Before I could get back to work on the Omars and Ozzies, I got the following letter in "Hey, Bill":

 

 

Could you please include George McBride & Everett Scott in your Omars and Ozzies article?  They both barely miss your cutoff of 1700 games but they are just the type of shortstops that fit the profile otherwise.  Both played over 1600 games at SS and both had a career OBP of .281.  Both had outstanding conventional fielding stats. McBride led the AL in DPs 6 times and Fielding AVG 6 times.  Scott led the AL in Fielding AVG 8 straight years.

 

--Mark R.

 

 

            OK, but what am I going to say to the people who think I should include Frankie Crosetti, Marty Marion, Bud Harrelson, Art Fletcher and Freddie Patek?  In a weak moment I decided to add McBride and Scott to the study, because otherwise we wouldn’t have had a single player in the study named "Everett", and. ..well, I enjoy having the kind of readers who care about George McBride and Everett Scott.    So we’re now up to 32 shortstops—Posnanski’s number—and we’ll have eight in this group, leaving us three groups of six to get through.   McBride, chronologically, should have come before Joe Tinker.

 

GEORGE MC BRIDE
Yr	City	Lg	Age	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1901	Milwaukee	AL	20	3	12	0	2	0	0	0	0	1	0	.167	.167	.231	.397
1905	Pittsburgh	NL	24	27	87	9	19	4	0	0	7	6	2	.218	.264	.277	.541
1905	St. Louis	NL	24	81	281	22	61	1	2	2	34	14	10	.217	.256	.264	.520
1906	St. Louis	NL	25	90	313	24	53	8	2	0	13	17	5	.169	.208	.215	.422
1908	Washington	AL	27	155	518	47	120	10	6	0	34	41	12	.232	.274	.292	.566
1909	Washington	AL	28	156	504	38	118	16	0	0	34	36	17	.234	.266	.294	.560
1910	Washington	AL	29	154	514	54	118	19	4	1	55	61	11	.230	.288	.321	.609
1911	Washington	AL	30	154	557	58	131	11	4	0	59	52	15	.235	.269	.312	.581
1912	Washington	AL	31	152	521	56	118	13	7	1	52	38	17	.226	.284	.288	.572
1913	Washington	AL	32	150	499	52	107	18	7	1	52	43	12	.214	.285	.286	.571
1914	Washington	AL	33	156	503	49	102	12	4	0	24	43	12	.203	.243	.274	.516
1915	Washington	AL	34	146	476	54	97	8	6	1	30	29	10	.204	.252	.251	.503
1916	Washington	AL	35	139	466	36	106	15	4	1	36	23	8	.227	.283	.271	.555
1917	Washington	AL	36	50	141	6	27	3	0	0	9	10	1	.191	.213	.265	.477
1918	Washington	AL	37	18	53	2	7	0	0	0	1	0	1	.132	.132	.132	.264
1919	Washington	AL	38	15	40	3	8	1	1	0	4	3	0	.200	.275	.256	.531
1920	Washington	AL	39	13	41	6	9	1	0	0	3	2	0	.220	.244	.256	.500
Totals				1659	5526	516	1203	140	47	7	447	419	133	.218	.264	.281	.544

 

 

GEORGE MC BRIDE or, if you prefer, McBRIDE
Career Won-Lost Contribution:	166-209			Offensive Winning Percentage:	.268
Offensive W-L Contribution:	70-190			Defensive Winning Percentage:	.841
Defensive W-L Contribution:	96-18			Defensive WPct, Up to Age 30:	.725
Team Success Percentage:	.491			Defensive WPct, After 30:	1.016
Estimated Innings at Short:	14392			Best Season:	1910
Career Putouts at Shortstop:	3585				20-15
Expected Putouts at Short:	3362	+223		Career 20-Win Seasons:	1
Career Assists at Shortstop:	5274			.800 Seasons (as a regular):	0
Expected Assists at Short:	5117	+157		.700 Seasons (as a regular):	0
Career Double Plays a/Short:	610			.600 Seasons (as a regular):	0
Expected DPs at Shortstop:	580	+30		.500 Seasons (as a regular):	4
Career Errors at Shortstop:	484			Sub-.500 seasons a/regular:	7
Expected Errors at Short:	641	+157		Career Value:	144

 

            But actually, now that I’ve done the work, McBride and Scott don’t quite fit the profile of this group.  Although both were brilliant defensive players, their careers ended a little bit before the 1700 game mark because both were terrible hitters, which reduced their overall value enough to shorten their careers.    Both McBride and Scott were worse hitters than anyone else we have seen yet.   Mark Belanger sort of defines "bad hitting shortstop" for my generation, and both were far worse hitters than Belanger.   Omar Vizquel and Ozzie Smith were not bad hitters; they’re not great hitters, but they’re not bad.   The players who define this group—Aparicio, Joe Tinker, Herman Long, Dave Bancroft—are not bad hitters.   McBride and Scott were more in the Hal Lanier/Dal Maxvill/Rey Sanchez group—the guys who really couldn’t hit.  We have one other guy later on who was a worse hitter than McBride and Scott, but among the 32 players now in the group they rank 30^th and 31^st with the bat.   But McBride was a sensational fielder, and Scott was better than McBride was.

