Home > NFL > NFL Power Rankings: Week One

NFL Power Rankings: Week One

Judging a team based solely on its record is flawed (see: 2011 Pittsburgh Pirates, 1st half). A team sometimes wins a game where they were outplayed but, say, got a few good bounces and some breaks from the officials. Conversely the team on the opposite end of those lousy bounces and missed calls gets a win that they probably deserved taken away from them. To make sense of this madness people have created power rankings, which are theoretically designed to get down to the bare bones of where teams rank in relation to each other in spite of these unfair wins and losses.
Unfortunately, power rankings tend to be assembled based on the opinion of one, or a few, or even a panel of “Experts.” In the NFL, somehow this entity (be it an individual, tribunal, quorum, whatever) is supposed to have enough knowledge of 32 teams that it can accurately rank where each team exists in relation to one another. To me, this is insane.
ESPN has the Raiders ranked 18th and the Redskins 19th. The raiders won on the road last night, defeating the Broncos, and the Redskins handled a banged-up Giants team at home. How do the voters at ESPN determine that the Raiders are slightly better than the Redskins? Before the season started, the Raiders were ranked 22nd and the Redskins 28th. I cannot say with certainty how ESPN’s panel of Experts arrived at these rankings, but it would seem as though the only available data is 1) last year’s performances, 2) offseason personnel changes, and 3) preseason records. But why do rankings formulated mostly on last years’ performance, personnel changes of unknown significance, and exhibition games factor in to how a team will perform this year. Experts hypothesize on what the Raiders will be like, but before we see them play can we realistically have any idea?
Let’s take a look at how the five Experts at ESPN voted for these two teams after week one:
Sando Clayton Kuharsky Walker Fox Average
Raiders 18 25 14 24 12 18.6
Redskins 17 18 21 18 22 19.2
So:
  • Clayton and walker think the Raiders are one of the nine worst teams in football while Kuharsky and Fox believe they are in the top half.
  • In spite of a victory, Clayton demotes the Raiders three spots from where the panel had them ranked before the year began.
  • Three Experts think the Redskins are the better team, while two Experts think the Raiders are better.
How is this possible? How can five Experts see the same team so differently? How is it that three of the five Experts can think the Redskins are a better team, yet they are ranked below the Raiders.
These rankings raise more questions than answers. Doesn’t that defeat the purpose of power rankings?
The biggest issue I have with the rankings at ESPN (as well as all examples given above, I just chose to pick on ESPN because their website makes it particularly easy) is that they appear to be the opinions of five people, averaged out. Do you really think those five Experts watched all 16 games this weekend and can accurately rank all 32 teams in relation to one another?
This is why I use numbers.
Numbers are not perfect on their own. But if you use them together, over time you can begin to see a more accurate picture of just how good or bad a team is. When I compile my rankings I literally enter numbers from a variety of categories that I think are important into a spreadsheet. These numbers are weighted based on how important I think they are. For instance, I believe yards allowed per game is a more important statistic than yards gained per game, and so I weigh them accordingly.
I use running averages throughout the year, so week one often ends up looking a little strange. Over the course of 17 weeks things even out. Teams like the Bills (who are #1 this week) regress to the mean. Teams like the Steelers (#31) stop committing seven turnovers per game (one of the big reasons they’re all the way down there; turnover differential is part of the formula) and their ranking moderates accordingly.
The Rankings_Index on the far right is the number the spreadsheet spits out at me when all the weighted numbers are averaged out. The closer a team is to a 1.0, the better.

 TEAM

LAST_WEEK

RECORD

RANKINGS_INDEX

1.

Bills

 1-0

        .7430

2.

Ravens

 1-0

        .7270

3.

Texans

1-0

        .6908

4.

Eagles

1-0

        .6409

5.

49ERS

 1-0

        .6392

6.

BEARS

 1-0

        .6343

7.

Patriots

 1-0

        .6268

8.

Chargers

 1-0

        .6267

9.

Redskins

 1-0

        .6200

10.

Lions

 1-0

        .6092

11.

Bengals

 1-0

        .6042

12.

Raiders

 1-0

        .5710

13.

Packers

 1-0

        .5694

14.

Jaguars

 1-0

        .5662

15.

Jets

 1-0

        .5617

16.

Cardinals

 1-0

        .5534

17.

Cowboys

 0-1

        .4478

18.

Panthers

 0-1

        .4470

19.

Broncos

 0-1

        .4462

20.

Titans

 0-1

        .4444

21.

Saints

 0-1

        .4398

22.

Browns

 0-1

        .4127

23.

Giants

 0-1

        .4017

24.

Buccaneers

 0-1

        .3932

25.

Seahawks

 0-1

        .3918

26.

Vikings

 0-1

        .3856

27.

Falcons

 0-1

        .3781

28.

Rams

 0-1

        .3698

29.

Dolphins

 0-1

        .3608

30.

Colts

 0-1

        .3224

31.

Steelers

 0-1

        .2847

32.

Chiefs

0-1        .2735
Advertisements
Categories: NFL Tags:
  1. No comments yet.
  1. No trackbacks yet.

What's on your mind?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: