials.) I still have the work on that. Do you think 6 games too lenient or not lenient enough? Your item in the stat section brought the DD work to mind. Asked by: Kev Answered: April 16, 2010 Thanks.   It would pretty easy, in the modern world, to do some larger studies along this line.

 

            I realize now, reviewing my answer, that I didn’t actually answer his question; sorry about that.   Anyway, my son Isaac created for me a series of spreadsheets which have all of the game lines for every game in the Retrosheet data, going back to 1952.  I realized that I could use that data to analyze this question very thoroughly.

            OK, here’s what I did.   I started by listing all of the teams in each decade by their winning percentage.   I divided these into four groups:

            A Teams were the top 25% of teams in each decade,

            B Teams were the second 25%,

            C Teams were the third 25%, and

            D Teams were the bottom 25%.  

 

            A Teams, in general, had winning percentages over .555,

            B Teams generally had winning percentages of .500 to .555,

            C Teams generally had winning percentages of .445 to .500, and

            D Teams generally had winning percentages under .445.

 

            But the specific percentages shifted a little bit from decade to decade.

 

            I then ran the data for every pitcher since 1952.   My study includes only starts, not relief appearances, and it includes only the data that is in the Retrosheet files from 1952 through 2009.    Also—full disclosure—I had to throw a few pitchers overboard because my data couldn’t distinguish between them.  I think there were three sets of pitchers who weren’t distinguishable—two Bobby Jones, two or three Bob Millers, and one other one.   I took the time to sort out the Kevin Browns, because one of them is an important pitcher, and I didn’t confuse the two Bob Gibsons or anything like that.

            So let me get to a little bit of the data.   This is the career data for Don Drysdale:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Don	Drysdale	92	657.2	31	33	.484	457	141	3.37	A
Don	Drysdale	158	1102.0	64	67	.489	828	306	3.32	B
Don	Drysdale	107	730.1	46	36	.561	490	176	3.01	C
Don	Drysdale	106	836.2	61	23	.726	635	193	2.04	D
Don	Drysdale	250	1759.2	95	100	.487	1285	447	3.34	Good
Don	Drysdale	213	1567.0	107	59	.645	1125	369	2.49	Weak
Don	Drysdale	463	3326.2	202	159	.560	2410	816	2.94	Total

 

            Don Drysdale had 465 career starts; we have 463 of them, so Retrosheet may be missing a couple of Drysdale’s games; either that, or I mixed something up.    Anyway, in the data that I have, Drysdale was 31-33 against “A” quality teams and 64-67 against “B” quality teams, whereas he was 61-23 with a 2.04 ERA against the bottom-of-the-league type of teams.   Kev had him at 49-48 against contending teams, so that’s reasonably consistent with our study.

            Can we conclude, then, that Drysdale was a “bully”, so to speak, beating up on the league’s weakest teams, but struggling to hold his own against the other strong teams?

            Well. . .not really.    Drysdale was 31-33 with a 3.37 ERA against “A” quality competition, but let’s look at some other guys’ records against the top 25%.   I’ll highlight the Hall of Famers on this list, and the guys who aren’t highlighted are good pitchers, too:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA
Don	Drysdale	92	657.2	31	33	.484	457	141	3.37
Dennis	Eckersley	100	676.0	38	41	.481	475	166	4.33
Jamie	Moyer	151	931.0	50	54	.481	560	260	4.61
Ferguson	Jenkins	143	987.1	59	64	.480	660	244	3.84
Bob	Lemon	53	361.1	22	24	.478	132	159	3.91
Tim	Hudson	80	524.1	27	30	.474	339	181	3.97
Bob	Gibson	96	737.2	41	46	.471	593	254	3.33
Steve	Carlton	133	932.1	48	54	.471	784	382	3.69
Mark	Buehrle	90	574.2	31	36	.463	362	152	4.35
Dennis	Martinez	166	1111.2	61	71	.462	622	352	3.65
Jim	Kaat	141	880.0	48	56	.462	469	227	4.07
Jack	Morris	127	877.2	48	57	.457	559	376	4.50
Bert	Blyleven	187	1356.2	70	84	.455	1037	366	3.56
Don	Sutton	172	1125.0	55	66	.455	778	301	3.79
Tom	Seaver	138	998.0	50	60	.455	775	319	3.27
Luis	Tiant	130	878.1	45	54	.455	583	262	3.83
Catfish	Hunter	114	772.0	41	50	.451	461	248	3.80
Dave	Stieb	105	726.1	35	43	.449	375	269	3.40
Frank	Viola	117	768.0	39	48	.448	504	253	4.15
Mike	Scott	80	489.2	26	32	.448	318	145	3.86
Steve	Rogers	103	733.1	37	46	.446	451	232	3.60
Vida	Blue	131	849.1	45	56	.446	509	300	3.80
Phil	Niekro	183	1320.2	64	80	.444	818	507	3.44
Dave	Stewart	69	452.1	24	30	.444	327	201	4.32
Chris	Carpenter	58	349.0	16	20	.444	242	142	5.16
Jerry	Koosman	128	888.2	47	59	.443	567	285	3.71
Orel	Hershiser	104	661.2	31	39	.443	440	232	3.99
Curt	Simmons	98	684.2	34	43	.442	337	207	3.48
Ken	Holtzman	72	446.2	26	33	.441	241	163	4.57
Robin	Roberts	129	955.0	49	63	.437	439	166	3.79
Jim	Perry	99	643.2	33	43	.434	288	220	3.87
Billy	Pierce	105	721.1	37	50	.425	464	227	3.53
Mickey	Lolich	114	812.1	39	53	.424	609	244	3.37
Frank	Tanana	168	1075.2	53	73	.421	704	341	3.90
Camilo	Pascual	101	636.2	32	46	.410	468	251	4.23
Early	Wynn	94	630.0	32	46	.410	386	251	4.29
Gaylord	Perry	179	1290.1	59	85	.410	823	339	3.52
Warren	Spahn	77	538.2	27	40	.403	246	141	3.79
Nolan	Ryan	208	1421.2	66	98	.402	1499	700	3.18
Lew	Burdette	68	447.0	21	32	.396	172	122	4.35
Sam	McDowell	89	583.0	26	42	.382	544	284	3.95
Bob	Friend	110	729.0	32	52	.381	342	184	3.84
C.C.	Sabathia	67	425.1	18	30	.375	321	146	4.17
Denny	McLain	65	399.2	20	34	.370	247	140	4.82
Zack	Greinke	32	185.0	6	18	.250	131	57	4.28
Don	Larsen	39	243.2	6	21	.222	117	119	4.06
Herb	Score	28	155.2	4	14	.222	127	108	3.99

