UPDATED: Pythagenport Relative Power Index Week 5

by Richard Wade on Oct 8, 2008 7:11 PM PDT

Here's this week's Power Rankings updated this morning.

Rk	Team	G	PF	PA	PCT	Opp	OppOpp	RPI	PF/G	PA/G	LW
1	Washington Redskins	5	109	98	.576	.710	.465	.615	21.8	19.6	3
2	Philadelphia Eagles	5	127	97	.689	.551	.547	.585	25.4	19.4	1
3	Chicago Bears	5	128	87	.755	.522	.535	.583	25.6	17.4	4
4	Carolina Panthers	5	114	70	.796	.498	.539	.583	22.8	14.0	7
5	Tampa Bay Buccaneers	5	114	94	.636	.601	.485	.581	22.8	18.8	2
6	Tennessee Titans	5	115	56	.878	.393	.593	.564	23.0	11.2	10
7	Arizona Cardinals	5	147	120	.653	.536	.516	.560	29.4	24.0	17
8	San Diego Chargers	5	148	129	.606	.552	.515	.556	29.6	25.8	8
9	Dallas Cowboys	5	151	111	.721	.441	.602	.551	30.2	22.2	5
10	Baltimore Ravens	4	75	56	.688	.491	.523	.548	18.8	14.0	13
11	Miami Dolphins	4	79	74	.546	.563	.515	.547	19.8	18.5	25
12	Minnesota Vikings	5	101	109	.445	.632	.475	.546	20.2	21.8	14
13	New Orleans Saints	5	138	130	.546	.538	.558	.545	27.6	26.0	6
14	New York Jets	4	115	116	.493	.576	.512	.539	28.8	29.0	19
15	Pittsburgh Steelers	5	103	79	.677	.440	.555	.528	20.6	15.8	9
16	New York Giants	4	127	49	.943	.259	.632	.523	31.8	12.3	11
17	Denver Broncos	5	149	130	.606	.445	.548	.511	29.8	26.0	16
18	Green Bay Packers	5	133	128	.530	.489	.535	.511	26.6	25.6	12
19	Cincinnati Bengals	5	74	118	.211	.685	.432	.503	14.8	23.6	21
20	Jacksonville Jaguars	5	100	111	.425	.560	.443	.497	20.0	22.2	18
21	Atlanta Falcons	5	117	107	.566	.432	.543	.493	23.4	21.4	23
22	Indianapolis Colts	4	83	94	.409	.456	.578	.475	20.8	23.5	24
23	Cleveland Browns	4	46	78	.197	.574	.514	.465	11.5	19.5	26
24	Houston Texans	4	83	130	.199	.597	.462	.464	20.8	32.5	20
25	Buffalo Bills	5	126	104	.639	.329	.544	.460	25.2	20.8	15
26	New England Patriots	4	79	79	.500	.396	.524	.454	19.8	19.8	30
27	Oakland Raiders	4	78	101	.318	.493	.471	.444	19.5	25.3	22
28	San Francisco 49ers	5	115	127	.426	.400	.509	.434	23.0	25.4	29
29	Kansas City Chiefs	5	65	131	.120	.557	.453	.422	13.0	26.2	31
30	St. Louis Rams	4	43	147	.024	.624	.412	.421	10.8	36.8	27
31	Detroit Lions	4	66	147	.077	.569	.461	.419	16.5	36.8	32
32	Seattle Seahawks	4	83	124	.225	.508	.403	.411	20.8	31.0	28

The Redskins have taken over the top spot, which I think is where I'd rank them were I doing this subjectively. Three teams moved at least 10 spots from last week's rankings. Miami's win over the Chargers moved them from 25th to 11th. Arizona and Buffalo moved in opposite directions based on the outcome of their game. Both the Steelers and Giants dropped despite winning, though only the Giants fell five spots while improving their pythagenport winning percentage. They fell entirely due to strength of schedule.