 

 

Mickey Doolan
Yr	City	Lg	Age	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1905	Philadelphia	NL	25	136	492	53	125	27	11	1	48	24	17	.254	.360	.292	.651
1906	Philadelphia	NL	26	154	535	41	123	19	7	1	55	27	16	.230	.297	.270	.567
1907	Philadelphia	NL	27	145	509	33	104	19	7	1	47	25	18	.204	.275	.243	.518
1908	Philadelphia	NL	28	129	445	29	104	25	4	2	49	17	5	.234	.321	.267	.588
1909	Philadelphia	NL	29	147	493	39	108	12	10	1	35	37	10	.219	.290	.276	.566
1910	Philadelphia	NL	30	148	536	58	141	31	6	2	57	35	16	.263	.354	.315	.670
1911	Philadelphia	NL	31	146	512	51	122	23	6	1	49	44	14	.238	.313	.301	.614
1912	Philadelphia	NL	32	146	532	47	137	26	6	1	62	34	6	.258	.335	.302	.637
1913	Philadelphia	NL	33	151	518	32	113	12	6	1	43	29	17	.218	.270	.262	.533
1914	Baltimore	FL	34	145	486	58	119	23	6	1	53	40	30	.245	.323	.311	.634
1915	Baltimore	FL	35	119	404	41	75	13	7	2	21	24	10	.186	.267	.238	.506
1915	Chicago	FL	35	24	86	9	23	1	1	0	9	2	5	.267	.302	.292	.594
1916	New York	NL	36	18	51	4	12	3	1	1	3	2	1	.235	.392	.264	.656
1916	Chicago	NL	36	28	70	4	15	2	1	0	5	8	0	.214	.271	.295	.566
1918	Brooklyn	NL	38	92	308	14	55	8	2	0	18	22	8	.179	.218	.233	.451
Totals				1728	5977	513	1376	244	81	15	554	370	173	.230	.306	.279	.584

 

 

 

MICKEY DOOLAN
Career Won-Lost Contribution:	188-208			Offensive Winning Percentage:	.318
Offensive W-L Contribution:	88-189			Defensive Winning Percentage:	.836
Defensive W-L Contribution:	100-20			Defensive WPct, Up to Age 30:	.815
Team Success Percentage:	.562			Defensive WPct, After 30:	.858
Estimated Innings at Short:	14465			Best Season:	1910
Career Putouts at Shortstop:	3279				20-12
Expected Putouts at Short:	3194	+85		Career 20-Win Seasons:	2
Career Assists at Shortstop:	4858			.800 Seasons (as a regular):	0
Expected Assists at Short:	4591	+267		.700 Seasons (as a regular):	0
Career Double Plays a/Short:	519			.600 Seasons (as a regular):	1
Expected DPs at Shortstop:	507	+12		.500 Seasons (as a regular):	5
Career Errors at Shortstop:	592			Sub-.500 seasons a/regular:	7
Expected Errors at Short:	682	+90		Career Value:	177

 

            Doolan was a light hitter in the tradition of Scott and McBride, although even Doolan—whose career OPS was .584—was not as bad a hitter as those two.   I think Doolan was a college teammate of Christy Mathewson’s; anyway they were both born in 1880 and they both attended Bucknell, so I would assume they were teammates.

            Do you guys use Lee Sinins’ Complete Baseball Encyclopedia?   I use it all the time, although I have an antique version of it when it used to be called the Sabermetric Encyclopedia.  In doing this series of reports I probably had to look up Mickey Doolan in the Sabermetric Encyclopedia 15 to 20 times, looking up some factoid or another. . .each player in here; I’m sure I looked up Donie Bush 15 to 20 times, and Germany Smith.

            Anyway, you can find players in the Sabermetric Encyclopedia by using truncated versions of their names, and. . .stupid observation of the week. . .I began looking these guys up by truncating their names.    Mickey Doolan, for example, can be found by typing in "Ickey Doo", which somehow is a lot more fun than looking up Mickey Doolan.   Germany Smith can be found by typing in "Ermany Smi", Donie Bush can be found by typing in "Onie Bu", and even George McBride can be found by typing in "Orge McB".   Ah, the wonders of a good search engine. . ..

            Ickey Doo, who was hardly a good hitter, finished 14^th in the NL MVP voting in 1911, the first year there was such an award, and 13^th in 1913, when he hit just .218 and scored just 32 runs (!!) in 151 games.   That’s actually the third-fewest runs ever scored in 150 or more games—a fact which I found by using the Sabermetric Encyclopedia.   Doolan was past 30 and his best years were behind him before the award started, and the fact that he did show up in the voting certainly testifies to his defensive excellence.

            In fact, we could generalize from this that "the voters" placed much more value on Doolan’s D than we would; I, after all, have Doolan’s won-lost contributions at 15-17 in 1911 and 13-20 in 1913, which wouldn’t place him among the league’s top 30 players either year.     It’s a dangerous inference.   Doolan’s performance in MVP voting probably only really means that one or two voters, out of a panel of 8, had some belief in his value.   We can say that "the MVP voters valued his defense more highly than modern analysts", but this might be incorrect for about 7 of the 8 voters.   George McBride, who was also born in 1880 (like Doolan and Christy Mathewson) was also mentioned in the MVP voting in 1911, 1912, 1913 and 1914.   Most of the players in this study were mentioned in MVP voting multiple times in their careers, if there was an MVP vote in their league.

