The Systems vs. the actions of the Committee

This is an effort by which, over time, I wish to determine if there are certain systems which can be useful for predicting at-large inclusion in the NCAA tournament or tournament seed. This is not to say that the systems at the bottom of the list are inferior. They may have been written for a purpose that is very different from this. I also would not put to much stock in just one year. I hope to look at this trend over time. I would also like to see if the systems with certain intentions do better at predicting inclusion or seed.

The systems are sorted by a crude average of the results. It is pretty clear that some systems do better than others from year to year but no system is the dominant indicator of anything. I think over the last three years DWHoops has done the best overall job of predicting the bracket.

2003 Results

Name	Best Rank Left Out)	Worst Rank Let In	Number of the 34 At-Large Correct(i)	Total Rank of the At-Large teams excluding teams with automatic bids	Total Seed Error	Teams not within 1 seed	Teams not within 2 seeds
Bobcat	39	47	31	607	55	10	3
Wolfe	42	47	32	601	63	13	6
Comparison	40	49	33	602	63	12	4
Rothman	39	47	32	605	61	15	1
DWHoops	41	47	32	604	63	14	5
The Sports Report	40	47	32	603	65	12	7
Greenfield	42	51	31	606	61	12	6
Wobus	43	49	32	600	67	14	8
Sauceda	38	47	32	604	67	14	5
Hermrats	39	54	32	603	71	14	8
Bihl	46	52	32	601	79	18	9
RPI	40	53	32	606	71	18	7
Sagarin	31	50	32	612	67	14	5
ARGH	32	51	33	609	69	16	6
Massey	42	54	31	614	71	15	7
Colley	41	56	33	606	73	18	11
Sagarin-ELO	42	51	32	608	77	18	10
Dolphin	41	54	30	615	75	17	7
Whitlock	44	63	32	607	75	20	10
Gindin	45	60	33	605	87	21	9
Brzezinski	42	63	32	622	79	21	6
Pomeroy	34	57	30	630	75	17	6
Markov	26	57	31	629	87	23	9
Entropy	31	63	31	626	89	26	9
Eck	34	69	32	622	117	28	16
Moore	28	74	30	686	91	27	12
Holland	40	77	30	648	151	36	24
White	27	81	28	685	119	31	19
Elrod	17	92	29	706	105	31	12
Dunkel	24	82	28	719	141	32	23
Strength	34	116	25	815	159	39	27

2002 Results

Name	Best Rank Left Out	Worst Rank Let In	Number of the 34 At-Large Correct	Total Rank of the At-Large teams excluding teams with automatic bids	Total Seed Error	Teams not within 1 seed	Teams not within 2 seeds
Wobus	42	53	32	606	69	17	9
DWHoops	42	57	32	614	63	16	8
Self	40	54	30	626	63	16	5
Bihl	39	55	31	613	77	16	11
Greenfield	40	57	31	625	69	18	8
Comparison	39	58	30	637	63	14	5
Bobcat	40	59	30	629	69	17	8
Massey	38	58	31	617	85	22	10
RPI	43	63	31	626	79	21	10
FACT	36	62	29	655	61	14	7
Whitlock	38	65	30	637	79	19	10
Sagarin	38	64	29	653	67	17	9
Sauceda	37	69	29	658	65	14	10
Brzezinski	33	67	28	680	63	15	5
Eck	44	76	31	622	107	25	17
ARGH	28	70	30	676	85	22	11
Pomeroy	34	75	29	675	91	24	12
Holland	30	71	30	673	121	31	17
Freeze	25	74	25	742	93	22	14
Entropy	19	77	28	745	95	26	13
Markov	17	80	28	749	97	26	13
Moore	21	70	24	780	97	26	15
Dunkel	15	100	26	839	123	27	18

2001 Results

Name	Best Rank Left Out	Worst Rank Let In	Number of the 34 At-Large Correct	Total Rank of the At-Large teams excluding teams with automatic bids	Total Seed Error	Teams not within 1 seed	Teams not within 2 seeds
Sorenson	40	46	33	599	55	12	5
Greenfield	34	44	33	602	59	13	4
DWHoops	29	49	33	610	51	9	4
Wobus	33	47	32	607	59	12	5
Comparison	37	49	32	606	63	17	6
Sauceda	37	50	32	606	69	16	9
Bobcat	34	52	32	609	65	15	8
Sagarin	36	49	32	606	71	17	9
Bihl	34	51	32	609	71	16	6
RPI	40	49	32	605	77	19	12
McComas	41	54	31	610	77	17	11
Massey	43	54	32	606	89	26	12
Whitlock	34	59	31	628	69	15	10
ARGH	32	58	31	619	79	19	12
Packard	32	56	30	630	73	18	10
Pomeroy	34	58	31	634	85	23	9
Moore	32	60	30	645	81	19	12
SW!-TECH	35	61	30	644	85	24	9
E-Ratings	33	66	30	655	99	26	12
Flyman	26	71	30	666	135	31	23
Holland	35	88	28	717	167	36	27
Dunkel	23	116	26	876	141	38	22

Bihl Rankings Home