 

            In having a losing record against the very best opponents, Don Drysdale has lots of really, really good company.  

            There were 702 pitchers in my study who made 100 or more starts.    Among those 702 pitchers, Drysdale ranks fifth in his overall effectiveness against the weakest teams, fifth out of 702.   That’s really good.

            But Drysdale also ranks 54^th among the 702 pitchers in his effectiveness against the strongest teams.   That’s pretty good, too.   A 31-33 record against “A” quality competition, over the course of a career, is borderline outstanding.

            We’ll deal at the end of the article with some of the pitchers whose records against “A” competition were even better than Drysdale’s, but in any case Drysdale was a little bit of a bully, in the sense that he gained more against weak competition than did the average pitcher.  Drysdale gained much more against weak competition than did Koufax, for example.

            How do we measure the extent to which each pitcher dominated inferior competition?    I looked at six factors relative to that issue, which were:

 

            1)  The percentage of the pitcher’s wins that came over “D” quality competition,

            2)  The difference in the pitcher’s winning percentage versus “A & B” teams and his winning percentage versus “C & D” teams,

            3)  The difference in the pitcher’s ERA versus “A & B” teams and his ERA versus “C & D” teams,

            4)  The difference in the pitcher’s overall effectiveness RANK (1 to 702) versus “A & B” teams and his overall effectiveness rank versus “C & D” teams,

            5)  The difference in the pitcher’s overall effectiveness rank (1 to 702) versus “A” teams compared to his overall effectiveness rank versus all teams, and

            6)   The player’s career win total versus “A & B” teams compared to his career wins versus “C & D” teams.

 

            I made up an index of these six indicators, which I called the “Bully Factor”; a high Bully Factor indicates that the pitcher pitched much better against weak competition than against strong competition—much better, or in some cases much more.   Later, I’ll list the pitchers at the top and bottom of the chart, but first, let’s look at the guys with the most “normal” data, the guys in the center of the chart.

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Dan	Petry	85	536.2	26	36	.419	241	208	4.28	A
Dan	Petry	66	420.0	27	25	.519	231	162	3.88	B
Dan	Petry	78	494.2	29	28	.509	272	218	4.13	C
Dan	Petry	71	475.0	36	15	.706	249	187	3.39	D
Dan	Petry	151	956.2	53	61	.465	472	370	4.10	Good
Dan	Petry	149	969.2	65	43	.602	521	405	3.77	Weak
Dan	Petry	300	1926.1	118	104	.532	993	775	3.93	Total

 

            Petry was 36-15 against “D” opponents, 26-36 against “A” opponents.   However, he was better against “B” opponents than “C”, and overall, his data shows a normal improvement against weaker teams, neither more nor less.    Jamie Moyer:

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Jamie	Moyer	151	931.0	50	54	.481	560	260	4.61	A
Jamie	Moyer	144	862.0	59	51	.536	501	263	4.46	B
Jamie	Moyer	159	1009.1	70	42	.625	601	274	3.90	C
Jamie	Moyer	155	979.2	72	45	.615	592	269	4.01	D
Jamie	Moyer	295	1793.0	109	105	.509	1061	523	4.54	Good
Jamie	Moyer	314	1989.0	142	87	.620	1193	543	3.95	Weak
Jamie	Moyer	609	3782.0	251	192	.567	2254	1066	4.23	Total

 

            Again, a normal improvement against weaker teams.   If we figure that an average pitcher should have a .450 winning percentage against “Good” teams and a .550 winning percentage against “Weak” teams, Moyer is .059 better-than-average against good teams, and .070 better-than-average against weak teams, with a corresponding improvement in his ERA.   A normal improvement.   Tom Seaver:

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Tom	Seaver	138	998.0	50	60	.455	775	319	3.27	A
Tom	Seaver	168	1242.2	80	52	.606	892	347	2.76	B
Tom	Seaver	193	1393.2	85	65	.567	1042	411	3.01	C
Tom	Seaver	148	1142.0	95	26	.785	925	310	2.42	D
Tom	Seaver	306	2240.2	130	112	.537	1667	666	2.99	Good
Tom	Seaver	341	2535.2	180	91	.664	1967	721	2.74	Weak
Tom	Seaver	647	4776.1	310	203	.604	3634	1387	2.86	Total

 

            And Mike Mussina:

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Mike	Mussina	141	923.2	57	52	.523	702	247	4.11	A
Mike	Mussina	126	845.0	65	33	.663	700	178	3.58	B
Mike	Mussina	108	722.1	59	25	.702	564	152	3.36	C
Mike	Mussina	161	1068.0	89	43	.674	846	208	3.61	D
Mike	Mussina	267	1768.2	122	85	.589	1402	425	3.86	Good
Mike	Mussina	269	1790.1	148	68	.685	1410	360	3.51	Weak
Mike	Mussina	536	3559.0	270	153	.638	2812	785	3.68	Total

 

            That’s what normal, “central tendency” data looks like.   Don Drysdale was a little bit of a bully, just a little bit.    These guys all had Bully Factors around .500; Drysdale was at .614.   That’s not really a noteworthy number; it just means he picked up a little bit more than average against weaker competition.   But he was no Jack Morris.

