Looking at League Quality

[cubbieinexile]

I wanted to see if the league quality has improved and if so could it be found using the statistics. My parameters were as follows:

American League only

Players must have at least 300 at bats in a season

 

I then looked at the batting average of the individuals as well as tallied the batting average of all the selected players combined. I then found the standard deviation for each season. This is what I found:

 

Year	STD	AL-AVG	Rank-STD	Rank-AVg	Rank-Dis	Distance
1901	0.035	0.291	19	22	60	-3
1902	0.037	0.292	11	21	70	-10
1903	0.036	0.273	18	58	93	-40
1904	0.034	0.258	25	97	100	-72
1905	0.030	0.254	47	101	96	-54
1906	0.037	0.265	15	90	101	-75
1907	0.029	0.263	62	94	89	-32
1908	0.029	0.252	63	103	93	-40
1909	0.037	0.263	13	93	103	-80
1910	0.037	0.266	10	85	101	-75
1911	0.044	0.296	1	18	76	-17
1912	0.043	0.285	2	29	85	-27
1913	0.039	0.279	6	39	90	-33
1914	0.035	0.266	20	82	98	-62
1915	0.035	0.266	21	80	97	-59
1916	0.036	0.266	17	87	99	-70
1917	0.039	0.264	3	92	104	-89
1918	0.037	0.273	14	56	95	-42
1919	0.034	0.285	23	27	64	-4
1920	0.039	0.302	4	7	60	-3
1921	0.032	0.307	35	3	15	32
1922	0.038	0.300	8	10	59	-2
1923	0.038	0.299	7	11	64	-4
1924	0.032	0.306	34	4	18	30
1925	0.039	0.308	5	2	52	3
1926	0.036	0.297	16	14	55	2
1927	0.037	0.303	12	6	48	6
1928	0.034	0.297	22	15	46	7
1929	0.034	0.302	24	8	34	16
1930	0.037	0.309	9	1	44	8
1931	0.033	0.298	31	13	32	18
1932	0.031	0.293	44	20	25	24
1933	0.030	0.288	46	23	28	23
1934	0.030	0.300	48	9	9	39
1935	0.025	0.295	96	19	1	77
1936	0.033	0.306	29	5	25	24
1937	0.033	0.298	27	12	39	15
1938	0.028	0.296	72	16	2	56
1939	0.028	0.296	69	17	4	52
1940	0.033	0.286	28	26	55	2
1941	0.034	0.283	26	32	66	-6
1942	0.028	0.276	71	45	21	26
1943	0.026	0.262	92	95	60	-3
1944	0.027	0.275	80	50	18	30
1945	0.025	0.270	98	73	24	25
1946	0.033	0.277	30	42	72	-12
1947	0.030	0.272	50	65	75	-15
1948	0.032	0.281	36	33	52	3
1949	0.026	0.281	88	34	3	54
1950	0.032	0.286	39	25	40	14
1951	0.026	0.278	84	41	6	43
1952	0.027	0.270	79	68	41	11
1953	0.026	0.277	93	43	5	50
1954	0.031	0.271	40	66	84	-26
1955	0.030	0.275	57	51	48	6
1956	0.032	0.276	37	46	68	-9
1957	0.033	0.272	32	63	86	-31
1958	0.029	0.274	61	53	44	8
1959	0.031	0.269	43	74	86	-31
1960	0.023	0.272	103	61	7	42
1961	0.028	0.274	78	52	21	26
1962	0.026	0.270	91	67	25	24
1963	0.024	0.261	100	96	50	4
1964	0.023	0.264	101	91	42	10
1965	0.025	0.258	95	98	60	-3
1966	0.023	0.255	104	100	50	4
1967	0.028	0.253	66	102	92	-36
1968	0.026	0.249	86	104	78	-18
1969	0.025	0.267	99	77	29	22
1970	0.030	0.265	58	89	86	-31
1971	0.026	0.265	87	88	58	-1
1972	0.026	0.257	90	99	68	-9
1973	0.028	0.266	67	81	74	-14
1974	0.029	0.266	64	84	80	-20
1975	0.029	0.266	60	83	83	-23
1976	0.029	0.266	65	86	82	-21
1977	0.031	0.273	41	54	73	-13
1978	0.023	0.270	102	70	15	32
1979	0.028	0.277	70	44	21	26
1980	0.030	0.279	53	37	34	16
1981	0.028	0.273	75	59	34	16
1982	0.027	0.272	81	60	30	21
1983	0.030	0.273	49	57	67	-8
1984	0.030	0.272	45	64	79	-19
1985	0.027	0.269	83	76	46	7
1986	0.030	0.270	59	69	70	-10
1987	0.031	0.272	42	62	80	-20
1988	0.030	0.270	54	71	76	-17
1989	0.026	0.269	85	75	42	10
1990	0.026	0.267	94	78	34	16
1991	0.032	0.270	38	72	91	-34
1992	0.025	0.267	97	79	32	18
1993	0.030	0.275	52	49	52	3
1994	0.033	0.284	33	31	55	2
1995	0.030	0.280	51	35	34	16
1996	0.030	0.286	56	24	15	32
1997	0.028	0.279	73	38	13	35
1998	0.028	0.278	77	40	12	37
1999	0.028	0.284	68	30	10	38
2000	0.030	0.285	55	28	20	27
2001	0.027	0.276	82	47	13	35
2002	0.028	0.273	76	55	30	21
2003	0.026	0.276	89	48	8	41
2004	0.028	0.280	74	36	10	38