 

 

 

DONIE BUSH
YR	City	Lg	Ag	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1908	Detroit	AL	20	20	68	13	20	1	1	0	4	7	2	.294	.338	.360	.698
1909	Detroit	AL	21	157	532	114	145	18	2	0	33	88	53	.273	.314	.380	.694
1910	Detroit	AL	22	142	496	90	130	13	4	3	34	78	49	.262	.323	.365	.687
1911	Detroit	AL	23	150	561	126	130	18	5	1	36	98	40	.232	.287	.349	.636
1912	Detroit	AL	24	144	511	107	118	14	8	2	38	117	35	.231	.301	.377	.679
1913	Detroit	AL	25	153	597	98	150	19	10	1	40	80	44	.251	.322	.344	.665
1914	Detroit	AL	26	157	596	97	150	18	4	0	32	112	35	.252	.295	.373	.668
1915	Detroit	AL	27	155	561	99	128	12	8	1	44	118	35	.228	.283	.364	.648
1916	Detroit	AL	28	145	550	73	124	5	9	0	34	75	19	.225	.267	.319	.587
1917	Detroit	AL	29	147	581	112	163	18	3	0	24	80	34	.281	.322	.370	.691
1918	Detroit	AL	30	128	500	74	117	10	3	0	22	79	9	.234	.266	.340	.606
1919	Detroit	AL	31	129	509	82	124	11	6	0	26	75	22	.244	.289	.343	.632
1920	Detroit	AL	32	141	506	85	133	18	5	1	33	73	15	.263	.324	.357	.681
1921	Detroit	AL	33	104	402	72	113	6	5	0	27	45	8	.281	.321	.355	.676
1921	Washington	AL	33	23	84	15	18	1	0	0	2	12	2	.214	.226	.313	.539
1922	Washington	AL	34	41	134	17	32	4	1	0	7	21	1	.239	.284	.342	.626
1923	Washington	AL	35	10	22	6	9	0	0	0	0	0	0	.409	.409	.409	.818
Totals				1946	7210	1280	1804	186	74	9	436	1158	403	.250	.300	.356	.656

 

 

 

DONIE BUSH
Career Won-Lost Contribution:	242-231			Offensive Winning Percentage:	.483
Offensive W-L Contribution:	161-173			Defensive Winning Percentage:	.581
Defensive W-L Contribution:	81-58			Defensive WPct, Up to Age 30:	.626
Team Success Percentage:	.492			Defensive WPct, After 30:	.416
Estimated Innings at Short:	16561			Best Season:	1909
Career Putouts at Shortstop:	4038				26-11
Expected Putouts at Short:	3990	+48		Career 20-Win Seasons:	6
Career Assists at Shortstop:	6119			.800 Seasons (as a regular):	0
Expected Assists at Short:	6018	+101		.700 Seasons (as a regular):	1
Career Double Plays a/Short:	585			.600 Seasons (as a regular):	2
Expected DPs at Shortstop:	755	-170		.500 Seasons (as a regular):	6
Career Errors at Shortstop:	689			Sub-.500 seasons a/regular:	7
Expected Errors at Short:	724	+35		Career Value:	247

 

 

            Donie Bush, the shortstop for the Ty Cobb Tigers from 1908 to 1921, was different from most of the shortstops in this group in two respects:  1, that he walked a great deal, and 2, that he was hopeless at turning the double play.     Bush, years after his career, was the co-owner of the Louisville team that sold Pee Wee Reese to the Dodgers.    Telling this story from memory, but. . .I think the co-owners of the Louisville American Association franchise were Bush, Tom Yawkey, and a third man; Bush managed the team as well as being one of the owners.   When the team came up with Pee Wee Reese in 1938, Yawkey assumed that the Red Sox would be able to buy him cheap because he was co-owner of the team, but Bush and the third owner  thought "Wait a minute.   We have here a young player who is worth more than the franchise is.   Why should we give him to Boston, just because Yawkey is a co-owner?"    They sold him to the Dodgers, greatly annoying Yawkey.

            Fred Haney used to tell a story about when he came up with the Tigers; he hit .352 as a rookie in 1922, 213 at bats.   Ty Cobb was his manager.  His story was that one time he decided to sit on a fastball and take big rip at it, and he hit a single, but when he got back to the bench Ty Cobb tore him a new one.   He wasn’t up there to rip the ball; that was Harry Heilmann’s job, and Veach’s and Cobb’s.   His job was to get on base.   Take pitches, defend the plate, slap the ball, get on base.

            The Tigers had a group of players like that—Bassler, Bush, Topper Rigney, even Lu Blue, the first baseman; their job was to get on base and let the big boys drive them in.  That’s a very different philosophy than that which prevails in modern baseball, in which everyone is expected to drive the ball.   I am asked frequently what kind of market inefficiencies there might be remaining in baseball, which could be exploited by cash-strapped teams to gain an advantage.   There are 50 or 100 obvious market inefficiencies in baseball, but this is one of the largest.  Defend the plate, slap the ball, get on base. . .nobody does that stuff in modern baseball.   The scouts don’t respect it, the organizations don’t encourage it, the managers don’t insist on it.  "Bat speed" is the deal now; everybody is supposed to have bat speed.   Everybody is supposed to drive the ball.   People overlook the fact that you still CAN slap and punch the ball and get on base, if you focus on that.

            Bush’s high walk totals gave him a certain offensive value, but his extremely poor double play numbers offset much of that value.   Bush’s double play totals are so poor that one suspects it must represent a deliberate decision—that is, that Bush must have been instructed by an early manager not to worry about the double play, but just to focus on getting the out at second base.

