Power Rankings 2007 Week 17

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	NE	10.5759
2	IND	10.3170
3	DAL	10.1534
4	PIT	9.8614
5	TB	9.5910
6	JAX	9.5862
7	GB	9.2658
8	WAS	9.1119
9	MIN	9.1075
10	PHI	9.0675
11	SD	8.7100
12	NYG	8.6673
13	TEN	8.4478
14	SEA	8.3719
15	DEN	8.2001
16	HOU	7.9178
17	CLE	7.7195
18	CIN	7.5846
19	CAR	7.4754
20	ARI	7.3438
21	BUF	7.2895
22	NO	7.2215
23	ATL	7.1775
24	DET	7.1663
25	BAL	6.7068
26	NYJ	6.6356
27	CHI	6.5820
28	STL	6.4483
29	KC	6.2941
30	MIA	6.1462
31	SF	5.6674
32	OAK	5.5888

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	NE	10.5837
2	PIT	10.2658
3	IND	10.1597
4	DAL	10.0700
5	TB	9.8437
6	MIN	9.6704
7	GB	9.5724
8	JAX	9.5190
9	SEA	8.8647
10	SD	8.8134
11	PHI	8.7488
12	WAS	8.7349
13	NYG	8.5790
14	TEN	8.4120
15	DEN	8.2860
16	CLE	8.2273
17	CIN	7.9928
18	ARI	7.8841
19	HOU	7.6742
20	NO	7.3013
21	CAR	7.1811
22	DET	7.0157
23	ATL	6.9466
24	BUF	6.8609
25	NYJ	6.6046
26	BAL	6.5961
27	STL	6.4422
28	KC	6.1708
29	CHI	6.1171
30	MIA	5.7952
31	OAK	5.5368
32	SF	5.5296

Read More......

Power Rankings 2007 Week 16

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	NE	10.9184
2	IND	10.6703
3	DAL	10.4441
4	TB	9.8023
5	JAX	9.6747
6	PIT	9.4925
7	MIN	9.1715
8	PHI	9.1059
9	GB	9.0902
10	SD	8.9651
11	WAS	8.8545
12	SEA	8.4820
13	NYG	8.3811
14	TEN	8.1760
15	DEN	7.9775
16	HOU	7.8321
17	CLE	7.6168
18	CIN	7.4845
19	NO	7.3679
20	DET	7.3406
21	CAR	7.3120
22	BUF	7.2980
23	ARI	7.1704
24	ATL	7.0634
25	BAL	6.9607
26	STL	6.5877
27	CHI	6.5075
28	KC	6.4352
29	NYJ	6.3864
30	MIA	6.2834
31	SF	5.7132
32	OAK	5.4341

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	NE	10.9108
2	IND	10.4595
3	DAL	10.3775
4	TB	9.9997
5	MIN	9.8078
6	PIT	9.7576
7	JAX	9.6016
8	GB	9.3454
9	SEA	8.9187
10	SD	8.8611
11	PHI	8.7189
12	WAS	8.5318
13	NYG	8.4430
14	DEN	8.2738
15	TEN	8.2570
16	CLE	8.0482
17	CIN	7.8102
18	HOU	7.6528
19	ARI	7.6204
20	NO	7.2802
21	DET	7.2723
22	CAR	7.0987
23	BAL	7.0671
24	BUF	6.9034
25	ATL	6.8773
26	STL	6.6190
27	NYJ	6.2978
28	KC	6.2621
29	CHI	5.9982
30	MIA	5.9183
31	SF	5.6119
32	OAK	5.3978

Read More......

Power Rankings 2007 Week 15

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	NE	10.7795
2	DAL	10.5095
3	IND	10.4331
4	MIN	10.1635
5	TB	10.0476
6	PIT	9.8106
7	GB	9.4977
8	JAX	9.0800
9	SEA	8.8732
10	SD	8.7411
11	PHI	8.5580
12	DEN	8.4992
13	WAS	8.3256
14	TEN	8.2315
15	NYG	8.1936
16	CLE	8.1169
17	HOU	7.8518
18	CIN	7.6882
19	ARI	7.6803
20	NO	7.4573
21	DET	7.2631
22	BUF	7.1796
23	BAL	7.1213
24	CAR	7.1122
25	ATL	6.7046
26	STL	6.6001
27	NYJ	6.2123
28	KC	6.0553
29	MIA	6.0417
30	OAK	5.8586
31	CHI	5.7423
32	SF	5.5710

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	NE	10.9732
2	IND	10.6267
3	DAL	10.5591
4	TB	9.9386
5	PIT	9.6408
6	MIN	9.4450
7	JAX	9.4199
8	GB	9.3922
9	PHI	8.9921
10	SD	8.8438
11	WAS	8.5297
12	SEA	8.3678
13	TEN	8.2583
14	NYG	8.1932
15	DEN	8.1181
16	HOU	7.8097
17	CLE	7.7665
18	DET	7.5346
19	BUF	7.4797
20	NO	7.4048
21	CIN	7.3765
22	ARI	7.3366
23	CAR	7.1503
24	BAL	6.9467
25	ATL	6.8989
26	STL	6.4431
27	KC	6.3914
28	NYJ	6.2920
29	CHI	6.2671
30	MIA	6.2115
31	OAK	5.8398
32	SF	5.5523

Read More......

A Power Ranking System that Ranks Detroit 9th (Adjusted Win Probabilities)

The first regular feature I post every week after the games is the expected win probabilities for each game given the box score stats (i.e. in-game only performance of both teams). These probabilities are the result of a logistic regression model trained on the 1996-2006 seasons.

P(team X wins | box score)=99.9% means that team X completely dominated the game statistically. It is easier to dominate the 49ers than the Steelers. Therefore, I tried adjusting the win probabilities for each game based on opponent expected win percentage. The method for opponent adjustments is similar to what I use for adjusting the VOLA stats seen on the site. A concrete example follows the math...



adjXWinProb(team, opponent) = unadjXWinProb(team,opponent) * .5 / (1 - unadjXWinPercentage(opponent))

The win probability can also be interpreted as an expected number of wins (if the matchup were played 1000 times, the expected number of wins would be 1000 * P(win), but it's only played once so 1*P(win)=P(win)).

Example: The Colts had a 77.43% probability of beating the Pats in their meeting this season given the box score. The Patriots have an .8314 expected winning percentage this year. So the Colts' adjusted probability of winning the game is:
.7743 * .5/(1-.8314) = 2.2963 (adjusted expected wins)

Obviously, a 229.63% probability of winning is impossible, but the system gives Indianapolis a lot of credit for performing so well against New England. Similarly, New England is heavily penalized for nearly losing to Baltimore, accruing only .2498 adjusted expected wins. Although one or two games can get weighed very heavily by the adjustment process, the results do not look terribly out of whack.

Below is a table of every team's adjusted expected win total prorated to a 16-game season ranked in order of wins.

Rank	Team	Unadj. Expected Wins	Adj. Expected Wins
1	IND	13.5230	16.0278
2	NE	13.3027	14.7816
3	DAL	12.7584	12.9264
4	PIT	11.1355	11.2092
5	GB	10.9467	10.4008
6	JAX	10.3409	10.1535
7	TB	10.4636	9.8695
8	SEA	11.2993	9.7173
9	DET	8.0443	9.1228
10	PHI	8.4323	8.9552
11	TEN	7.7004	8.6468
12	MIN	8.8850	8.2582
13	KC	6.6498	8.0524
14	WAS	7.3755	7.9169
15	NYG	7.9746	7.5721
16	DEN	8.1019	7.4848
17	SD	8.3092	7.4617
18	CIN	7.7629	7.3479
19	ARI	6.5629	7.2998
20	ATL	6.9780	7.1539
21	CLE	7.6921	7.0279
22	NO	7.3527	6.9686
23	BUF	6.7337	6.5243
24	HOU	7.2044	6.5216
25	BAL	5.9070	6.5084
26	NYJ	5.4645	5.4883
27	MIA	5.1138	5.1293
28	CAR	5.8493	4.9563
29	OAK	5.4123	4.9256
30	STL	5.9473	4.9045
31	CHI	4.3431	4.5478
32	SF	2.4330	2.1388

For most teams, their ranking ends up similar to what's seen in my other power ranking system with a few noticable exceptions.

Kansas City moves up from 27 to 13, thanks largely to its performance against Indy in week 11.
San Diego moves down from 10 to 17 with their most valuable win coming against Denver (1.0128 expected wins). Wins against Baltimore, Kansas City, and Tennessee are devalued by the opponent adjustments.

And then there's Detroit. They gained 2.0234 adjusted expected wins for their performance against Dallas. Strong performances against Tampa Bay, Minnesota, and Denver also help. Of the teams in the NFC wild card chase, Detroit comes out on top because they were the most impressive against tougher opponents.

Read More......

Power Rankings 2007 Week 14

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	NE	10.9917
2	DAL	10.8534
3	IND	10.6660
4	PIT	9.9950
5	TB	9.9396
6	GB	9.6261
7	MIN	9.4555
8	JAX	9.1018
9	PHI	8.8448
10	SD	8.5902
11	SEA	8.5582
12	NYG	8.4449
13	WAS	8.3827
14	DEN	8.3703
15	TEN	8.2756
16	DET	7.8931
17	BUF	7.7278
18	CLE	7.6245
19	CIN	7.5566
20	ARI	7.5225
21	HOU	7.3846
22	NO	7.1999
23	BAL	7.0026
24	CAR	6.8037
25	ATL	6.6809
26	CHI	6.4073
27	KC	6.2831
28	MIA	6.2465
29	NYJ	6.2225
30	STL	6.1820
31	OAK	5.6712
32	SF	5.4952

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	DAL	10.7944
2	NE	10.7207
3	IND	10.4349
4	PIT	10.3109
5	TB	10.0348
6	MIN	9.9974
7	GB	9.6121
8	SEA	9.0437
9	JAX	8.9223
10	DEN	8.7471
11	SD	8.4706
12	PHI	8.4676
13	NYG	8.4518
14	TEN	8.1530
15	WAS	8.1198
16	CLE	7.8871
17	ARI	7.8171
18	CIN	7.6125
19	HOU	7.5984
20	DET	7.5810
21	BUF	7.3879
22	NO	7.2593
23	BAL	7.0096
24	CAR	6.9199
25	ATL	6.7258
26	STL	6.4545
27	NYJ	6.2821
28	KC	6.0006
29	CHI	5.9061
30	MIA	5.9046
31	OAK	5.8811
32	SF	5.4911

Read More......