            Jack Morris, of course, is famous for his post-season performances, and one might suppose, based on that, that Morris was always at his best against quality opposition.   In fact, among major pitchers, Morris is the most notable “bully” of the last 50 years:

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Jack	Morris	127	877.2	48	57	.457	559	376	4.50	A
Jack	Morris	125	864.1	44	57	.436	592	321	4.14	B
Jack	Morris	142	1025.2	77	39	.664	670	346	3.64	C
Jack	Morris	133	978.2	82	29	.739	622	313	3.42	D
Jack	Morris	252	1742.0	92	114	.447	1151	697	4.32	Good
Jack	Morris	275	2004.1	159	68	.700	1292	659	3.53	Weak
Jack	Morris	527	3746.1	251	182	.580	2443	1356	3.90	Total

 

            Twenty-two games under .500 against first-division teams but 91 games over .500 against second-division opponents, Morris has a Bully Factor of .937.    The only big-name pitcher who has a higher one is Early Wynn:

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Early	Wynn	94	630.0	32	46	.410	386	251	4.29	A
Early	Wynn	81	594.2	36	35	.507	355	228	3.62	B
Early	Wynn	74	518.0	34	23	.596	340	230	2.92	C
Early	Wynn	109	810.2	70	19	.787	480	309	2.96	D
Early	Wynn	175	1224.2	68	81	.456	741	479	3.96	Good
Early	Wynn	183	1328.2	104	42	.712	820	539	2.95	Weak
Early	Wynn	358	2553.1	172	123	.583	1561	1018	3.43	Total

 

            70-19 against the weakest opponents; that is really taking care of business.    Of course, we’re missing almost half of Wynn’s career starts, and. . .that matters.    In fact, many of the pitchers from that generation, for whom we are missing their early-career data but have their late-career data. . .many of them have high or very high Bully Factors.   Like Johnny Antonelli:

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Johnny	Antonelli	45	278.0	13	22	.371	146	91	3.88	A
Johnny	Antonelli	51	341.1	15	27	.357	204	114	3.61	B
Johnny	Antonelli	87	621.1	46	25	.648	353	201	2.82	C
Johnny	Antonelli	43	277.0	20	12	.625	162	103	3.05	D
Johnny	Antonelli	96	619.1	28	49	.364	350	205	3.73	Good
Johnny	Antonelli	130	898.1	66	37	.641	515	304	2.90	Weak
Johnny	Antonelli	226	1517.2	94	86	.522	865	509	3.24	Total

 

 

            And even Warren Spahn:

 

Last	First	G	IP	W	L	Pct	SO	BB	ERA	Group
Warren	Spahn	77	538.2	27	40	.403	246	141	3.79	A
Warren	Spahn	125	968.2	64	49	.566	441	247	3.22	B
Warren	Spahn	132	1034.1	79	37	.681	474	257	2.86	C
Warren	Spahn	91	695.2	57	20	.740	373	169	2.81	D
Warren	Spahn	202	1507.1	91	89	.506	687	388	3.43	Good
Warren	Spahn	223	1730.0	136	57	.705	847	426	2.84	Weak
Warren	Spahn	425	3237.1	227	146	.609	1534	814	3.11	Total

 

            Spahn’s Bully Factor based on the data we have was .716, which is a little bit high, not really high like Jack Morris, Early Wynn and Johnny Antonelli.   Johnny Antonelli, for those of you who don’t live in the past, was kind of like Warren Spahn, only he didn’t last; he had about a six- or seven-year period where he was on the same level as Warren Spahn, only Spahn stayed on that level for seventeen years.

            Anyway, we are missing the “early career” data for Wynn, Antonelli and Spahn, and this may be quite significant, because there may be a syndrome wherein pitchers of that quality are effective against high-quality opposition early in their careers, and then not so much later in their careers:

            a)  Jim Bunning in the 1950s, when he was young, was 12-7 against “A” quality opponents, with a 3.37 ERA.  From 1960 on he was 32-38 against A quality opposition.

            b)  Don Drysdale in the 1950s was 7-1 with a 2.51 ERA against A quality opponents.   From 1960 on he was 24-32 against them.

            c)  Steve Blass in the 1960s was 12-9 with a 3.05 ERA against A quality opponents.   From 1970 on he was 4-8 against them, with an ERA near 4.00.

            d)  Joaquin Andujar in the 1970s—when he was really not a top-flight pitcher, although he had a great arm—was 13-10 against A quality opponents.   In the 1980s, when he was a top-rank pitcher, he was 25-28 against them, with a higher ERA.

            d)  Rich Gale in the late 1970s, as a young pitcher, was 8-4 against top-quality competition.  From 1980 on he was 5-14 against them.

            e)  Ron Guidry in the 1970s was 17-7 against A quality opponents.   From 1980 on he was 22-22 against them.

            f)  Jack Morris in the 1970s was 8-2 against A quality opponents, 3.25 ERA.   From 1980 on he was 40-55 against A quality opposition.

            g)   Rick Rhoden in the 1970s was 8-3 with a 3.09 ERA against A opposition.   From 1980 on he was 34-30 with a 4.04 ERA against them.

            h)  Dwight Gooden in the 1980s was 24-8 with a 2.44 ERA against A quality opposition—one hell of a record, given that these are games against the best teams in the league.   From 1990 on he was 19-20 against A quality opponents, with an ERA well over 4.00.

            i)  Roger Clemens in the 1980s was 19-11 with a 2.94 ERA against A quality opponents.   From 1990 on he was 41-48 against A quality opposition.