            That was one thing when Bush came up.   From 1908 to 1911 there were 770 to 809 double plays per season in the American League—about 100 double plays per team per season--and many of those were caught stealing/throw ‘em out double plays or runners thrown out by outfielders.  Catchers, in 1911, participated in more than 20% of the American League’s double plays.

            But over the course of Bush’s career double play totals increased markedly, from 770 in 1911 to 1,136 in 1922, and the relevant increase is much greater than that, because double plays by catchers and outfielders were not only not increasing, but actually decreasing.    The modern double play, the 4-6-3 or 6-4-3 double play, was becoming substantially more important, making Bush’s almost comical inability to turn the DP constantly more central to his game.   Bush in his career participated in only 46 double plays per 1,300 innings, while Peckinpaugh—coming to the majors two years later—participated in 71, and Maranville, who came up two years after Peckinpaugh, participated in 80 per 1,300 innings.    Bush’s DP/inning ratio is the lowest of any player in this study.

 

 

ROGER PECKINPAUGH
Yr	City	Lg	Age	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1910	Cleveland	AL	19	15	45	1	9	0	0	0	6	1	3	.200	.200	.234	.434
1912	Cleveland	AL	21	69	236	18	50	4	1	1	22	16	11	.212	.250	.262	.512
1913	Cleveland	AL	22	1	0	1	0	0	0	0	0	0	0	.000	.000	.000	.000
1913	New York	AL	22	95	340	35	91	10	7	1	32	24	19	.268	.347	.316	.663
1914	New York	AL	23	157	570	55	127	14	6	3	51	51	38	.223	.284	.288	.572
1915	New York	AL	24	142	540	67	119	18	7	5	44	49	19	.220	.307	.289	.596
1916	New York	AL	25	146	552	65	141	22	8	4	58	62	18	.255	.346	.332	.678
1917	New York	AL	26	148	543	63	141	24	7	0	41	64	17	.260	.330	.340	.670
1918	New York	AL	27	122	446	59	103	15	3	0	43	43	12	.231	.278	.303	.581
1919	New York	AL	28	122	453	89	138	20	2	7	33	59	10	.305	.404	.390	.794
1920	New York	AL	29	139	534	109	144	26	6	8	54	72	8	.270	.386	.356	.742
1921	New York	AL	30	149	577	128	166	25	7	8	71	84	2	.288	.397	.380	.777
1922	Washington	AL	31	147	520	62	132	14	4	2	48	55	11	.254	.308	.329	.636
1923	Washington	AL	32	154	568	73	150	18	4	2	62	64	10	.264	.320	.340	.660
1924	Washington	AL	33	155	523	72	142	20	5	2	73	72	11	.272	.340	.360	.700
1925	Washington	AL	34	126	422	67	124	16	4	4	64	49	13	.294	.379	.367	.746
1926	Washington	AL	35	57	147	19	35	4	1	1	14	28	3	.238	.299	.360	.659
1927	Chicago	AL	36	68	217	23	64	6	3	0	23	21	2	.295	.350	.360	.710
Totals				2012	7233	1006	1876	256	75	48	739	814	207	.259	.335	.336	.672

 

 

 

ROGER PECKINPAUGH
Career Won-Lost Contribution:	252-211			Offensive Winning Percentage:	.430
Offensive W-L Contribution:	140-186			Defensive Winning Percentage:	.817
Defensive W-L Contribution:	112-25			Defensive WPct, Up to Age 30:	.790
Team Success Percentage:	.647			Defensive WPct, After 30:	.871
Estimated Innings at Short:	17464			Best Season:	1919
Career Putouts at Shortstop:	3919				22-5
Expected Putouts at Short:	4078	-159		Career 20-Win Seasons:	6
Career Assists at Shortstop:	6337			.800 Seasons (as a regular):	1
Expected Assists at Short:	6110	+227		.700 Seasons (as a regular):	1
Career Double Plays a/Short:	952			.600 Seasons (as a regular):	3
Expected DPs at Shortstop:	805	+147		.500 Seasons (as a regular):	8
Career Errors at Shortstop:	553			Sub-.500 seasons a/regular:	5
Expected Errors at Short:	646	+93		Career Value:	272

 

 

            In the analytical box I’ve broken players’ defense down into Defensive Winning Percentage up to age 30, and Defensive Winning Percentage after age 30.    I’m interested in studying defensive aging, and it makes a clear point; I believe that 27 of the 32 shortstops in the study have higher defensive winning percentages before age 30 than after age 30, generally by a pretty predictable margin.

            There are a handful of shortstops, however, who have higher defensive winning percentages after age 30 than before, including three in this group, (Peckinpaugh, Doolan and McBride.)  We shouldn’t be too concerned about that; there’s an occasional hitter who has his best years after age 30, probably there’s an occasional fielder as well.   But I would attribute these fluke cases to four causes:

            1)  Conditioning variables,

            2)  Error tendencies,

            3)  Hitting abilities, and

            4)  Measurement error.

            The error tendencies are relevant in this way:  that fielding percentages predictably improve as a player ages.   If a young shortstop is particularly error-prone, like Garry Templeton for example, then it is predictable that, as he ages, this will become less of an issue, and that type of a defensive player may be more likely to have his best defensive years later in his career.

            Hitting creates a "performance floor" for the player.   Hitting helps define the player’s relationship to replacement level.   If a player is a good-hitting shortstop, he may play regularly until he is 37, 38 years old (or more), and has lost a couple of steps in the field.   If he is not such a good hitter, he is likely to be forced out of the game at a younger age, thus less likely to have an "aging drag" on his overall fielding record.