UPDATE: I thought about it a bit and then decided to steal the idea of reducing opponent adjustments by 50%. The new rankings certainly agree more with DVOA, so if you believe in those numbers these are better:

Team	G	PF	PA	PCT	Opp	OppOpp	RPI
New York Giants	4	127	49	.943	.259	.632	.733
Tennessee Titans	5	115	56	.878	.393	.593	.721
Carolina Panthers	5	114	70	.796	.498	.539	.690
Chicago Bears	5	128	87	.755	.522	.535	.669
Philadelphia Eagles	5	127	97	.689	.551	.547	.637
Dallas Cowboys	5	151	111	.721	.441	.602	.636
Baltimore Ravens	4	75	56	.688	.491	.523	.618
Tampa Bay Buccaneers	5	114	94	.636	.601	.485	.608
Arizona Cardinals	5	147	120	.653	.536	.516	.606
Pittsburgh Steelers	5	103	79	.677	.440	.555	.602
Washington Redskins	5	109	98	.576	.710	.465	.596
San Diego Chargers	5	148	129	.606	.552	.515	.581
Denver Broncos	5	149	130	.606	.445	.548	.558
Buffalo Bills	5	126	104	.639	.329	.544	.550
Miami Dolphins	4	79	74	.546	.563	.515	.546
New Orleans Saints	5	138	130	.546	.538	.558	.546
Atlanta Falcons	5	117	107	.566	.432	.543	.529
Green Bay Packers	5	133	128	.530	.489	.535	.520
New York Jets	4	115	116	.493	.576	.512	.516
Minnesota Vikings	5	101	109	.445	.632	.475	.495
New England Patriots	4	79	79	.500	.396	.524	.477
Jacksonville Jaguars	5	100	111	.425	.560	.443	.461
Indianapolis Colts	4	83	94	.409	.456	.578	.442
San Francisco 49ers	5	115	127	.426	.400	.509	.430
Oakland Raiders	4	78	101	.318	.493	.471	.381
Cincinnati Bengals	5	74	118	.211	.685	.432	.357
Houston Texans	4	83	130	.199	.597	.462	.331
Cleveland Browns	4	46	78	.197	.574	.514	.331
Seattle Seahawks	4	83	124	.225	.508	.403	.318
Kansas City Chiefs	5	65	131	.120	.557	.453	.271
Detroit Lions	4	66	147	.077	.569	.461	.248
St. Louis Rams	4	43	147	.024	.624	.412	.223

This FanPost was written by a member of the Bolts From The Blue community and does not necessarily reflect the views of the Bolts From The Blue editors or SB Nation.

Tweet Comment 21 comments | 0 recs |

Do you like this story?

Comments

Hochuli and Gramatica included in the calculations?

How are Tampa Bay, New Orleans, and San Diego all ranked ahead of Denver when Denver has beaten all of them?

Honest question… just curious how this works.

by Leukadian on Oct 8, 2008 9:18 AM PDT reply actions

Well I'll tell you

There is no subjectivity at all. It’s all based on the numbers you see above.

People the world over have always been more impressed by the power of our example than by the example of our power.

by Sam (sdsuaztec4) on Oct 8, 2008 11:32 AM PDT up reply actions

No

Because this is based entirely off of total points for and against combined off the strength of opponents (also based off of points for and against). For instance, because of the Hochuli mistake the Broncos have an extra 8 points (15 if you count the botched Bailey int call) and the Vikings an extra 7. If you used a stat based on the strength of each play run a team runs based on offense, defense and special teams combine with opponent strength then the effects of the Hochuli incidents would be minimized (it’s only 1-3 plays among many many plays). Gramatica would still count though, since he is the one (not) executing plays.

Homer: Ohhhh, The Denver Broncos.
Marge: Whats wrong with the Denver Broncos?
Homer: Marge you just don't understand football.

by Wonko on Oct 8, 2008 12:47 PM PDT up reply actions

From My Week 3 Post:

Pythagenport is (pfx)/(pfx + pfx) where x is ((pf+pf)/g).285.

RPI is 25% Pythagenport Winning Percentage, 50% Opponents’ Pythagenport Winning Percentage and 25% Opponents’ Opponents’ Pythagenport Winning Percentage.

The rest of the stats should already be familiar.

Intentional Blog on Balls

by Richard Wade on Oct 8, 2008 3:43 PM PDT up reply actions

ouuuuuuuuuch

MAAAAAAAAATH!!! I HATE MAAAAATH!!!

RAAAAAARRRRR

(d’s for degrees got me through stats in case you couldnt tell. thank god poli sci majors dont have to take calculus)

www.wellbelowthemendozaline.blogspot.com

by justdave on Oct 8, 2008 8:33 PM PDT up reply actions

so PA is not factored into RPI or PCT

and Pythagenport from your explanation = PCT in the chart

by Leukadian on Oct 9, 2008 12:19 PM PDT up reply actions

Oh, son of a -

I really should have proofread that. Sorry.