            I’m not saying that it is absolutely true that young, very talented power pitchers, working with a little extra adrenaline, are able to stuff the highest quality opponents, whereas later in their careers they are less able to do this.   I am not saying that is true; I am saying there is enough reason to believe that that might be true that, in looking at the “end of career” data for pitchers like Early Wynn, Warren Spahn and Johnny Antonelli, we should remember that we may be missing the data for their best seasons against high-quality opposition.   It’s possible. 

 

            Of course, it is also possible that we’re not really measuring anything here, other than the skill levels of the pitchers and some random events and some scheduling discrepancies.   That could be.   Let me make my best argument that there’s something real going on here:  Whitey Ford and Bob Turley.

            Whitey Ford and Bob Turley were teammates from 1955 to 1962—two-thirds of Turley’s career.    These are their records against “D” quality opponents:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Bob	Turley	77	539.2	43	11	.796	413	326	3.02	D
Whitey	Ford	118	864.2	75	20	.789	553	305	2.70	D

 

            Not much difference, right?   Turley’s Winning Percentage is a little bit better; Ford’s ERA is a little bit better.   But these are their records against “A” quality opponents:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Bob	Turley	51	327.0	11	30	.268	245	213	4.21	A
Whitey	Ford	107	729.2	51	29	.637	456	238	2.55	A

 

            This does not appear to be a random difference.   At this point I should caution us all that the fact that the difference is not random does not mean that it indicates exactly what it might superficially seem to indicate.    Ford had a better career ERA against “A” opponents than against “D” opponents.    One reason for this is that the “A” opponents for him, in the 1960s, were very often the Chicago White Sox, who played in an extremely low-run environment.   In those days the home team had custody of the baseballs until they were given to the umpires before the game.   It is alleged that the White Sox used to store their baseballs in a freezer so that the balls would be dead, which would maximize the White Sox ability to win by manufacturing runs, and minimize their inability to hit home runs.   I don’t think this is just a story; my belief is that that’s actually true.    I believe, subject to correction, that the policy of delivering the baseballs to the home team, rather than the umpires, was changed because the American League realized that the White Sox were in fact doing this.

            Anyway, there could be factors in the data that cause Ford to excel against superior opponents and Turley to struggle, other than actual differences between Ford and Turley.   But. . .there’s something going on there.

            In 1963, as Turley moved on, the Yankees came up with Al Downing, who looked in 1963 like he might be on the level of Koufax, and in 1964 they came up with Mel Stottlemyre, who was a sinker ball pitcher who became the Yankees ace during the Horace Clarke years and the Bobby Murcer years; homage to Rob Neyer there, Neyer’s been “naming” team eras.   In 1963 Downing, essentially a rookie, was 13-5 with 171 strikeouts in 176 innings, 2.56 ERA; in 1964 Stottlemyre, truly a rookie, was 9-3 with a 2.06 ERA.   These are their career records against different qualities of opposition:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Al	Downing	141	899.2	36	56	.391	632	378	3.53	Good
Al	Downing	176	1189.2	80	46	.635	863	466	2.98	Weak
Mel	Stottlemyre	181	1358.0	81	72	.529	594	420	3.04	Good
Mel	Stottlemyre	175	1296.2	83	66	.557	661	384	2.88	Weak

 

            Against “C” and “D” opponents, Downing was as good as Stottlemyre.   Against “A” and “B” opponents, Stottlemyre was far better than Downing. 

 

            Another matched set for you:  Gary Bell and Jim Perry.  In the 1950s the Cleveland Indians had a legendary pitching machine.   For most of the 1950s the Indians would literally put an outstanding starting pitcher on the mound every day of the season, and they seemed to be able to produce these at will, until Herb Score got hurt.   In 1958, while the image of the powerhouse Indians’ pitching continued to linger, the Indians came up with Gary Bell, who was 12-10 as a rookie, then went 16-11 in his second season.    In 1959 they came up with Jim Perry, who was 12-10 as a rookie, then went 18-10 in his second season.

At the time, Bell and Perry (and Mudcat Grant) were looked upon as the young lions of the Indians new pitching dynasty.  (You will notice that I didn’t say “young chiefs” or “young braves”.   I want props for that.)  But it didn’t work out for them.   Bell, after going 16-11 in 1959, went 9-10 in 1960, 12-16 in 1961.   Still believing that Bell had outstanding stuff, the Indians shuffled Bell into the bullpen, not even a key role in the bullpen, but making him a spot starter and garbage man reliever.   He worked in that role for several years while the Indians trotted out more outstanding young arms (Sam McDowell, Luis Tiant, Tommy John, Sonny Siebert, Steve Hargan), becoming known now as the organization that produced outstanding pitching prospects but didn’t have any idea what to do with them.   In 1966, after four years wandering in the desert, Bell finally got back in the rotation, going 14-15 in 254 innings, but with a good ERA and a good strikeout to walk ratio (194-79).   Traded to the Red Sox in early 1967, Bell went 12-8 for the Red Sox in 1967, helping them to the pennant, and had another pretty good season as a starter in 1968.