 

First three columns should be self-explanatory. The next three columns are rankings. First is for the STD spread, second if for the league average, and the third is for the difference between the two (more on that below), and the final column is that actual raw difference of the two.

 

Theoritically what you are looking for to determine a high league quality is a league that has a low numerical rank for average (1, 2, 3, . . .)and a high numerical rank for STD spread (101, 102, 103, . . ).

 

To measure that I simply subtract AVG-Rank from STD-Rank and see who has the highest number. The winners?

Distance	Year
77	1935
56	1938
54	1949
52	1939
50	1953
43	1951
42	1960
41	2003
39	1934
38	2004
38	1999
37	1998
35	1997
35	2001
32	1921
32	1996
32	1978

 

According to this the years that one could say had the highest league quality were all right before integration and then when they barely had any integration. The top 6 years in this list one could argue that blacks had no effect or little effect on the league as a whole in those years.

 

The losers?

-32	1907
-33	1913
-34	1991
-36	1967
-40	1903
-40	1908
-42	1918
-54	1905
-59	1915
-62	1914
-70	1916
-72	1904
-75	1906
-75	1910
-80	1909
-89	1917

 

Almost all of them are deadball era years. With only 1967 and 1991 messing up the sweep. So theoritically the deadball era was the era with least overall quality.

 

What I take from this follows:

 

It appears if one wants to assume that a low spread and high average as an indicator of quality that white baseball had probably hit its peak just before WWII. That the infrastucture of baseball for whites was setup well enough that they were employing the best that the white could offer.

Once integration happened in full force league quality went down. Why? Personally I think it is because it turned out that the lower tiered whites were not as good as the upper tiered blacks but they were not pushed out because the blacks were given a limited role. Meaning middle and lowered tiered blacks were still not allowed to play. Until around 1996 were the spread stays consistently small and the Average stays consistently high.

 

Now does that mean I personally believe the AL was in such upheaval and radically redefining itself for over 50 years? No I don't I think the pitchers and DH have some effect on the league average which would definitely effect the rankings. When I look at the NL I will know more.

 

I do think that it does show that early AL was a league with low quality. In otherwords a league with good chunk of great players but also a lot of bad players.

[Guest]

Theoritically what you are looking for to determine a high league quality is a league that has a low numerical rank for average (1, 2, 3, . . .)and a high numerical rank for STD spread (101, 102, 103, . . ).

 

To measure that I simply subtract AVG-Rank from STD-Rank and see who has the highest number. The winners?

Making that assumption (namely, the best leagues are the ones where most players are grouped tightly around th mean) wouldn't it be simpler and more accurate to just express the STD as a percentage of the mean and rank them that way? Here's how the 1901-present NL stacks up using this method on OPS:

 

http://img.photobucket.com/albums/v384/bobskeeper/NL-League1.jpg

 

(Note that using this method the "best" leagues have the lowest percentages.)

[cubbieinexile]

I don't think a simple % of the mean would do it. Because we are not simply looking for the lowest spread but also a high average. Or if we were to include the pitchers (to make sure it isn't simply the best players gravitating to one side or the other) so kind of middle ground where we would feel there is harmony between hitting and pitching.

 

Would we really think a league is of high quality if they have a low overall average but a relatively low STD spread?

I personally wouldn't consider that high quality league, at least in regards to hitters.