Power Rankings 2007 Week 13

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	DAL	11.2079
2	NE	10.8186
3	IND	10.3761
4	TB	9.9500
5	PIT	9.8970
6	MIN	9.5840
7	GB	9.4376
8	PHI	8.8882
9	SD	8.6438
10	JAX	8.6307
11	NYG	8.5136
12	SEA	8.4706
13	WAS	8.3979
14	TEN	8.2366
15	DEN	7.7704
16	CLE	7.7464
17	ARI	7.6760
18	CIN	7.4911
19	DET	7.3911
20	HOU	7.3772
21	BUF	7.3031
22	BAL	7.1272
23	CAR	7.0606
24	ATL	7.0049
25	NO	6.9037
26	MIA	6.8243
27	KC	6.6999
28	CHI	6.4554
29	STL	6.4279
30	NYJ	6.2846
31	OAK	5.8642
32	SF	5.5392

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	DAL	11.1339
2	NE	10.8625
3	PIT	10.6398
4	IND	10.1367
5	TB	10.1171
6	MIN	9.8523
7	GB	9.1847
8	SEA	8.9511
9	PHI	8.5295
10	SD	8.5163
11	JAX	8.5104
12	NYG	8.4719
13	TEN	8.2418
14	ARI	8.0126
15	DEN	7.9752
16	WAS	7.9187
17	HOU	7.6785
18	CLE	7.6648
19	BAL	7.4479
20	DET	7.4054
21	CAR	7.3351
22	CIN	7.2654
23	ATL	6.9836
24	NO	6.8303
25	BUF	6.6862
26	KC	6.6477
27	STL	6.6255
28	MIA	6.3597
29	NYJ	6.2952
30	OAK	6.2507
31	CHI	5.9139
32	SF	5.5559

Read More......

Power Rankings 2007 Week 12

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	NE	11.2441
2	DAL	11.1332
3	IND	10.6209
4	PIT	9.9642
5	TB	9.9244
6	GB	9.3945
7	MIN	9.3234
8	PHI	9.0560
9	WAS	8.8586
10	NYG	8.7037
11	SEA	8.4958
12	SD	8.4589
13	JAX	8.3526
14	TEN	8.1060
15	DEN	8.0833
16	CLE	7.7783
17	DET	7.6137
18	ARI	7.6130
19	MIA	7.6014
20	HOU	7.5289
21	CIN	7.2008
22	ATL	7.1665
23	NO	6.9120
24	KC	6.9006
25	CAR	6.8935
26	BUF	6.8154
27	BAL	6.5092
28	CHI	6.4680
29	STL	6.2714
30	NYJ	5.7587
31	SF	5.6797
32	OAK	5.5694

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	NE	11.3002
2	DAL	11.2019
3	PIT	10.7758
4	IND	10.3835
5	TB	9.9057
6	MIN	9.4825
7	GB	9.4620
8	SEA	9.0465
9	PHI	8.5803
10	JAX	8.4789
11	NYG	8.4063
12	WAS	8.2131
13	SD	8.1418
14	DEN	8.1411
15	TEN	8.0375
16	ARI	7.9962
17	DET	7.9144
18	HOU	7.8943
19	CLE	7.6587
20	CIN	7.2868
21	BAL	7.2055
22	NO	7.1142
23	ATL	7.0253
24	KC	6.9362
25	CAR	6.9291
26	MIA	6.8095
27	STL	6.3635
28	BUF	6.2475
29	OAK	5.9599
30	CHI	5.8376
31	SF	5.7099
32	NYJ	5.5542

Read More......

Power Rankings 2007 Week 11

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	NE	11.4795
2	DAL	11.3244
3	IND	10.7248
4	TB	10.2407
5	PIT	9.8901
6	MIN	9.1499
7	GB	9.0889
8	NYG	8.8950
9	PHI	8.8749
10	WAS	8.7871
11	SEA	8.3319
12	TEN	8.3209
13	JAX	8.1216
14	SD	8.1022
15	DEN	7.9506
16	ARI	7.8998
17	DET	7.7828
18	CLE	7.7737
19	MIA	7.5919
20	HOU	7.5603
21	CAR	7.3835
22	BUF	7.1744
23	ATL	7.0890
24	CIN	7.0321
25	KC	6.9455
26	BAL	6.7021
27	CHI	6.5708
28	NO	6.4245
29	STL	6.1375
30	NYJ	5.8191
31	SF	5.5370
32	OAK	5.2937

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	NE	11.6398
2	DAL	11.1056
3	PIT	10.8020
4	IND	10.3668
5	TB	10.1863
6	MIN	9.4168
7	GB	9.1591
8	SEA	8.8782
9	PHI	8.6894
10	NYG	8.6627
11	DET	8.3091
12	TEN	8.2302
13	WAS	8.1641
14	JAX	8.1178
15	HOU	8.0090
16	ARI	7.8899
17	SD	7.8145
18	DEN	7.7899
19	CLE	7.5595
20	BAL	7.4997
21	CAR	7.3289
22	ATL	7.1751
23	CIN	7.0558
24	KC	7.0239
25	MIA	6.8786
26	NO	6.5305
27	BUF	6.4760
28	STL	6.4508
29	CHI	5.9604
30	NYJ	5.6965
31	OAK	5.5697
32	SF	5.5634

Read More......

Power Rankings 2007 Week 10

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

New England is mediocre in run offense and defense according to yards per play. That is the sole reason they are #2 to Dallas. Indy and San Diego stay more or less in place despite sloppy games. Minnesota is still really high because Adrian Peterson has been so ridiculously good and so has their run defense. By yards per play, they are 43% above league average running the ball and 27% above league average stopping the run. They are 16% and 7% below average in pass offense and defense respectively. Philadelphia is at #9 and looks poised to make a comeback in the division, though the ranking seems to be based a lot on the strength of their running game. Similarly, Miami is considered a middle of the pack team with 4.6 yards per carry on offense and an improving but still sucky defense.

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	DAL	11.6155
2	NE	11.4932
3	IND	11.0721
4	PIT	10.4505
5	TB	10.0557
6	MIN	9.2326
7	GB	9.1011
8	NYG	9.0704
9	PHI	8.8011
10	TEN	8.7476
11	SEA	8.4169
12	WAS	8.3712
13	JAX	8.2691
14	SD	7.9571
15	ARI	7.9354
16	CLE	7.7619
17	DEN	7.7385
18	MIA	7.6025
19	DET	7.5167
20	CAR	7.4233
21	ATL	7.3070
22	HOU	7.2890
23	BUF	7.1928
24	CIN	7.0074
25	BAL	6.7075
26	NO	6.6808
27	KC	6.4052
28	CHI	6.3966
29	STL	6.2136
30	SF	5.5396
31	NYJ	5.3895
32	OAK	5.2385

Read More......

Power Rankings 2007 Week 9

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

After making a huge comeback to #6, San Diego drops down to the middle of the pack. And somehow Dallas vaulted over New England and Indianapolis to take the #1 spot. Minnesota and the Giants are surprisingly strong despite spotty QB play.

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	DAL	11.6828
2	NE	11.5958
3	IND	11.3737
4	PIT	10.2007
5	TB	10.0710
6	MIN	9.4761
7	NYG	9.0151
8	TEN	8.9846
9	GB	8.6627
10	WAS	8.5938
11	PHI	8.5899
12	SEA	8.5218
13	JAX	8.2240
14	CLE	7.7882
15	SD	7.7499
16	DET	7.7243
17	MIA	7.7148
18	CAR	7.6460
19	ARI	7.5698
20	DEN	7.3299
21	HOU	7.2592
22	BUF	7.2020
23	ATL	7.1826
24	CIN	7.0208
25	NO	7.0126
26	KC	6.6185
27	BAL	6.6078
28	CHI	6.2554
29	STL	5.9045
30	OAK	5.7940
31	NYJ	5.4532
32	SF	5.1744

Read More......

Power Rankings 2007 Week 8

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

San Diego is making a huge comeback to #6 after being bottom of the barrel to start. The Saints are climbing upwards from dead last towards mediocrity.

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	NE	11.5856
2	IND	11.5305
3	DAL	11.0685
4	TB	10.0434
5	PIT	9.6449
6	SD	8.9480
7	TEN	8.8901
8	NYG	8.8737
9	WAS	8.8413
10	PHI	8.6632
11	SEA	8.6215
12	JAX	8.6040
13	GB	8.4589
14	MIN	8.3500
15	CIN	7.9932
16	DEN	7.8561
17	CLE	7.7744
18	ARI	7.6103
19	MIA	7.5862
20	CAR	7.3970
21	DET	7.3160
22	HOU	7.2904
23	BAL	6.9235
24	KC	6.9199
25	NO	6.8708
26	ATL	6.7046
27	BUF	6.6202
28	CHI	6.1950
29	OAK	6.1287
30	STL	6.0256
31	NYJ	5.4028
32	SF	5.2618

Read More......

Projected Final Standings 2007 Week 8

Using the projected win probabilities from the logistic regression power rankings, I estimated which games each team would win on the rest of their respective schedules. And then I summed the projected wins with the actual wins so far to get their projected win totals. The teams are sorted by division.

Projected Final Standings
Team	Projected Win Total
AFC East
BUF	5
MIA	4
NE	15
NYJ	2
AFC North
BAL	6
CIN	7
CLE	9
PIT	13
AFC South
HOU	6
IND	16
JAX	10
TEN	11
AFC West
DEN	8
KC	6
OAK	4
SD	9
NFC East
DAL	15
NYG	11
PHI	8
WAS	8
NFC North
CHI	4
DET	8
GB	13
MIN	8
NFC South
ATL	3
CAR	7
NO	5
TB	12
NFC West
ARI	8
STL	1
SF	3
SEA	11

Read More......

Power Rankings 2007 Week 7

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities). Adjusted and unadjusted rankings included.

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	IND	11.4905
2	NE	11.0967
3	DAL	11.0366
4	TB	9.7785
5	PIT	9.5315
6	WAS	9.3918
7	NYG	9.1441
8	TEN	9.0731
9	JAX	8.7938
10	SD	8.7205
11	SEA	8.4443
12	PHI	8.4076
13	MIN	8.4011
14	GB	8.3821
15	DEN	8.2858
16	ARI	7.9302
17	CIN	7.8663
18	CLE	7.7844
19	HOU	7.6298
20	MIA	7.3475
21	CAR	7.2920
22	DET	7.2118
23	KC	7.0735
24	BAL	6.8962
25	ATL	6.8204
26	NO	6.5023
27	CHI	6.2434
28	BUF	6.1996
29	STL	6.0449
30	OAK	5.9200
31	NYJ	5.6766
32	SF	5.5834

Power Rankings, Opponent-adjusted
Rank	Team	Expected Wins
1	IND	11.4905
2	NE	11.0967
3	DAL	11.0366
4	TB	9.7785
5	PIT	9.5315
6	WAS	9.3918
7	NYG	9.1441
8	TEN	9.0731
9	JAX	8.7938
10	SD	8.7205
11	SEA	8.4443
12	PHI	8.4076
13	MIN	8.4011
14	GB	8.3821
15	DEN	8.2858
16	ARI	7.9302
17	CIN	7.8663
18	CLE	7.7844
19	HOU	7.6298
20	MIA	7.3475
21	CAR	7.2920
22	DET	7.2118
23	KC	7.0735
24	BAL	6.8962
25	ATL	6.8204
26	NO	6.5023
27	CHI	6.2434
28	BUF	6.1996
29	STL	6.0449
30	OAK	5.9200
31	NYJ	5.6766
32	SF	5.5834

Power Rankings, Unadjusted
Rank	Team	Expected Wins
1	NE	11.4613
2	IND	11.1798
3	DAL	11.0478
4	PIT	10.6005
5	TB	9.6857
6	WAS	9.3622
7	NYG	9.2132
8	PHI	9.0649
9	SD	9.0305
10	SEA	8.8872
11	MIN	8.6806
12	TEN	8.6470
13	JAX	8.4937
14	BAL	8.3515
15	CAR	8.2901
16	GB	8.0074
17	DEN	7.9614
18	ARI	7.8753
19	DET	7.6292
20	KC	7.6011
21	CIN	7.5983
22	CLE	7.5684
23	HOU	7.4236
24	ATL	6.7335
25	MIA	6.4393
26	OAK	6.2238
27	STL	5.7106
28	NO	5.6194
29	NYJ	5.4792
30	SF	5.4567
31	CHI	5.4103
32	BUF	5.2663

Read More......

Power Rankings 2007 Week 6

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities). Adjusted and unadjusted rankings included.


Using adjusted stats

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	IND	11.1290
2	NE	11.0792
3	DAL	10.7311
4	TB	10.0030
5	WAS	9.8468
6	NYG	9.6412
7	PIT	9.5458
8	JAX	9.2684
9	PHI	9.0883
10	TEN	9.0834
11	SD	8.5522
12	GB	8.4900
13	MIN	8.2463
14	SEA	8.1399
15	HOU	7.9821
16	CLE	7.8416
17	DEN	7.8045
18	ATL	7.6147
19	MIA	7.5847
20	CIN	7.5143
21	ARI	7.4882
22	CAR	7.2857
23	KC	7.2053
24	BAL	7.1100
25	DET	6.6432
26	STL	6.1356
27	BUF	6.1048
28	NO	6.0207
29	CHI	5.9590
30	NYJ	5.7973
31	OAK	5.6081
32	SF	5.4557

Using unadjusted stats

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	NE	11.3883
2	DAL	11.2170
3	IND	11.0017
4	PIT	10.9324
5	TB	10.0555
6	WAS	9.6780
7	JAX	9.5635
8	MIN	9.2922
9	PHI	9.2515
10	NYG	9.0960
11	SD	9.0145
12	SEA	8.5387
13	BAL	8.3563
14	TEN	8.3339
15	CAR	8.2877
16	DEN	8.0485
17	GB	8.0058
18	ARI	7.7705
19	HOU	7.6323
20	CLE	7.5779
21	KC	7.4160
22	DET	7.3518
23	MIA	7.1062
24	CIN	6.9493
25	ATL	6.7962
26	OAK	5.8485
27	STL	5.7550
28	NYJ	5.6024
29	SF	5.2688
30	CHI	5.0450
31	BUF	4.9765
32	NO	4.8422

Read More......

Power Rankings 2007 Week 5

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The first power rankings use linear regression to predict the final score margins of each such game. The score you see along with the rankings is the sum of all of those expected margins.

The second power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

From here on out, the power rankings will be based on the adjusted VOLA (value over league average) stats.

For the logisitc regression rankings based on opponent-adjusted stats, the Colts, Pats, and Cowboys form the top 3, though the Pats are #1 in the unadjusted stats power rankings. As it likely should be, Denver falls to #20 while the Chargers climb from the bottom to reach #18. Miami is still unusually high despite being 0-5 (#13). The unadjusted rankings make more sense, putting the Fins at #22 (about where FO's rankings put them). This might be an indictment of the way I adjust for opponent quality. Tampa Bay and Washington are #8 and #5 respectively (kudos to Football Outsiders). Opponent adjustments bring the Vikings down from #7 (really, really strong run defense) to #12.


Based on opponent-adjusted stats

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	IND	11.1960
2	NE	10.8164
3	DAL	10.1442
4	NYG	9.8800
5	WAS	9.8792
6	PIT	9.6669
7	PHI	9.5689
8	TB	9.4077
9	JAX	9.0579
10	SEA	8.9058
11	HOU	8.7405
12	MIN	8.7397
13	MIA	8.7044
14	TEN	8.4941
15	GB	8.2460
16	ARI	8.1529
17	CLE	7.8998
18	SD	7.8593
19	ATL	7.7065
20	DEN	7.6849
21	CIN	7.5833
22	BAL	7.0963
23	KC	7.0530
24	DET	6.6802
25	BUF	6.4155
26	CAR	6.4130
27	OAK	6.3809
28	STL	5.9181
29	CHI	5.6514
30	SF	5.5273
31	NYJ	5.5242
32	NO	5.0057