Perry’s biography has strong parallels.  After his 18-10 season in 1960 he dropped to 10-17 in 1961 and 12-12 in 1962.   The Indians traded him to the Twins and the Twins buried him in their bullpen, making him a spot starter and garbage man reliever for several years, although he started more in those years than Bell did.   After several years in that role he finally got back into the rotation in 1969.   He went 20-6 in 1969, helping the Twins to the division championship, and then went 24-12 in 1970, winning the Cy Young Award.

But while their career paths are similar, and one can see in the data that both were “spotted” against weaker opponents, Bell is one of the most extreme “bullies” in our data, whereas Perry had little tendency to dominate inferior opposition:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Jim	Perry	99	643.2	33	43	.434	288	220	3.87	A
Jim	Perry	103	691.1	48	38	.558	269	199	3.40	B
Jim	Perry	114	754.0	49	44	.527	368	219	3.39	C
Jim	Perry	131	867.0	60	38	.612	427	244	3.28	D
Jim	Perry	202	1335.0	81	81	.500	557	419	3.63	Good
Jim	Perry	245	1621.0	109	82	.571	795	463	3.33	Weak
Jim	Perry	447	2956.0	190	163	.538	1352	882	3.46	Total

 

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Gary	Bell	54	303.0	9	30	.231	192	150	5.11	A
Gary	Bell	43	280.1	11	23	.324	172	117	3.82	B
Gary	Bell	74	489.2	30	29	.508	327	200	4.01	C
Gary	Bell	61	438.1	36	13	.735	286	144	2.77	D
Gary	Bell	97	583.1	20	53	.274	364	267	4.49	Good
Gary	Bell	135	928.0	66	42	.611	613	344	3.42	Weak
Gary	Bell	232	1511.1	86	95	.475	977	611	3.84	Total

 

Perry’s winning percentage is 71 points better against second-division teams than first-division teams; Bell’s is 337 points better.   I will bow out of that debate now; I will leave it to other researchers to determine whether something real is being measured here—and, if so, what it is.

 

The Braves Three Aces

Another question I have been asked related to this deals with the three great pitchers for the Braves in the 1990s, Maddux, Glavine and Smoltz.    Maddux and Glavine were 300-game winners, but Smoltz—who was also a Hall of Famer, in my view—was the best of the three in post-season play.   “Do you think that’s because of the kind of pitcher he was?”  I have been asked.   “With more of a power mix?   Glavine was a lefty with good control and a sinker, a change, Maddux had fantastic control, great change and cutter, but Smoltz had the four-pitch mix and the high fastball.”

But actually, while all of the three had normal, mid-range Bully Factors, Tom Glavine actually had the best won-lost record of the three (although the highest ERA) against “A” quality competition in regular season, an outstanding 66-49 W and L:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Tom	Glavine	152	978.1	66	49	.574	567	362	3.90	A
Tom	Glavine	200	1268.1	77	64	.546	703	456	3.79	B
Tom	Glavine	185	1220.1	93	46	.669	730	394	3.27	C
Tom	Glavine	145	946.1	69	44	.611	607	288	3.16	D
Tom	Glavine	352	2246.2	143	113	.559	1270	818	3.84	Good
Tom	Glavine	330	2166.2	162	90	.643	1337	682	3.22	Weak
Tom	Glavine	682	4413.1	305	203	.600	2607	1500	3.54	Total
Tom	Glavine				Bully Factor:				.428
First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Greg	Maddux	159	1032.0	67	63	.515	669	226	3.85	A
Greg	Maddux	200	1337.2	87	66	.569	813	284	3.20	B
Greg	Maddux	196	1324.2	102	54	.654	920	269	2.96	C
Greg	Maddux	185	1306.2	99	43	.697	964	216	2.73	D
Greg	Maddux	359	2369.2	154	129	.544	1482	510	3.49	Good
Greg	Maddux	381	2631.1	201	97	.674	1884	485	2.85	Weak
Greg	Maddux	740	5001.0	355	226	.611	3366	995	3.15	Total
Greg	Maddux				Bully Factor:				.568
First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
John	Smoltz	89	592.1	34	34	.500	503	189	3.66	A
John	Smoltz	143	960.1	59	48	.551	866	289	3.45	B
John	Smoltz	137	917.0	56	45	.554	775	278	3.44	C
John	Smoltz	112	742.0	60	22	.732	660	204	3.09	D
John	Smoltz	232	1552.2	93	82	.531	1369	478	3.53	Good
John	Smoltz	249	1659.0	116	67	.634	1435	482	3.28	Weak
John	Smoltz	481	3211.2	209	149	.584	2804	960	3.40	Total
John	Smoltz				Bully Factor:				.421

 

 

The Leader Lists

 

These are the pitchers with the highest and lowest “Bully Factors” in our data:

 

First	Last	Highest			First	Last	Lowest
Mike	Norris	Bully Factor:	1.496		Warren	Hacker	Bully Factor:	-.470
Rob	Bell	Bully Factor:	1.351		Ed	Halicki	Bully Factor:	-.451
Virgil	Trucks	Bully Factor:	1.298		Ken	McBride	Bully Factor:	-.423
Gary	Bell	Bully Factor:	1.298		Ryan	Franklin	Bully Factor:	-.419
Ken	Kravec	Bully Factor:	1.286		Victor	Zambrano	Bully Factor:	-.413
Eric	Show	Bully Factor:	1.271		John	Dopson	Bully Factor:	-.288
Mickey	McDermott	Bully Factor:	1.230		Jason	Marquis	Bully Factor:	-.275
Bob	Turley	Bully Factor:	1.210		Scott	Kazmir	Bully Factor:	-.263
John	Buzhardt	Bully Factor:	1.206		Melido	Perez	Bully Factor:	-.246
Tom	Brewer	Bully Factor:	1.200		Francisco	Cordova	Bully Factor:	-.239

 