            And, of course, our measurements are not perfect.   I believe in my system; I believe in my results—but we have to be realistic.   While we may have very accurate measurements of a player’s overall value, when you divide that record into two parts (offense and defense) then you have a smaller base, and when you have a smaller base you have less accurate measurements.   Defense is the smaller of the two parts, even for shortstops. 

            Then we divide defense into two parts—defense up to age 30, defense after age 30.   Defense after age 30 is the smaller of those two.   By this time the base is so small that we’re vulnerable to flukes in the data.   It may be true that Peckinpaugh’s defense did improve as he aged, or it may be that there’s a fluke in the data.  

 

 

 

RABBIT MARANVILLE
Yr	City	Lg	Age	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1912	Boston	NL	20	26	86	8	18	2	0	0	8	9	1	.209	.233	.292	.524
1913	Boston	NL	21	143	571	68	141	13	8	2	48	68	25	.247	.308	.330	.638
1914	Boston	NL	22	156	586	74	144	23	6	4	78	45	28	.246	.326	.306	.632
1915	Boston	NL	23	149	509	51	124	23	6	2	43	45	18	.244	.324	.308	.632
1916	Boston	NL	24	155	604	79	142	16	13	4	38	50	32	.235	.325	.296	.620
1917	Boston	NL	25	142	561	69	146	19	13	3	43	40	27	.260	.357	.312	.668
1918	Boston	NL	26	11	38	3	12	0	1	0	3	4	0	.316	.368	.381	.749
1919	Boston	NL	27	131	480	44	128	18	10	5	43	36	12	.267	.377	.319	.696
1920	Boston	NL	28	134	493	48	131	19	15	1	43	28	14	.266	.371	.305	.676
1921	Pittsburgh	NL	29	153	612	90	180	25	12	1	70	47	25	.294	.379	.347	.727
1922	Pittsburgh	NL	30	155	672	115	198	26	15	0	63	61	24	.295	.378	.355	.733
1923	Pittsburgh	NL	31	141	581	78	161	19	9	1	41	42	14	.277	.346	.327	.673
1924	Pittsburgh	NL	32	152	594	62	158	33	20	2	71	35	18	.266	.399	.307	.706
1925	Chicago	NL	33	75	266	37	62	10	3	0	23	29	6	.233	.293	.308	.602
1926	Brooklyn	NL	34	78	234	32	55	8	5	0	24	26	7	.235	.312	.312	.624
1927	St. Louis	NL	35	9	29	0	7	1	0	0	0	2	0	.241	.276	.290	.566
1928	St. Louis	NL	36	112	366	40	88	14	10	1	34	36	3	.240	.342	.310	.652
1929	Boston	NL	37	146	560	87	159	26	10	0	55	47	13	.284	.366	.344	.710
1930	Boston	NL	38	142	558	85	157	26	8	2	43	48	9	.281	.367	.344	.711
1931	Boston	NL	39	145	562	69	146	22	5	0	33	56	9	.260	.317	.329	.646
1932	Boston	NL	40	149	571	67	134	20	4	0	37	46	4	.235	.284	.295	.579
1933	Boston	NL	41	143	478	46	104	15	4	0	38	36	2	.218	.266	.274	.539
1935	Boston	NL	43	23	67	3	10	2	0	0	5	3	0	.149	.179	.186	.365
Totals				2670	10078	1255	2605	380	177	28	884	839	291	.258	.340	.318	.658

 

 

RABBIT MARANVILLE
Career Won-Lost Contribution:	317-317			Offensive Winning Percentage:	.399
Offensive W-L Contribution:	177-267			Defensive Winning Percentage:	.737
Defensive W-L Contribution:	140-50			Defensive WPct, Up to Age 30:	.822
Team Success Percentage:	.564			Defensive WPct, After 30:	.642
Estimated Innings at Short:	19259			Best Season:	1914
Career Putouts at Shortstop:	5139				24-13
Expected Putouts at Short:	4837	+302		Career 20-Win Seasons:	7
Career Assists at Shortstop:	7354			.800 Seasons (as a regular):	0
Expected Assists at Short:	7194	+160		.700 Seasons (as a regular):	0
Career Double Plays a/Short:	1183			.600 Seasons (as a regular):	6
Expected DPs at Shortstop:	1171	+12		.500 Seasons (as a regular):	11
Career Errors at Shortstop:	631			Sub-.500 seasons a/regular:	6
Expected Errors at Short:	782	+151		Career Value:	317

 

            Rabbit Maranville was a .500 player in a very long career, and no doubt some people will disapprove of the fact that a .500 player is in the Hall of Fame.   That’s a little unfair to Maranville.   Maranville’s won-lost contributions, beginning in 1914, are 24-13, 20-14, 24-16, 22-13, 1-1, 19-11, 15-15, 24-14, 22-19, 19-17, 20-19.    He hung around for eleven years after that as a sub-.500 who was still over replacement level, and this very long, very slow decline phase pulled his career winning percentage down to .500.   But the fact that he aged very slowly does not mean that he was not a good player for ten years.

