NCAA pointspreads and winners

I made use of the paper several times, for example discussing the odds of Ohio State getting to the BCS title game in 2007. Right after the loss to Illinois, the probability of Ohio State getting to the BCS title game was around 1%. So many teams above Ohio State lost in the subsequent weeks, that by the time I wrote that article the probability was over 50%. The probability of finishing #1 BCS (instead of just "top two") was only 2%, but that happened as well. It was a crazy season.

I recognized at the time that I was really abusing the numbers in Stern's paper. His numbers came from NFL pointspread data. NFL teams are subject to more scrutiny, and I believe that they perform on a more consistent level than college teams. I suspected that the probability of a 7-point NFL favorite losing (30%) might be different from the probability of a 7-point NCAA favorite losing.

Over the last month or two, with the generous help of Phil Steele, I've been assembling a database of all pointspreads and outcomes for BCS teams from 1999 through 2010 inclusive. The result is 9,626 games entered. (Note that each BCS-vs-BCS game is entered twice: For example, "Kansas -2 at Kansas State, result 24-10" would also show up in the data as "Kanas State +2 hosting Kansas, result 10-24." A game like Akron at Ohio State would only be listed once.) I divided the data into buckets by pointspread and computed the percentage of time that the favorite won in each bucket. For a 7-point favorite, that exact bucket yielded 69% wins by the favorite. Also, the best-fit curve seems likely to yield a similar value, so the NCAA numbers seem to be similar to the NFL ones.

The result, accumulatd by a simple perl script and then charted in Excel, is shown below. The Y-axis is the percent of teams which were favored by the exact amount that won their game; the X-axis is the size of the pointspread. There's one pointspread bucket per half-point (teams with a "0" line, teams favored by 0.5 points teams favored by 1 point, teams favored by 1.5 points, etc.). The chart shows three plotted data sets against those axes: the full data set (red), home favorites only (green) and road favorites only (purple).

Note that some of the bumps in the higher pointspreads are exaggerated due to the small number of samples that go into those data points. For 1-point favorites through 7-point favorites, every bucket has a minimum of 200 games. But consider the purple data point around -30 which falls on the 60% line. For that bucket -- "away games where the visiting team is a 30.5-point favorite" -- there are only three data points (two BCS-vs-BCS games each counted twice, and one other). Rutgers in 1999 beat Syracuse despite being 30.5-point underdogs, which (as a double-counted game, yielding only 3/5 wins by the favorite) by itself accounts for that deviation.

That's also true for the overly-high purple data point on the extreme left of the graph: there was only one 0.5-point road favorite in my data set. That team happened to win the game, to yield 100% winners for that bucket. With college overtime, it's not possible for a game to end in a tie score. As a result spreads of +0.5, 0, and -0.5 are all effectively the same.

The raw data that goes into the chart, is:

Spread	Favorite Wins	Games	Win Pct	Home Fav wins	Home Games	Home Win %	Away Fav wins	Away Games	Away Win %
0	25	46	54.35	8	17	47.06	11	18	61.11
0.5	2	4	50.00	0	2	0.00	2	2	100.00
1	125	243	51.44	52	99	52.53	48	99	48.48
1.5	95	211	45.02	44	89	49.44	42	100	42.00
2	132	206	64.08	51	93	54.84	64	83	77.11
2.5	179	396	45.20	85	192	44.27	75	160	46.88
3	328	587	55.88	143	263	54.37	154	264	58.33
3.5	293	483	60.66	124	208	59.62	129	208	62.02
4	156	241	64.73	76	121	62.81	58	88	65.91
4.5	145	241	60.17	78	125	62.40	51	88	57.95
5	115	173	66.47	61	95	64.21	39	60	65.00
5.5	124	206	60.19	71	104	68.27	41	82	50.00
6	178	276	64.49	93	142	65.49	70	106	66.04
6.5	266	372	71.51	158	215	73.49	87	118	73.73
7	299	433	69.05	151	211	71.56	128	182	70.33
7.5	243	328	74.09	143	183	78.14	81	110	73.64
8	126	186	67.74	67	104	64.42	41	61	67.21
8.5	129	179	72.07	75	110	68.18	42	57	73.68
9	104	142	73.24	58	75	77.33	30	46	65.22
9.5	115	162	70.99	73	96	76.04	36	57	63.16
10	190	248	76.61	105	141	74.47	67	83	80.72
10.5	116	155	74.84	69	94	73.40	36	48	75.00
11	102	141	72.34	64	92	69.57	36	43	83.72
11.5	98	124	79.03	52	71	73.24	43	48	89.58
12	105	124	84.68	61	72	84.72	34	40	85.00
12.5	105	123	85.37	78	93	83.87	24	26	92.31
13	111	147	75.51	68	92	73.91	38	49	77.55
13.5	206	253	81.42	143	174	82.18	60	76	78.95
14	182	220	82.73	121	146	82.88	47	56	83.93
14.5	126	148	85.14	83	95	87.37	35	43	81.40
15	95	106	89.62	67	74	90.54	25	29	86.21
15.5	82	96	85.42	56	65	86.15	26	29	89.66
16	77	97	79.38	49	60	81.67	28	37	75.68
16.5	109	129	84.50	71	88	80.68	35	38	92.11
17	118	130	90.77	86	94	91.49	30	34	88.24
17.5	111	125	88.80	80	90	88.89	29	33	87.88
18	72	77	93.51	52	57	91.23	20	20	100.00
18.5	71	76	93.42	51	56	91.07	18	18	100.00
19	72	81	88.89	45	52	86.54	25	27	92.59
19.5	58	69	84.06	46	51	90.20	12	18	66.67
20	100	106	94.34	72	78	92.31	27	27	100.00
20.5	74	80	92.50	52	56	92.86	19	21	90.48
21	139	141	98.58	101	103	98.06	36	36	100.00
21.5	74	78	94.87	56	60	93.33	18	18	100.00
22	71	78	91.03	49	54	90.74	18	20	90.00
22.5	44	47	93.62	31	33	93.94	13	13	100.00
23	61	65	93.85	39	39	100.00	22	26	84.62
23.5	61	64	95.31	49	50	98.00	12	14	85.71
24	69	73	94.52	53	57	92.98	16	16	100.00
24.5	72	78	92.31	55	58	94.83	14	17	82.35
25	45	50	90.00	30	31	96.77	15	19	78.95
25.5	38	39	97.44	35	36	97.22	2	2	100.00
26	36	38	94.74	29	31	93.55	6	6	100.00
26.5	40	40	100.00	31	31	100.00	7	7	100.00
27	65	70	92.86	51	54	94.44	14	16	87.50
27.5	49	49	100.00	40	40	100.00	6	6	100.00
28	62	62	100.00	46	46	100.00	11	11	100.00
28.5	31	32	96.88	28	29	96.55	3	3	100.00
29	34	37	91.89	28	31	90.32	4	4	100.00
29.5	34	35	97.14	30	31	96.77	3	3	100.00
30	51	51	100.00	34	34	100.00	17	17	100.00
30.5	33	35	94.29	28	28	100.00	3	5	60.00
31	30	31	96.77	22	23	95.65	7	7	100.00
31.5	22	22	100.00	20	20	100.00	2	2	100.00
32	34	34	100.00	28	28	100.00	6	6	100.00
32.5	18	18	100.00	16	16	100.00	2	2	100.00
33	33	33	100.00	26	26	100.00	7	7	100.00
33.5	17	17	100.00	17	17	100.00	-	-	-
34	25	25	100.00	24	24	100.00	1	1	100.00
34.5	21	21	100.00	21	21	100.00	-	-	-
35	31	32	96.88	31	32	96.88	-	-	-
35.5	25	25	100.00	23	23	100.00	2	2	100.00
36	20	20	100.00	19	19	100.00	1	1	100.00
36.5	15	15	100.00	13	13	100.00	2	2	100.00
37	13	15	86.67	10	12	83.33	3	3	100.00
37.5	11	11	100.00	8	8	100.00	3	3	100.00
38	17	17	100.00	15	15	100.00	2	2	100.00
38.5	4	4	100.00	4	4	100.00	-	-	-
39	10	10	100.00	6	6	100.00	4	4	100.00
39.5	7	7	100.00	7	7	100.00	-	-	-
40	8	8	100.00	6	6	100.00	2	2	100.00
40.5	12	12	100.00	11	11	100.00	1	1	100.00
41	21	23	91.30	18	20	90.00	3	3	100.00
41.5	9	9	100.00	8	8	100.00	1	1	100.00
42	12	12	100.00	12	12	100.00	-	-	-
42.5	6	6	100.00	4	4	100.00	2	2	100.00
43	4	4	100.00	4	4	100.00	-	-	-
43.5	6	6	100.00	6	6	100.00	-	-	-
44	6	6	100.00	6	6	100.00	-	-	-
44.5	5	5	100.00	4	4	100.00	-	-	-
45	10	10	100.00	10	10	100.00	-	-	-
45.5	4	4	100.00	4	4	100.00	-	-	-
46	3	3	100.00	3	3	100.00	-	-	-
46.5	3	3	100.00	3	3	100.00	-	-	-
47	4	4	100.00	4	4	100.00	-	-	-
47.5	1	1	100.00	1	1	100.00	-	-	-
48	4	4	100.00	2	2	100.00	2	2	100.00
48.5	1	1	100.00	1	1	100.00	-	-	-
49	4	4	100.00	4	4	100.00	-	-	-
49.5	2	2	100.00	2	2	100.00	-	-	-
50	2	2	100.00	2	2	100.00	-	-	-
51	2	2	100.00	2	2	100.00	-	-	-
51.5	1	1	100.00	1	1	100.00	-	-	-
52	1	1	100.00	1	1	100.00	-	-	-
53	2	2	100.00	2	2	100.00	-	-	-
56	1	1	100.00	1	1	100.00	-	-	-
59.5	1	1	100.00	1	1	100.00	-	-	-

For those interested in the raw data (all the individual game results), I hope to make it available soon. For now, you can click here to download an Excel spreadsheet containing the chart above and the collated data that goes into it. Also, I will work on further analysis on the data, coming up with a best-fit curve and a conversion chart for pointspread to win-percentage for that.