Power Rankings, Linear Regression
Rank	Team	Expected Point Differential
1	IND	6.8148
2	NE	5.8376
3	DAL	4.1081
4	NYG	3.7282
5	WAS	3.5712
6	PIT	3.1057
7	PHI	2.7101
8	TB	2.6651
9	JAX	2.1593
10	SEA	1.5993
11	TEN	1.4235
12	HOU	1.3816
13	MIA	1.0793
14	MIN	0.8678
15	GB	0.7365
16	SD	0.1476
17	ARI	0.0165
18	CLE	-0.0998
19	ATL	-0.2166
20	DEN	-0.5562
21	CIN	-0.6609
22	BAL	-1.3696
23	KC	-2.0231
24	CAR	-3.1265
25	BUF	-3.1674
26	DET	-3.1868
27	OAK	-3.3282
28	STL	-4.1517
29	NYJ	-4.5165
30	CHI	-4.6071
31	SF	-5.2781
32	NO	-5.6636

Based on unadjusted stats

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	NE	12.1346
2	DAL	11.6354
3	IND	11.0115
4	PIT	10.9498
5	TB	9.9379
6	WAS	9.7616
7	MIN	9.1091
8	JAX	9.0684
9	PHI	9.0539
10	NYG	8.7927
11	TEN	8.6625
12	HOU	8.6241
13	ARI	8.6119
14	SEA	8.5327
15	GB	8.1879
16	BAL	8.1222
17	DEN	8.0696
18	SD	7.7974
19	MIA	7.5571
20	CAR	7.5435
21	DET	7.3763
22	KC	7.3458
23	CLE	7.0781
24	ATL	7.0505
25	OAK	6.8493
26	CIN	6.6544
27	NYJ	5.5382
28	STL	5.5208
29	SF	5.3025
30	CHI	5.0989
31	BUF	4.9965
32	NO	4.0248

Power Rankings, Linear Regression
Rank	Team	Expected Point Differential
1	NE	8.9585
2	DAL	7.8410
3	IND	6.4341
4	PIT	6.1485
5	TB	3.5604
6	WAS	3.5340
7	JAX	2.2024
8	PHI	1.7959
9	MIN	1.6991
10	TEN	1.5469
11	NYG	1.4447
12	HOU	1.2199
13	SEA	1.0339
14	ARI	0.8937
15	GB	0.6481
16	BAL	0.5864
17	DEN	0.0700
18	SD	-0.1107
19	CAR	-1.1566
20	ATL	-1.4169
21	MIA	-1.4296
22	KC	-1.5473
23	DET	-1.6314
24	CLE	-1.8795
25	OAK	-2.2035
26	CIN	-2.5017
27	NYJ	-4.5871
28	STL	-4.8010
29	SF	-5.8941
30	CHI	-5.8992
31	BUF	-6.5439
32	NO	-8.0149

Read More......

Power Rankings 2007 Week 4

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The first power rankings use linear regression to predict the final score margins of each such game. The score you see along with the rankings is the sum of all of those expected margins.

The second power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

From here on out, the power rankings will be based on the adjusted VOLA (value over league average) stats.

Power Rankings, Linear Regression
Rank	Team	Expected Point Differential
1	IND	5.9404
2	NE	5.6201
3	DAL	5.5001
4	SEA	3.3232
5	NYG	3.2484
6	TB	2.7994
7	DEN	2.7454
8	PHI	2.4690
9	ARI	2.3274
10	HOU	2.2301
11	JAX	1.5286
12	MIN	1.4882
13	WAS	1.2148
14	PIT	1.1036
15	MIA	0.3592
16	TEN	0.3514
17	CIN	0.2501
18	ATL	-0.4528
19	GB	-0.6509
20	BAL	-0.6928
21	CLE	-1.3041
22	DET	-1.5032
23	OAK	-1.7889
24	KC	-1.8304
25	CAR	-2.0165
26	SD	-3.1041
27	BUF	-4.1486
28	NYJ	-4.2575
29	CHI	-4.3569
30	NO	-5.1338
31	STL	-5.5779
32	SF	-5.6809

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	IND	11.0179
2	DAL	10.8174
3	NE	10.6105
4	SEA	9.7515
5	NYG	9.6060
6	DEN	9.5204
7	ARI	9.3961
8	PHI	9.3928
9	TB	9.3426
10	HOU	9.1012
11	MIN	8.9477
12	JAX	8.7916
13	WAS	8.6958
14	PIT	8.6907
15	MIA	8.4258
16	TEN	8.2059
17	CIN	8.0082
18	DET	7.5299
19	ATL	7.4956
20	BAL	7.4924
21	GB	7.4798
22	CLE	7.3255
23	OAK	7.1399
24	KC	7.0549
25	CAR	6.9271
26	SD	6.1265
27	BUF	5.8696
28	CHI	5.8082
29	NYJ	5.6858
30	SF	5.3615
31	NO	5.1985
32	STL	5.1828

Bonus Rankings, based on unadjusted stats

Power Rankings, Linear Regression
Rank	Team	Expected Point Differential
1	NE	10.4453
2	DAL	9.4955
3	IND	6.5819
4	TB	6.0274
5	PIT	5.4543
6	DEN	4.5223
7	SEA	3.1511
8	PHI	1.7881
9	ARI	1.6081
10	MIN	1.5651
11	TEN	1.5560
12	HOU	1.3897
13	WAS	0.7357
14	DET	0.4886
15	NYG	0.4728
16	JAX	0.2966
17	BAL	0.2485
18	KC	0.1948
19	GB	-0.5833
20	CLE	-1.2640
21	MIA	-1.2768
22	ATL	-1.3180
23	CAR	-1.7562
24	OAK	-2.1185
25	CIN	-2.4856
26	SD	-4.0176
27	NYJ	-4.3221
28	CHI	-6.0643
29	SF	-6.5469
30	STL	-6.8063
31	BUF	-7.4840
32	NO	-9.9784