 

 

 

EVERETT SCOTT
Yr	City	Lg	Age	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1914	Boston	AL	21	144	539	66	129	15	6	2	37	32	9	.239	.301	.286	.586
1915	Boston	AL	22	100	359	25	72	11	0	0	28	17	4	.201	.231	.237	.468
1916	Boston	AL	23	123	366	37	85	19	2	0	27	23	8	.232	.295	.283	.578
1917	Boston	AL	24	157	528	40	127	24	7	0	50	20	12	.241	.313	.268	.581
1918	Boston	AL	25	126	443	40	98	11	5	0	43	12	11	.221	.269	.242	.510
1919	Boston	AL	26	138	507	41	141	19	0	0	38	19	8	.278	.316	.306	.621
1920	Boston	AL	27	154	569	41	153	21	12	4	61	21	4	.269	.369	.300	.669
1921	Boston	AL	28	154	576	65	151	21	9	1	60	27	5	.262	.335	.295	.630
1922	New York	AL	29	154	557	64	150	23	5	3	45	23	2	.269	.345	.304	.649
1923	New York	AL	30	152	533	48	131	16	4	6	60	13	1	.246	.325	.266	.591
1924	New York	AL	31	153	548	56	137	12	6	4	64	21	3	.250	.316	.278	.593
1925	New York	AL	32	22	60	3	13	0	0	0	4	2	0	.217	.217	.242	.459
1925	Washington	AL	32	33	103	10	28	6	1	0	18	4	1	.272	.350	.299	.649
1926	Chicago	AL	33	40	143	15	36	10	1	0	13	9	1	.252	.336	.296	.632
1926	Cincinnati	NL	33	4	6	1	4	0	0	0	1	0	0	.667	.667	.667	1.33
Totals				1654	5837	552	1455	208	58	20	549	243	69	.249	.315	.281	.596

 

 

EVERETT SCOTT
Career Won-Lost Contribution:	176-204			Offensive Winning Percentage:	.277
Offensive W-L Contribution:	74-194			Defensive Winning Percentage:	.907
Defensive W-L Contribution:	102-10			Defensive WPct, Up to Age 30:	.926
Team Success Percentage:	.643			Defensive WPct, After 30:	.842
Estimated Innings at Short:	14244			Best Season:	1921
Career Putouts at Shortstop:	3351				19-17
Expected Putouts at Short:	3316	+35		Career 20-Win Seasons:	0
Career Assists at Shortstop:	5053			.800 Seasons (as a regular):	0
Expected Assists at Short:	4847	+206		.700 Seasons (as a regular):	0
Career Double Plays a/Short:	710			.600 Seasons (as a regular):	0
Expected DPs at Shortstop:	675	+35		.500 Seasons (as a regular):	5
Career Errors at Shortstop:	306			Sub-.500 seasons a/regular:	6
Expected Errors at Short:	401	+95		Career Value:	162

 

            There are a bunch of these players who played for Washington in the early 1920s.   McBride was there until 1920, Donie Bush (as an old player) in ’22-’23, Roger Peckinpaugh from 1922 to 1926, Everett Scott as an old player in 1925.

            Roger Peckinpaugh was the Yankee shortstop from 1913 to 1921.   On December 20, 1921, the Yankees traded a package of players including Peckinpaugh and a pile of money to the Red Sox for a package of players including Everett.    The Yankees were a dominant organization at that time, the Red Sox a submissive organization that was selling off talent.  The inclusion of Peckinpaugh and Scott in this trade almost certainly indicates that the Yankees preferred Scott to Peckinpaugh. 

            This—like the MVP votes for Dooley, McBride and others—indicates that contemporary observers placed a higher value on defense than I do.   My analytical method here places a higher value on the defense of these guys (Dooley, McBride, Scott and others). . . my analytical method credits these players with .850+ winning percentages as defensive players, and thus places a very high value on their defense.    But—while agreeing that Scott was a great defensive player—I still see Peckinpaugh as a better all-around player than Scott, and thus see the Yankees as weakening their team by trading Peckinpaugh for Scott.

            We need to respect the opinions of on-scene observers.   We cannot and should not assume that we’re right and they are wrong.   We need to respect the fact that they saw these men play with their own eyes, and we didn’t.

            That’s not to say that they are right and we’re wrong; I still believe in my method.   But. . .it is possible that the value of their defense was even greater than the value I have placed on it.

 

 

 

DAVE BANCROFT
Yr	City	Lg	Age	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1915	Philadelphia	NL	24	153	563	85	143	18	2	7	30	77	15	.254	.330	.346	.676
1916	Philadelphia	NL	25	142	477	53	101	10	0	3	33	74	15	.212	.252	.323	.574
1917	Philadelphia	NL	26	127	478	56	116	22	5	4	43	44	14	.243	.335	.307	.641
1918	Philadelphia	NL	27	125	499	69	132	19	4	0	26	54	11	.265	.319	.338	.656
1919	Philadelphia	NL	28	92	335	45	91	13	7	0	25	31	8	.272	.352	.333	.686
1920	Philadelphia	NL	29	42	171	23	51	7	2	0	5	9	1	.298	.363	.337	.700
1920	New York	NL	29	108	442	79	132	29	7	0	31	33	7	.299	.396	.349	.745
1921	New York	NL	30	153	606	121	193	26	15	6	67	66	17	.318	.441	.389	.830
1922	New York	NL	31	156	651	117	209	41	5	4	60	79	16	.321	.418	.397	.815
1923	New York	NL	32	107	444	80	135	33	3	1	31	62	8	.304	.399	.391	.789
1924	Boston	NL	33	79	319	49	89	11	1	2	21	37	4	.279	.339	.356	.694
1925	Boston	NL	34	128	479	75	153	29	8	2	49	64	7	.319	.426	.400	.826
1926	Boston	NL	35	127	453	70	141	18	6	1	44	64	3	.311	.384	.399	.783
1927	Boston	NL	36	111	375	44	91	13	4	1	31	43	5	.243	.307	.322	.629
1928	Brooklyn	NL	37	149	515	47	127	19	5	0	51	59	7	.247	.303	.326	.629
1929	Brooklyn	NL	38	104	358	35	99	11	3	1	44	29	7	.277	.332	.331	.663
1930	New York	NL	39	10	17	0	1	1	0	0	0	2	0	.059	.118	.158	.276
Totals				1913	7182	1048	2004	320	77	32	591	827	145	.279	.358	.355	.714

