Whoo!  We are going to have some FUN with the responses to my Natural Power Rankings (NAPOs) today!  Why?  The biggest reason, I imagine, is that the team most people see as the worst team in the league... is in the top 5. 

I made a few changes this week to the process - I cut the period of consideration back down to 3 weeks, which I've been hoping to pull off for a while.  It's been impractical, since half the league's teams had byes in that time.  The reasoning was that, while three games is the ideal window, four was better than two.  I'm considering, this time, that two might be better.  Because of this, I had to establish a new rule.

For those who are new readers to NAPOs, I'm going to explain a few things, though it's not going to stop you from bashing me.  Everyone else, you can skip to the rankings, or at least to rule D if you're interested in the new change.

1^st.  The goal of NAPOs is to gauge who in the league is most primed for a win and who is primed for a loss.  It's a ranking of the teams, from most likely to win to most likely to lose, only for the following Sunday.  It has no weight for the rest of the season or who I think will be in the playoffs.  This wasn't the original goal, but as they have proved surprisingly accurate at predicting "upsets," that's how the purpose has evolved.

2^nd.  Again, I'm only looking at the past three weeks.  Some say this is short-sighted; I say a lot has changed since week 3.

3^rd.  This is purely mathematical and my opinion has nothing to do with it. 

4^th.  I have developed a formula to determine Strength Of Victory (SOV) that calculates who "won better."

Here's how the NAPOs are determined.  I progress through the following steps until everyone is ranked.

• A) If team A beat team B, A is ranked higher.
• B) If team A and team B both beat team C, the team with a higher SOV is ranked higher. If A and B both lost to C, the team with a lower SOV is ranked higher.
• C) I follow similar logic progressing through degrees of separation. The next degree is 1.5, where team A lost to team D, and team B lost to team C, who lost to team D. I add up the D/C and C/B SOV and compare it to the D/A SOV. If D/A is lower, A is ranked higher than B. (Lost yet?) Same thing at 2 degrees, 2.5 degrees, 3 degrees, etc. But I only follow chains of all wins or all losses, otherwise it gets too complicated (gotta draw a line somewhere).

•D)    This is the new rule.  If team A and team B do not share a degree, I look at all of the scores involving either team.  I find the lowest SOV, and rank those two teams as close together as possible.  Then I do the same with the second-lowest until all the teams are ranked.

This happened with Houston, Atlanta, and San Francisco.  I had Indianapolis over Tennessee after steps A-C had been completed, followed by Tennessee, Oakland, and Carolina.  HOU, ATL, and SFO needed to fit between IND and TEN, but the question was, where?  So I looked at the scores.  The lowest SOV was SFO-ATL, with an SOV of 1.59.  So they wouldn't be split up.  The next lowest was HOU-OAK, with an SOV of 2.70.  So I put them as close together as possible, giving me an order of IND, ATL, SFO, HOU, TEN, OAK, CAR.

So here's the complete order, with the record over the past three weeks in parentheses:

• 1) New England (2-0)
• 2) Dallas (2-0)
• 3) Buffalo (3-0)
• 3) NY Giants (1-1) - my first actual tie. They have the same SOV over Miami (identical 13-10 score, for an SOV of 1.39).
• 5) Miami (0-2)
• 6) Pittsburgh (3-0)
• 7) Cleveland (2-1)
• 8) Seattle (1-1)
• 9) Cincinnati (1-2)
• 10) St. Louis (1-1)
• 11) New Orleans (2-1)
• 12) Jacksonville (2-1)
• 13) Tampa Bay (1-1)
• 14) Arizona (1-1)
• 15) Detroit (2-1)
• 16) Green Bay (3-0)
• 17) Chicago (1-1)
• 18) Denver (1-2)
• 19) Kansas City (0-2)
• 20) Philadelphia (2-1)
• 21) Minnesota (1-2)
• 22) San Diego (2-1)
• 23) Indianapolis (1-2)
• 24) Atlanta (2-0)
• 25) San Francisco (0-3)
• 26) Houston (1-1)
• 27) Tennessee (2-1)
• 28) Oakland (0-3)
• 29) Carolina (0-3)
• 30) Washington (1-2)
• 31) NY Jets (0-2)
• 32) Baltimore (0-2)

So here are the predictions for week 11:

Tampa Bay over Atlanta

NY Giants over Detroit

Miami over Philadelphia - if this actually happens, you may all call me His Dudeness.

Minnesota over Oakland

Jacksonville over San Diego

Kansas City over Indianapolis

New Orleans over Houston

Green Bay over Carolina

Cincinnati over Arizona

Cleveland over Baltimore

Pittsburgh over NY Jets

Dallas over Washington

Seattle over Chicago

St. Louis over San Francisco

New England over Buffalo

Denver over Tennessee

It's worth mentioning that the NAPOs only missed four games last week - only one of the top 50 on the Peter King Challenge did better.  The top pickers have a correct % of .722.  Since I've been using NAPOs to pick games, I'm at .712, which would be good for 3^rd place out of over 15000 people.  My personal predictions are at .597.  So unless your name is phdrunner15 or sam65010 (the top two), think twice about bashing me.