I realize that I haven’t shown you any of the data for pitchers with the lowest Bully Factors, so let me begin:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Warren	Hacker	39	260.0	12	20	.375	114	46	4.19	A
Warren	Hacker	30	204.1	14	12	.538	70	40	2.91	B
Warren	Hacker	30	176.1	7	17	.292	79	43	5.21	C
Warren	Hacker	7	44.0	4	2	.667	22	10	3.68	D
Warren	Hacker	69	464.1	26	32	.448	184	86	3.62	Good
Warren	Hacker	37	220.1	11	19	.367	101	53	4.90	Weak
Warren	Hacker	106	684.2	37	51	.420	285	139	4.04	Total

 

Warren Hacker was a 1950s pitcher with the Chicago Cubs.   The Cubs were the weak sister of the National League at that time (the Cubs and the Pirates), and, as Hacker didn’t get to pitch against his own team, this left him with more opportunities to pitch against the better teams.   Also, this is the era in which Retrosheet has some games but doesn’t have others, and, as it is likely that Retrosheet has more of the games of the better teams, it is likely that we are missing more games which the Cubs played some other wastrel than games when they played the good teams.   The odd result is that we have Hacker making 39 starts against “A” opponents and only 7 against “D” opponents, which is a striking thing, but. . .not sure what we can do with it.   Let’s do Ed Halicki, who is second on the list:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Ed	Halicki	45	284.2	15	19	.441	189	78	3.54	A
Ed	Halicki	42	288.1	18	14	.562	201	89	2.65	B
Ed	Halicki	34	224.2	9	17	.346	150	69	3.20	C
Ed	Halicki	36	204.0	12	15	.444	125	67	5.07	D
Ed	Halicki	87	573.0	33	33	.500	390	167	3.09	Good
Ed	Halicki	70	428.2	21	32	.396	275	136	4.09	Weak
Ed	Halicki	157	1001.2	54	65	.454	665	303	3.52	Total

 

Halicki is more interesting.   Halicki pitched better, against good teams, than. . .well, better than Steve McCatty, Mike McCormick, Jack McDowell, Sam McDowell, Denny McLain and Dave McNally, sticking to the “Mc” part of the list and the best pitchers in that era.   He pitched better against good teams than 96% of the pitchers in our data.   Not many of the Hall of Famers had a 3.09 ERA against first-division opponents.  Halicki just did not take care of business against bad teams.   Three active pitchers have shown the same tendency:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Scott	Kazmir	44	245.1	14	14	.500	246	122	3.45	A
Scott	Kazmir	36	219.0	16	12	.571	229	83	3.12	B
Scott	Kazmir	41	249.1	18	9	.667	264	98	3.32	C
Scott	Kazmir	29	153.2	9	11	.450	157	86	6.27	D
Scott	Kazmir	80	464.1	30	26	.536	475	205	3.30	Good
Scott	Kazmir	70	403.0	27	20	.574	421	184	4.44	Weak
Scott	Kazmir	150	867.1	57	46	.553	896	389	3.83	Total

 

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Jason	Marquis	29	178.2	13	10	.565	106	61	4.63	A
Jason	Marquis	65	405.1	26	20	.565	224	149	3.75	B
Jason	Marquis	68	410.1	27	27	.500	227	169	4.98	C
Jason	Marquis	69	404.0	26	26	.500	249	159	4.37	D
Jason	Marquis	94	584.0	39	30	.565	330	210	4.02	Good
Jason	Marquis	137	814.1	53	53	.500	476	328	4.67	Weak
Jason	Marquis	231	1398.1	92	83	.526	806	538	4.40	Total

 

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Ryan	Franklin	39	247.0	12	16	.429	124	76	3.90	A
Ryan	Franklin	21	129.2	7	8	.467	58	43	4.93	B
Ryan	Franklin	17	100.0	2	13	.133	52	32	6.57	C
Ryan	Franklin	29	191.1	6	12	.333	96	45	3.72	D
Ryan	Franklin	60	376.2	19	24	.442	182	119	4.25	Good
Ryan	Franklin	46	291.1	8	25	.242	148	77	4.70	Weak
Ryan	Franklin	106	668.0	27	49	.355	330	196	4.45	Total

 

This is data as a starting pitcher only; I don’t know what Franklin’s splits are as a reliever.   But you know, now that it comes up, I remember that, when he was starting for the Mariners, we’d see Franklin on the schedule and kind of pencil that one in as a game we ought to win, but then he’d beat us.     How does a guy go 12-16 with a 3.90 ERA against the best teams in the league, but then go 8-25 with a 4.70 ERA against the second division?

 

But Franklin made 106 starts in his career, and maybe it’s not enough for everything to even out.    When you have a list of 702 players and you rank them one through 702, what will almost always happen is that those who barely have enough playing time to qualify for the list will dominate the top and bottom of the chart.   That’s what happens here; most of our “extreme” pitchers are guys who had short careers.   This is a list of the most extreme Bully Factors for pitchers with 250 or more starts:

 

First	Last	Highest			First	Last	Lowest
Denny	McLain	Bully Factor:	1.113		Rick	Wise	Bully Factor:	-.172
C.C.	Sabathia	Bully Factor:	1.074		Bob	Knepper	Bully Factor:	-.073
Greg	Swindell	Bully Factor:	1.024		Rick	Mahler	Bully Factor:	-.056
Early	Wynn	Bully Factor:	1.020		Bob	Purkey	Bully Factor:	-.033
Chris	Carpenter	Bully Factor:	1.010		Buddy	Black	Bully Factor:	.005
Andy	Messersmith	Bully Factor:	.972		Ed	Whitson	Bully Factor:	.011
Geoff	Zahn	Bully Factor:	.969		Tom	Browning	Bully Factor:	.052
Russ	Ortiz	Bully Factor:	.956		Frank	Lary	Bully Factor:	.059
Jack	Morris	Bully Factor:	.937		Mike	Krukow	Bully Factor:	.065
Dock	Ellis	Bully Factor:	.896		Pat	Dobson	Bully Factor:	.077
Bruce	Hurst	Bully Factor:	.892		John	Smiley	Bully Factor:	.090
Al	Downing	Bully Factor:	.869		Danny	Darwin	Bully Factor:	.119

