Pythagenpat Ratings Through 10 Games

by Chase Stuart on November 23, 2019

in Checkdowns

The Seattle Seahawks have been one of the biggest outliers in recent history when it comes to winning percentage and points differential. Seattle is 8-2, despite scoring 275 points and allowing 254 points. We can calculate a team’s expected winning percentage — it’s Pythagenpat winning percentage — by using a two-step formula.

First, we calculate the total points scored and points allowed in all Seahawks games (529) and divide it by the number of games played (10, leaving 52.9 total points scored per game). We then take that number and raise it to the 0.251 power, and get a result of 2.71. This is the exponent we use when calculating a team’s expected winning percentage using the traditional Pythagorean formula. The advantage here using this two-step process over a static exponent (like 2.37) is that it recognizes that higher-scoring games provide different environments than lower-scoring games. To calculate Seattle’s expected winning percentage, we would use this formula:

(Points Scored ^ Exponent) / [(Points Scored ^ Exponent) + (Points Allowed ^ Exponent)]

In Seattle’s case, that’s:

(275 ^ 2.71) / (275 ^ 2.71 + 254 ^ 2.71)

That gives a result of 0.554, which is Seattle’s expected winning percentage based on its points scored and points allowed. The graph below shows every team this season and their actual winning percentage on the X-Axis, and Pythagenpat winning percentage on the Y-Axis.

Seattle is the biggest outlier here, and by a good margin. The Seahawks have a winning percentage that is 0.246 higher than their Pythagenpat winning percentage; only the Saints (+0.187), Packers (+0.173), and Raiders (+0.169) are at over 10%, and none of those teams are all that close to Seattle. On the other side, the Bucs (-0.116), Broncos (-0.117), Chargers (-0.154), Cowboys (-0.128), and Bengals (-0.166) have underperformed by a large margin.

I performed the same analysis for every team from 1950 to 2018 through 10 games, and included the 2019 teams in red as well.

So how much of an outlier is Seattle? The Seahawks are the 6th-biggest outlier through 10 games since 1950.

Team	Year	PF	PA	Win %	EXP	Pyth Pag	Diff
OAK	1976	216	194	0.900	2.54	0.568	0.332
SDG	1987	195	189	0.800	2.50	0.520	0.280
ATL	2004	208	194	0.800	2.53	0.544	0.256
IND	2006	263	210	0.900	2.63	0.644	0.256
NOR	1973	114	234	0.400	2.44	0.148	0.252
SEA	2019	275	254	0.800	2.71	0.554	0.246
CAR	2003	198	180	0.800	2.49	0.559	0.241
PIT	1962	226	280	0.600	2.68	0.360	0.240
PIT	1989	123	240	0.400	2.46	0.162	0.238
IND	2012	210	260	0.600	2.63	0.363	0.237
CAR	2015	299	191	1.000	2.66	0.767	0.233
JAX	2010	220	270	0.600	2.66	0.367	0.233
NYJ	2013	183	268	0.500	2.60	0.270	0.230
DEN	1992	175	183	0.700	2.45	0.473	0.227
SFO	2006	182	265	0.500	2.60	0.274	0.226
OAK	2016	272	243	0.800	2.69	0.575	0.225
TEN	1999	206	181	0.800	2.50	0.580	0.220
ARI	2014	237	176	0.900	2.54	0.681	0.219
RAM	1969	276	169	1.000	2.59	0.781	0.219
MIN	2000	248	218	0.800	2.62	0.584	0.216
BUF	2007	153	222	0.500	2.48	0.284	0.216
WAS	2001	141	206	0.500	2.44	0.284	0.216
SFO	1990	253	151	1.000	2.53	0.787	0.213
CHI	1962	207	247	0.600	2.61	0.387	0.213
DEN	1979	147	149	0.700	2.34	0.492	0.208
CHI	1991	181	155	0.800	2.42	0.593	0.207
HOU	2016	181	215	0.600	2.52	0.393	0.207
NYG	1965	174	245	0.500	2.55	0.294	0.206
MIN	2005	174	245	0.500	2.55	0.294	0.206
JAX	2009	199	235	0.600	2.58	0.395	0.205
PHI	1960	262	191	0.900	2.60	0.695	0.205
KAN	2011	144	252	0.400	2.52	0.196	0.204
RAI	1982	287	210	0.900	2.67	0.697	0.203
BAL	1978	141	247	0.400	2.50	0.197	0.203
NYT	1961	207	286	0.500	2.66	0.297	0.203
IND	2009	269	157	1.000	2.56	0.799	0.201
HOU	1979	212	212	0.700	2.56	0.500	0.200
DEN	2007	187	258	0.500	2.59	0.303	0.197
OAK	1998	176	175	0.700	2.44	0.503	0.197
STL	1982	151	211	0.500	2.46	0.305	0.195
GNB	2011	355	212	1.000	2.76	0.805	0.195
MIA	2001	214	212	0.700	2.56	0.506	0.194
ATL	2012	270	193	0.900	2.62	0.707	0.193
MIN	1997	238	201	0.800	2.58	0.607	0.193
WAS	1951	153	259	0.400	2.54	0.208	0.192
PHI	1952	193	223	0.600	2.55	0.409	0.191
TAM	2010	209	206	0.700	2.55	0.509	0.191
CLE	1965	263	222	0.800	2.65	0.610	0.190
CIN	1986	237	271	0.600	2.68	0.411	0.189
BAL	1983	183	211	0.600	2.51	0.411	0.189

As always, please leave your thoughts in the comments.

{ 0 comments }