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	NE	12.5345
2	DAL	12.3302
3	IND	11.1651
4	TB	10.9552
5	PIT	10.6023
6	DEN	10.2181
7	SEA	9.5226
8	PHI	9.0219
9	MIN	9.0178
10	ARI	8.9989
11	TEN	8.7841
12	HOU	8.7401
13	DET	8.4257
14	WAS	8.3782
15	JAX	8.2485
16	NYG	8.1665
17	KC	8.1520
18	BAL	7.9735
19	MIA	7.6866
20	GB	7.5178
21	CLE	7.4590
22	CAR	7.2069
23	ATL	6.9473
24	OAK	6.9182
25	CIN	6.6996
26	SD	5.8440
27	NYJ	5.7809
28	CHI	5.0922
29	SF	5.0513
30	STL	4.6379
31	BUF	4.5527
32	NO	3.3702

Read More......

Power Rankings 2007 Week 3

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The first power rankings use linear regression to predict the final score margins of each such game. The score you see along with the rankings is the sum of all of those expected margins.

The second power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

From here on out, the power rankings will be based on the adjusted VOLA (value over league average) stats.

Power Rankings, Linear Regression
Rank	Team	Expected Point Differential
1	IND	5.1372
2	NE	4.6963
3	PHI	4.4882
4	DAL	4.4489
5	HOU	4.2759
6	JAX	2.7364
7	DEN	2.2194
8	PIT	2.0808
9	SEA	1.9967
10	WAS	1.0809
11	NYG	0.9683
12	TB	0.8943
13	TEN	0.3019
14	CAR	-0.0282
15	MIN	-0.3278
16	BAL	-0.3986
17	ARI	-0.4182
18	ATL	-0.4516
19	MIA	-0.4580
20	GB	-0.4797
21	CLE	-1.1080
22	CIN	-1.4180
23	BUF	-1.5398
24	DET	-1.6907
25	SD	-1.7952
26	SF	-2.6270
27	KC	-2.7353
28	NYJ	-2.9009
29	OAK	-2.9807
30	STL	-3.2558
31	CHI	-5.2796
32	NO	-5.4321

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	IND	10.7001
2	DAL	10.5789
3	PHI	10.4336
4	NE	10.3860
5	HOU	10.1382
6	JAX	9.4433
7	DEN	9.2701
8	SEA	9.1180
9	PIT	8.9880
10	WAS	8.6391
11	NYG	8.4719
12	TB	8.2504
13	TEN	8.1991
14	ARI	8.1641
15	MIN	8.0222
16	MIA	7.9489
17	CAR	7.8971
18	BAL	7.6365
19	GB	7.5572
20	ATL	7.5167
21	DET	7.3864
22	CLE	7.2844
23	BUF	7.1819
24	CIN	6.9683
25	SD	6.7951
26	SF	6.7815
27	KC	6.6613
28	OAK	6.5439
29	NYJ	6.3535
30	STL	6.1707
31	CHI	5.4655
32	NO	5.0481

Adddition: For those who are curious, here are the rankings based on unadjusted stats.

Power Rankings, Linear Regression
Rank	Team	Expected Point Differential
1	NE	11.2720
2	PIT	8.1241
3	DAL	7.4265
4	DEN	7.1949
5	IND	7.0557
6	TB	4.9459
7	PHI	3.8690
8	HOU	1.7319
9	TEN	1.5146
10	MIN	1.2234
11	ARI	0.8836
12	CAR	0.7423
13	BAL	0.6645
14	WAS	0.5989
15	SEA	0.4970
16	JAX	0.1947
17	GB	0.1204
18	MIA	-1.0742
19	CIN	-1.1317
20	NYG	-1.3867
21	ATL	-1.7138
22	CLE	-2.0001
23	DET	-2.0462
24	KC	-2.0476
25	SD	-3.2215
26	STL	-3.4520
27	OAK	-3.6659
28	SF	-4.5673
29	NYJ	-6.0060
30	CHI	-6.0192
31	BUF	-9.7513
32	NO	-9.9761

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	NE	12.9813
2	PIT	11.6263
3	DAL	11.6066
4	IND	11.3967
5	DEN	11.3499
6	TB	10.4152
7	PHI	10.0796
8	ARI	8.8136
9	MIN	8.7847
10	HOU	8.7794
11	TEN	8.7690
12	CAR	8.5410
13	WAS	8.3187
14	SEA	8.3100
15	BAL	8.2196
16	JAX	8.2002
17	GB	7.9602
18	MIA	7.6812
19	NYG	7.2284
20	CIN	7.1236
21	KC	7.0974
22	DET	7.0924
23	CLE	7.0654
24	ATL	6.7873
25	OAK	6.3478
26	SD	6.1342
27	STL	6.0664
28	SF	5.9218
29	CHI	5.2842
30	NYJ	4.9917
31	BUF	3.6334
32	NO	3.3927

Read More......

Power Rankings - Week 1

The power rankings for this site were developed in this article. The idea behind them is simple: if each team played one home game and one away game against every other team in the league, who would be the most successful?

The first power rankings use linear regression to predict the final score margins of each such game. The score you see along with the rankings is the sum of all of those expected margins.

The second power rankings use logistic regression to predict the probability of winning. The score you see along with the rankings is the expected win total over a 16-game season based on the expected winning percentages over the 62 games (i.e. the sum of those probabilities).