 

A few splits:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Denny	McLain	65	399.2	20	34	.370	247	140	4.82	A
Denny	McLain	54	372.1	26	16	.619	244	92	3.48	B
Denny	McLain	73	548.0	40	23	.635	392	145	2.69	C
Denny	McLain	72	528.1	42	16	.724	363	162	2.96	D
Denny	McLain	119	772.0	46	50	.479	491	232	4.17	Good
Denny	McLain	145	1076.1	82	39	.678	755	307	2.83	Weak
Denny	McLain	264	1848.1	128	89	.590	1246	539	3.39	Total

 

You may remember that in 1968 McLain won 31 games, but was not notably effective against the Cardinals in the World Series, and ceded his spot as the Tigers’ #1 pitcher to Mickey Lolich.   Lolich had a very low Bully Factor (.375), and a career ERA of 3.37 in 114 starts against “A” quality opponents.   Sabathia, without comment:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
C.C.	Sabathia	67	425.1	18	30	.375	321	146	4.17	A
C.C.	Sabathia	63	411.1	27	22	.551	373	137	3.92	B
C.C.	Sabathia	57	370.2	30	15	.667	312	126	3.86	C
C.C.	Sabathia	101	682.0	61	14	.813	584	181	2.97	D
C.C.	Sabathia	130	836.2	45	52	.464	694	283	4.04	Good
C.C.	Sabathia	158	1052.2	91	29	.758	896	307	3.28	Weak
C.C.	Sabathia	288	1889.1	136	81	.627	1590	590	3.62	Total

 

And Rick Wise:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Rick	Wise	106	702.1	41	41	.500	351	200	3.49	A
Rick	Wise	129	884.2	52	51	.505	451	204	3.48	B
Rick	Wise	123	770.0	39	54	.419	390	204	4.30	C
Rick	Wise	97	652.2	47	34	.580	383	163	3.53	D
Rick	Wise	235	1587.0	93	92	.503	802	404	3.48	Good
Rick	Wise	220	1422.2	86	88	.494	773	367	3.95	Weak
Rick	Wise	455	3009.2	179	180	.499	1575	771	3.70	Total

 

In a long career, Rick Wise pitched better against good teams than weak teams—an ERA half a run better, with 1400 innings each way.   Frank Lary, who is also on our list of pitchers who pitched well against good teams, was actually most famous, while active, for his ability to beat the Yankees, so he was actually known for this tendency. 

 

I tried to study the question of whether there was any type of pitcher who tended to have a high or low Bully Factor.   I thought that pitchers who have good stuff but marginal control (like Bob Turley) might tend to dominate poor teams, but struggle against good ones.    The data doesn’t really show that.    I looked at whether power pitchers or control pitchers do well against strong opposition, or left-handers.  I didn’t really find anything.   The only thing is that good pitchers do have a slightly higher Bully Factors than lower-quality pitchers, but that’s probably a natural effect from magnifying their advantage against weaker teams, or stating it on a larger scale.    Certain knuckleball pitchers (Charlie Hough and Tim Wakefield) have very high Bully Factors, but others (Phil Niekro and Tom Candiotti) have very low ones.    There are no obvious patterns.

 

The Dominators

 

            OK, let us get to one of the really big questions.    Who are the best pitchers ever against top quality competition?

            Probably the best guy ever. …best guy in the data. ..is Bret Saberhagen.    Saberhagen, Sandy Koufax, Whitey Ford:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Bret	Saberhagen	85	562.1	39	18	.684	390	104	3.15	A
Bret	Saberhagen	94	619.1	34	37	.479	425	130	4.08	B
Bret	Saberhagen	97	667.1	47	28	.627	412	110	2.99	C
Bret	Saberhagen	95	642.2	42	31	.575	454	109	3.21	D
Bret	Saberhagen	179	1181.2	73	55	.570	815	234	3.64	Good
Bret	Saberhagen	192	1310.0	89	59	.601	866	219	3.10	Weak
Bret	Saberhagen	371	2491.2	162	114	.587	1681	453	3.36	Total

 

            Saberhagen’s 39-18 record is the best we have against “A” quality competition.   The other guys’ ERAs are a little better, but that’s probably a context thing.     Sandy Koufax in the 1960s was totally dominant against all levels of competition--30-13 with a 2.46 ERA against “A” quality competitors from 1960 to 1966.  I would guess that it might be even better than that if we focused on the period 1963-1966:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Sandy	Koufax	62	445.2	33	16	.673	412	130	2.56	A
Sandy	Koufax	90	615.2	39	28	.582	631	200	2.92	B
Sandy	Koufax	71	452.0	30	20	.600	447	182	3.35	C
Sandy	Koufax	91	689.0	57	21	.731	784	229	2.19	D
Sandy	Koufax	152	1061.1	72	44	.621	1043	330	2.77	Good
Sandy	Koufax	162	1141.0	87	41	.680	1231	411	2.65	Weak
Sandy	Koufax	314	2202.1	159	85	.652	2274	741	2.71	Total

 