(pf ^ x)/(pf ^ x + pa ^ x), where x is ((pf+pa)/g)^.285

Intentional Blog on Balls

by Richard Wade on Oct 9, 2008 1:38 PM PDT up reply actions

whoa... x is an exponent

that is some heavy math…

And it now makes more sense why the Broncos are ranked so low (previously their PF should have had them nearer the top of the heap).

I like the new weighting of SOS… the rankings seem more reasonable to me. Although Carolina is a bit of a surprise.

Interesting that the PCT for the Chargers and Broncos is identical.

by Leukadian on Oct 9, 2008 2:53 PM PDT up reply actions

Chargers and Broncos with same PCT

Makes sense, since their PF and PA is almost identical. I guess it’s not the interesting…

by Leukadian on Oct 9, 2008 2:55 PM PDT up reply actions

Yeah, I can see why you were confused. There were something like 8 typos in my first posting of the equation.

It’s not quite identical. The Chargers’ is slightly (and I must stress slightly) better.

Intentional Blog on Balls

by Richard Wade on Oct 9, 2008 3:03 PM PDT up reply actions

I'm still curious if their is a correlation between these numbers and future winning percentage

I know there is a strong correlation in baseball. But if the correlation were there in football it seems like stat based sites like football outsiders would use Pythagenport instead of DVOA.

So is there any advantage that anyone knows of to using this over DVOA?

Homer: Ohhhh, The Denver Broncos.
Marge: Whats wrong with the Denver Broncos?
Homer: Marge you just don't understand football.

by Wonko on Oct 8, 2008 12:40 PM PDT reply actions

Good question.

I’d like to know the answer too if anyone can explain.

We're gonna dazzle you with our "sub par" play.

by soulSD on Oct 8, 2008 1:00 PM PDT up reply actions

FO does reference Pythagorean records.

Intentional Blog on Balls

by Richard Wade on Oct 8, 2008 3:44 PM PDT up reply actions

BTW

I’m not try to criticize anything here. I’m just making observations. Every stat has its own role in interpreting the data.

One big difference I see between this and DVOA is the NY Giants. DVOA has them as the #1 team in football on the strength of its #1 ranked offense and #7 ranked defense. Pythagenport has them at #16. A lot of it is due to the strength of schedule (Was, StL, Cin, Sea). Pythagenport seems to weight those points for and against as average performance whereas DVOA sees the plays they made on par with what would be expected of a top team vs far inferior opponents. I would think Pythagenport wouldn’t have a problem with a top team beating Sea and StL by the scores it did, but it is really hitting them hard for the close Cincinnati game.

Homer: Ohhhh, The Denver Broncos.
Marge: Whats wrong with the Denver Broncos?
Homer: Marge you just don't understand football.

by Wonko on Oct 8, 2008 12:59 PM PDT reply actions

The Giants are #1 by Pythagenport. They’re taking a hit from having so 75% of their games against terrible teams. I should probably change the weighting based on games played, but I haven’t looked at it closely enough to know what that weighting should be. If you have thoughts on that, I’d be interested to hear them.

Intentional Blog on Balls

by Richard Wade on Oct 8, 2008 3:47 PM PDT up reply actions

Also, when you look at points rather than plays you’re probably going to get a less accurate picture just as you would if you went by wins rather than points.

Intentional Blog on Balls

by Richard Wade on Oct 8, 2008 3:48 PM PDT up reply actions

Oops

I didn’t pay attention to the ordering by RPI. So it’s the RPI that is having a tough time with the Giants’ opponents.

Homer: Ohhhh, The Denver Broncos.
Marge: Whats wrong with the Denver Broncos?
Homer: Marge you just don't understand football.

by Wonko on Oct 8, 2008 4:00 PM PDT up reply actions

LOL

New Update Still has the Bolts over the donks.

People the world over have always been more impressed by the power of our example than by the example of our power.

by Sam (sdsuaztec4) on Oct 8, 2008 11:17 PM PDT reply actions

Take heart

I’ve been running down a rankings rabbit hole, and I found this guy who I thought Charger fans might find encouraging: http://fspi.blogspot.com/2008/10/nfl-week-5-power-rankings.html

by Leukadian on Oct 9, 2008 3:53 PM PDT reply actions