 

 

DAVE BANCROFT
Career Won-Lost Contribution:	262-183			Offensive Winning Percentage:	.518
Offensive W-L Contribution:	161-150			Defensive Winning Percentage:	.753
Defensive W-L Contribution:	101-33			Defensive WPct, Up to Age 30:	.854
Team Success Percentage:	.587			Defensive WPct, After 30:	.645
Estimated Innings at Short:	16647			Best Season:	1921
Career Putouts at Shortstop:	4623				27-9
Expected Putouts at Short:	4166	+457		Career 20-Win Seasons:	6
Career Assists at Shortstop:	6561			.800 Seasons (as a regular):	0
Expected Assists at Short:	6365	+196		.700 Seasons (as a regular):	3
Career Double Plays a/Short:	1017			.600 Seasons (as a regular):	7
Expected DPs at Shortstop:	1048	-31		.500 Seasons (as a regular):	10
Career Errors at Shortstop:	660			Sub-.500 seasons a/regular:	4
Expected Errors at Short:	696	+36		Career Value:	301

 

            There are two players in this group of eight who are in the Hall of Fame, Maranville and Bancroft.   There are also two players in this group of eight who have "Career Value" over 300—Maranville and Bancroft.    To this point in our study, every player who has had a career value higher than 300 has been selected to the Hall of Fame (Ward, Wallace, Tinker, Maranville and Bancroft), while no one with a career value of 300 or lower has been selected.  Herman Long was at 300 even.

            I’d better put a little more time into explaining the data.   I’ll move on to "Assists", because I think how we figure "Expected Assists" is a little easier to explain than putouts or double plays.

            The first thing you need to understand is that "assists" by a team represent ground balls.   A team has a few assists that do not result from ground balls, and they have a few ground ball outs that do not cause an assist, but only a relative handful of each.   Basically, ground ball outs are assists, and assists are ground ball outs.    Thus, we don’t need to adjust separately for the ground ball tendencies of the pitching staff because, in starting with the team’s assist total, we are, for all practical purposes, starting with their ground balls.  

            We then ask two questions:

            1)  How many assists did this team have, and

            2)  What percentage of the league’s assists were recorded by shortstops?

            National League, 1917.   There were 16,598 assists in the league, of which 4,155 were by shortstops.   Basically, one in four.   25%.    The percentage is higher now; now it’s more like 29%. 

            Dave Bancroft was the Philadelphia shortstop.  The Phillies had 2,106 assists as a team.   If 25% of those had been recorded by the Philadelphia shortstops, that would have been 527 assists.      We thus have an expectation that Philadelphia shortstops might have recorded about 527 assists on the season—the team assists total, times the percentage of assists that the shortstop normally records.

            But we have to adjust that for the left handed/right-handed mix of the pitching staff.   More left-handed pitching leads to more right-handed hitters batting for the other team, which leads to more ground balls to the third base side.   The team defensive statistics reflect the left/right balance of the pitching staff.   Teams with more left-handed pitching DO have more assists by third basemen and shortstops than teams with less left-handed pitching; not just in theory, they actually do.

             For third basemen, this is a pretty significant adjustment.   For shortstops, it is not nearly as significant, but here’s what we do.

            1)   We figure what percentage of the balls put in play against the team were put in play against left-handed pitchers,

            2)   We figure the same for the league,

            3)   We figure whether the team is over or under the league norm, and by how much.  

            The Phillies had 1,156 Balls in Play against left-handed pitchers, which is 257 fewer than the expectation based on the league norms.    There’s an increase of one assist by the shortstop for each 120 balls in play against left-handed pitchers, so that decreases the expected assists by Philly shortstops from 527 to 525.

            OK, that was kind of a waste of time; so maybe it wasn’t the best example.   It’s not always a waste of time.   In 1918, for example, the Phillies had no balls in play against left-handed pitchers; none.

            We take the 525 expected assists for Philadelphia shortstops in 1917, and we scale it back for Bancroft’s innings.   Bancroft played an estimated 1,066 innings at shortstop for the Phillies in 1917, out of a team total of 1,389 innings.    If we expect a team to have 525 shortstop assists in 1,389 innings, then we expect them to have 403 assists in 1,066 innings.

            Bancroft in 1917 actually had 439 assists at shortstop—36 more than expected.   So, for the year, he’s +36 assists. . ..439 actual vs. 403 expected.    Which was a fairly typical year for him at that time; he was +47 in 1916, +20 in 1918.  