            Whitey Ford had a career ERA of 2.74, but in the data we have it is 2.72—2.72 against good teams, 2.72 against second-division teams:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Whitey	Ford	107	729.2	51	29	.637	456	238	2.55	A
Whitey	Ford	92	637.0	37	29	.561	363	221	2.91	B
Whitey	Ford	108	711.0	54	21	.720	440	210	2.75	C
Whitey	Ford	118	864.2	75	20	.789	553	305	2.70	D
Whitey	Ford	199	1366.2	88	58	.603	819	459	2.72	Good
Whitey	Ford	226	1575.2	129	41	.759	993	515	2.72	Weak
Whitey	Ford	425	2942.1	217	99	.687	1812	974	2.72	Total

 

            Ford, Koufax and Saberhagen stand out as the pitchers who excelled against championship-quality opposition.   Zack Greinke, on the mound in a Royals uniform, reminds us all of Bret Saberhagen—but Greinke hasn’t done it against the best teams:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Zack	Greinke	32	185.0	6	18	.250	131	57	4.28	A
Zack	Greinke	38	236.0	13	15	.464	207	67	3.55	B
Zack	Greinke	36	218.2	13	14	.481	187	51	4.28	C
Zack	Greinke	30	188.2	13	5	.722	165	32	2.81	D
Zack	Greinke	70	421.0	19	33	.365	338	124	3.87	Good
Zack	Greinke	66	407.1	26	19	.578	352	83	3.60	Weak
Zack	Greinke	136	828.1	45	52	.464	690	207	3.74	Total

 

 

Fun Stats

 

            Fun Stats are off-balance numbers that result from not-very-good pitchers pitching against good teams, or outstanding pitchers pitching against bad teams.   Justin Verlander against “D” quality competition:

 

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Justin	Verlander	33	215.2	21	2	.913	203	72	2.55	D

 

            And here’s a few others:

 

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Johan	Santana	75	500.0	47	12	.797	523	124	2.93	D
Roy	Halladay	78	554.2	50	13	.794	383	96	3.03	D
Roger	Clemens	194	1357.0	118	31	.792	1317	392	2.62	D
Jim	Palmer	143	1098.0	94	25	.790	618	361	2.30	D
Mike	Norris	39	274.1	22	6	.786	153	114	2.30	D
Rich	Harden	33	200.1	18	5	.783	205	75	2.61	D
Virgil	Trucks	44	336.2	28	8	.778	177	111	2.54	D
Steve	Busby	35	248.1	21	6	.778	169	91	3.30	D

 

            And, on the other end:

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Russ	Kemmerer	34	183.0	3	21	.125	91	88	5.90	A
Dave	Freisleben	36	203.2	3	20	.130	101	101	4.82	A
Joe	Decker	18	85.1	2	13	.133	57	58	8.12	A
Skip	Lockwood	26	157.1	3	18	.143	92	73	3.83	A
Bryan	Rekar	33	183.2	3	18	.143	109	59	6.17	A
Ian	Snell	19	98.1	2	10	.167	86	51	6.13	A
Shawn	Boskie	33	177.1	4	20	.167	82	62	6.09	A
Moe	Drabowsky	29	158.0	4	18	.182	97	81	5.58	A
Bruce	Chen	34	181.0	4	17	.190	134	72	5.07	A
Paul	Wilson	42	254.0	5	21	.192	175	109	5.78	A
Mike	Smithson	58	323.1	7	29	.194	170	101	5.96	A
Jeff	Fassero	44	267.2	7	28	.200	225	118	5.28	A

 

           Here is a link to a spreadsheet with the full list of the pitchers with 100 or more starts, so you can download that and do with it what you will.

 

            And finally, we circle back to the question of Bob Gibson versus Juan Marichal.   It turns out I was right; it was bullshit to say that Marichal fattened up on weak opponents, while Gibson was better against the best teams.   In fact, Gibson was more of a Bully than was Marichal:

 

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Bob	Gibson	96	737.2	41	46	.471	593	254	3.33	A
Bob	Gibson	147	1142.0	63	61	.508	911	405	3.24	B
Bob	Gibson	105	852.1	60	27	.690	689	252	2.34	C
Bob	Gibson	134	1064.1	81	36	.692	851	366	2.62	D
Bob	Gibson	243	1879.2	104	107	.493	1504	659	3.28	Good
Bob	Gibson	239	1916.2	141	63	.691	1540	618	2.50	Weak
Bob	Gibson	482	3796.1	245	170	.590	3044	1277	2.88	Total
Bob	Gibson					Bully Factor:			.667

 

 

            Gibson’s Bully Factor was well above average.   Marichal’s was well below average.   Gibson’s winning percentages against different qualities of opponents descend from .692 against the weakest opponents to .471 against the best.   Marichal goes from .667 to .576:

 

 

First	Last	G	IP	W	L	Pct	SO	BB	ERA	Group
Juan	Marichal	100	747.2	49	36	.576	497	133	2.84	A
Juan	Marichal	126	961.0	66	39	.629	617	225	3.17	B
Juan	Marichal	116	892.1	59	33	.641	582	181	2.89	C
Juan	Marichal	115	875.1	64	32	.667	588	162	2.63	D
Juan	Marichal	226	1708.2	115	75	.605	1114	358	3.03	Good
Juan	Marichal	231	1767.2	123	65	.654	1170	343	2.76	Weak
Juan	Marichal	457	3476.1	238	140	.630	2284	701	2.89	Total
Juan	Marichal					Bully Factor:			.373

 

            Marichal was more effective against first-division opponents than was Gibson—or almost anybody else; he was one of the most effective pitchers against first-division teams in our data.    Gibson was less effective against contenders, but made up some of the difference against the weaker teams.   Ah, if only I knew where Tony Bandle was today.