Power Rankings, Linear Regression
Rank	Team	Expected Point Differential
1	NE	13.7325
2	IND	12.6659
3	HOU	11.3791
4	PIT	10.5062
5	DEN	9.8315
6	DAL	9.6199
7	CAR	7.1694
8	MIN	5.5084
9	DET	5.2108
10	SEA	4.9286
11	PHI	3.8638
12	WAS	3.8512
13	SD	3.3387
14	JAX	3.0969
15	CIN	1.6938
16	ARI	0.9178
17	SF	-0.9178
18	BAL	-1.6938
19	TEN	-3.0969
20	CHI	-3.3387
21	MIA	-3.8512
22	GB	-3.8638
23	TB	-4.9286
24	OAK	-5.2108
25	ATL	-5.5084
26	STL	-7.1694
27	NYG	-9.6199
28	BUF	-9.8315
29	CLE	-10.5062
30	KC	-11.3791
31	NO	-12.6659
32	NYJ	-13.7325

Power Rankings, Logistic Regression
Rank	Team	Expected Wins
1	IND	13.1524
2	NE	13.1070
3	PIT	12.3465
4	DEN	12.1270
5	DAL	12.0315
6	HOU	11.9760
7	CAR	11.4361
8	MIN	10.7716
9	SEA	10.2733
10	DET	10.2539
11	PHI	9.6394
12	WAS	9.5602
13	JAX	9.4403
14	SD	8.9667
15	CIN	8.3876
16	ARI	8.1964
17	SF	7.8036
18	BAL	7.6124
19	CHI	7.0333
20	TEN	6.5597
21	MIA	6.4398
22	GB	6.3606
23	OAK	5.7461
24	TB	5.7267
25	ATL	5.2284
26	STL	4.5639
27	KC	4.0240
28	NYG	3.9685
29	BUF	3.8730
30	CLE	3.6535
31	NYJ	2.8930
32	NO	2.8476

Read More......

Creating Power Rankings

Just as a lark, I figured I'd try devising some power rankings using my model. The basic idea behind them is this: If every team played each other once at home, once at away, who would win the most?

I tested this idea on the 2006 teams after the end of the regular season. Using my adj. rush, adj. pass, sack, adj. 3rd down conv., and turnover data, I trained a linear regression predictor on all games before 2006. Then I used the model to predict the outcomes of all 992 possible games. I created 2 separate rankings: one based on the sum of margins of victory/defeat, one based on the expected winning percentage.

Rank	Team	Sum of Exp. Margins
1	PHI	4.2027
2	SD	4.1368
3	BAL	3.8437
4	JAX	3.2917
5	DAL	2.1643
6	CHI	2.1514
7	PIT	1.9942
8	NO	1.8954
9	IND	1.7482
10	NE	1.5237
11	MIN	1.1935
12	CAR	1.1847
13	ATL	0.94039
14	NYG	0.84285
15	MIA	0.82191
16	GB	0.71736
17	CIN	0.52797
18	DEN	0.33802
19	KC	0.29031
20	SF	-0.685
21	WAS	-1.0359
22	STL	-1.4769
23	NYJ	-1.5889
24	SEA	-2.0378
25	BUF	-2.3402
26	TEN	-2.4492
27	TB	-2.8237
28	ARI	-3.2092
29	DET	-3.6094
30	OAK	-3.7756
31	HOU	-3.8764
32	CLE	-4.9007

Rank	Exp. Win %	Team
1t	BAL	0.83871
1t	PHI	0.83871
3	SD	0.80645
4	JAX	0.72581
5t	CHI	0.66129
5t	NO	0.66129
7	NE	0.64516
8t	DAL	0.62903
8t	CAR	0.62903
10t	PIT	0.6129
10t	IND	0.6129
10t	MIN	0.6129
13t	MIA	0.59677
13t	ATL	0.59677
15	NYG	0.58065
16	GB	0.56452
17t	CIN	0.54839
17t	DEN	0.54839
17t	KC	0.54839
20t	STL	0.43548
20t	SF	0.43548
22t	NYJ	0.41935
22t	WAS	0.41935
24	SEA	0.33871
25	TEN	0.32258
26	BUF	0.27419
27t	OAK	0.19355
27t	ARI	0.19355
29t	CLE	0.17742
29t	HOU	0.17742
29t	DET	0.17742
29t	TB	0.17742

Looking at the bottom, it's about what you'd expect. Seattle, a playoff team, shows up at #24 in both rankings, but they were 8-8 in a weak division and suffered some losses due to injury and free agency. The Jets, another playoff team, show up at #23 and #22 respectively due to weak rushing offense and defense and average passing offense and defense. Seattle and the Jets were #25 and #19 respectively in Football Outsider's 2006 DVOA rankings. Philadelphia, despite a 10-6 record, shows up at #1 with a very strong offense (14.721% VOLA rushing, 20.58% passing unadj.). Jacksonville also obtains a very high ranking (#4) despite its record, a mediocre 8-8, thanks to well above average rush offense and defense. New England is a little low compared to DVOA (#10 and #7, #5) with average and below average YPA numbers but high sack and interception rates. The Super Bowl champion Colts are at #9 and #10 (DVOA #7) with 20.142% VOLA in pass offense and 26.09% VOLA in pass defense(!too high!) but -28.29% VOLA rush defense.

So overall, the rankings turn out very similar to the DVOA rankings with individual teams gaining or losing a couple spots. There are no major disparities, however, that I can see, except maybe Dallas showing up too high in the ranking by points. It would be interesting to calculate these rankings over the season and compare these with other rankings.

Figures and article corrected on 7/12/2007 after finding errors in some box scores.

Read More......