            Small note.   For data prior to 1914, we don’t adjust the expected assists for the left/right mix of the pitching staff because, prior to 1914, teams did not platoon to any significant extent.  There was isolated platooning before 1914, here and there, and there were a handful of switch hitters in the game before 1914, but basically, platooning starts in 1914—therefore, we have to adjust the data beginning in 1914.  

 

 

 

DICK BARTELL
Yr	City	Lg	Age	G	AB	R	H	2B	3B	HR	BI	BB	SB	AVG	SLG	OBA	OPS
1927	Pittsburgh	NL	19	1	2	0	0	0	0	0	0	2	0	.000	.000	.500	.500
1928	Pittsburgh	NL	20	72	233	27	71	8	4	1	36	21	4	.305	.386	.377	.763
1929	Pittsburgh	NL	21	143	610	101	184	40	13	2	57	40	11	.302	.420	.347	.766
1930	Pittsburgh	NL	22	129	475	69	152	32	13	4	75	39	8	.320	.467	.378	.845
1931	Philadelphia	NL	23	135	554	88	160	43	7	0	34	27	6	.289	.392	.325	.717
1932	Philadelphia	NL	24	154	614	118	189	48	7	1	53	64	8	.308	.414	.379	.792
1933	Philadelphia	NL	25	152	587	78	159	25	5	1	37	56	6	.271	.336	.340	.675
1934	Philadelphia	NL	26	146	604	102	187	30	4	0	37	64	13	.310	.373	.384	.757
1935	New York	NL	27	137	539	60	141	28	4	14	53	37	5	.262	.406	.316	.722
1936	New York	NL	28	145	510	71	152	31	3	8	42	40	6	.298	.418	.355	.773
1937	New York	NL	29	128	516	91	158	38	2	14	62	40	5	.306	.469	.367	.836
1938	New York	NL	30	127	481	67	126	26	1	9	49	55	4	.262	.376	.347	.724
1939	Chicago	NL	31	105	336	37	80	24	2	3	34	42	6	.238	.348	.335	.683
1940	Detroit	AL	32	139	528	76	123	24	3	7	53	76	12	.233	.330	.335	.665
1941	Detroit	AL	33	5	12	0	2	1	0	0	1	2	0	.167	.250	.333	.583
1941	New York	NL	33	104	373	44	113	20	0	5	35	52	6	.303	.397	.394	.791
1942	New York	NL	34	90	316	53	77	10	3	5	24	44	4	.244	.342	.351	.692
1943	New York	NL	35	99	337	48	91	14	0	5	28	47	5	.270	.356	.371	.727
1946	New York	NL	38	5	2	0	0	0	0	0	0	0	0	.000	.000	.000	.000
Totals				2016	7629	1130	2165	442	71	79	710	748	109	.284	.391	.355	.747

 

 

DICK BARTELL
Career Won-Lost Contribution:	251-203			Offensive Winning Percentage:	.482
Offensive W-L Contribution:	159-170			Defensive Winning Percentage:	.739
Defensive W-L Contribution:	92-33			Defensive WPct, Up to Age 30:	.770
Team Success Percentage:	.582			Defensive WPct, After 30:	.644
Estimated Innings at Short:	14762			Best Season:	1937
Career Putouts at Shortstop:	3872				24-6
Expected Putouts at Short:	3475	+397		Career 20-Win Seasons:	4
Career Assists at Shortstop:	5590			.800 Seasons (as a regular):	0
Expected Assists at Short:	5495	+95		.700 Seasons (as a regular):	2
Career Double Plays a/Short:	1072			.600 Seasons (as a regular):	4
Expected DPs at Shortstop:	1132	-60		.500 Seasons (as a regular):	10
Career Errors at Shortstop:	471			Sub-.500 seasons a/regular:	5
Expected Errors at Short:	504	+33		Career Value:	275

 

            There are profound and obvious similarities between Dick Bartell and Dave Bancroft.   They were both National League shortstops, overlapping eras, similar hitters.   Both of them played short for Philly for several years, got traded or sold to the Giants, had four really good years with the Giants—winning three pennants each--bounced out of New York, came back to the Giants to finish their careers.

            In one of my books, one of the Historical Abstracts, I cited a string of these similarities between Bartell and Bancroft, and wrapped it up by saying that if Bancroft belonged in the Hall of Fame, Bartell belonged, and if Bancroft did, Bartell didn’t.   One thing that I did not know when I wrote that was how similar their defensive play was, as well; they both have huge positive numbers in putouts, smaller positives in assists and errors, and they’re both negative in double plays. . .a pattern that suggests that they had strong arms.   If a shortstop has a strong arm he backs up and tends to play further from second, which enables him to chase down more balls in shallow left, and still make the throw to first when he has to.   A shortstop with a weaker arm tends to creep in and cheat toward second, resulting in more assists but fewer putouts.  Bancroft and Bartell are defensively similar as well.

            But having said that, I now realize now that this was not the best logic.   What I was saying was that because Bancroft and Bartell have a long string of the same characteristics, that makes them equally qualified for the Hall of Fame.   It doesn’t.  That they have many of the same characteristics makes them similar; it doesn’t make them equal.   Bartell has a won-lost contribution of 251-203; Bancroft, of 262-183.   Those are similar records and it’s a small difference, but you have to draw a line somewhere.   It is not unreasonable to draw that line between Bartell and Bancroft, even though they are unusually similar.

            Which is the same argument as with Omar and Ozzie; yes, they are similar on many different levels.   But the fact that they are similar does not make them equal, and it may be reasonable to draw a line between them.