NFL Week 9 Picks

2008 Navigation Links

Picks for week 9, based on the week 8 rankings.
Read this for a description of APR.
Read this for a description of SRS.
Line values are taken from yahoo.com (except when noted otherwise), and are for comparison purposes only.

Game	APR	SRS	Line
Detroit Lions @ Chicago Bears	Bears ρ=1.4229	Bears δ=21.67	Bears -13
Tampa Bay Buccaneers @ Kansas City Chiefs	Buccaneers ρ=1.3477	Buccaneers δ=22.08	Buccaneers -9
Jacksonville Jaguars @ Cincinnati Bengals	Jaguars ρ=1.2392	Jaguars δ=11.16	Jaguars -7½
Philadelphia Eagles @ Seattle Seahawks	Eagles ρ=1.1926	Eagles δ=15.60	Eagles -7
Atlanta Falcons @ Oakland Raiders	Falcons ρ=1.1633	Falcons δ=13.62	Falcons -3
Arizona Cardinals @ St. Louis Rams	Cardinals ρ=1.1442	Cardinals δ=15.47	Cardinals -3
Green Bay Packers @ Tennessee Titans	Titans ρ=1.1342	Titans δ=6.01	Titans -5½
New York Jets @ Buffalo Bills	Bills ρ=1.0966	Bills δ=2.83	Bills -5½
Miami Dolphins @ Denver Broncos	Dolphins ρ=1.0956	Dolphins δ=5.53	Broncos -3
Houston Texans @ Minnesota Vikings	Vikings ρ=1.0544	Vikings δ=3.17	Vikings -4½
Baltimore Ravens @ Cleveland Browns	Ravens ρ=1.0541	Ravens δ=2.37	Browns -1½
Dallas Cowboys @ New York Giants	Giants ρ=1.0392	Giants δ=3.99	Giants -8
New England Patriots @ Indianapolis Colts	Colts ρ=1.0059	Colts δ=4.48	Colts -5½
Pittsburgh Steelers @ Washington Redskins	Steelers ρ=1.0045	Steelers δ=4.65	Redskins -1½

Bye: Carolina Panthers, New Orleans Saints, San Francisco 49ers, San Diego Chargers

Comments:

• Disagreements on Dolphins @ Broncos, Ravens @ Browns, and Steelers @ Redskins.

• APR rates the bottom two games as virtual toss-ups.

• The Titans seemed pretty beatable Monday night (the Colts had the lead for quite a while, and the Titans didn't pull ahead by more than 3 until late in the 4th quarter). The Ravens are the only team the Titans have played with a winning record, and the only one currently in the top half of the APR power rankings.

NFL Week 8 Power Rankings

2008 Navigation Links
Read this for a description of APR

Rank	Team	W-L	Power Index	This week
1 (-)	Tennessee Titans	7-0	1.5221 (LW × 0.999)	won by 10 vs Colts
2 (-)	Tampa Bay Buccaneers	5-3	1.4424 (LW × 0.983)	lost by 4 at Cowboys
3 (↑1)	Chicago Bears	4-3	1.4105 (LW × 1.000)	Bye
4 (↑3)	Philadelphia Eagles	4-3	1.4083 (LW × 1.006)	won by 13 vs Falcons
5 (↑10)	New York Giants	6-1	1.4053 (LW × 1.074)	won by 7 at Steelers
6 (↓1)	Carolina Panthers	6-2	1.4048 (LW × 1.000)	won by 4 vs Cardinals
7 (↑2)	Baltimore Ravens	4-3	1.3942 (LW × 1.022)	won by 19 vs Raiders
8 (↓5)	Pittsburgh Steelers	5-2	1.3696 (LW × 0.946)	lost by 7 vs Giants
9 (↑1)	Washington Redskins	6-2	1.3635 (LW × 1.005)	won by 8 at Lions
10 (↓2)	Arizona Cardinals	4-3	1.3594 (LW × 0.994)	lost by 4 at Panthers
11 (↑3)	Dallas Cowboys	5-3	1.3522 (LW × 1.032)	won by 4 vs Buccaneers
12 (-)	Green Bay Packers	4-3	1.3420 (LW × 1.014)	Bye
13 (↓7)	Buffalo Bills	5-2	1.3340 (LW × 0.950)	lost by 9 at Dolphins
14 (↑4)	Cleveland Browns	3-4	1.3227 (LW × 1.045)	won by 6 at Jaguars
15 (↓2)	San Diego Chargers	3-5	1.2889 (LW × 0.976)	lost by 5 at Saints
16 (↑7)	Miami Dolphins	3-4	1.2878 (LW × 1.050)	won by 9 vs Bills
17 (-)	Indianapolis Colts	3-4	1.2841 (LW × 1.002)	lost by 10 at Titans
18 (↓2)	Atlanta Falcons	4-3	1.2825 (LW × 0.984)	lost by 13 at Eagles
19 (↑1)	New England Patriots	5-2	1.2765 (LW × 1.014)	won by 7 vs Rams
20 (↓9)	Jacksonville Jaguars	3-4	1.2745 (LW × 0.960)	lost by 6 vs Browns
21 (↓2)	Minnesota Vikings	3-4	1.2709 (LW × 1.004)	Bye
22 (↓1)	New Orleans Saints	4-4	1.2626 (LW × 1.006)	won by 5 vs Chargers
23 (↓1)	New York Jets	4-3	1.2165 (LW × 0.980)	won by 4 vs Chiefs
24 (↑2)	Houston Texans	3-4	1.2054 (LW × 1.033)	won by 29 vs Bengals
25 (↓1)	St. Louis Rams	2-5	1.1880 (LW × 0.996)	lost by 7 at Patriots
26 (↑2)	Seattle Seahawks	2-5	1.1808 (LW × 1.062)	won by 21 at 49ers
27 (↓2)	Denver Broncos	4-3	1.1754 (LW × 0.987)	Bye
28 (↓1)	Oakland Raiders	2-5	1.1024 (LW × 0.968)	lost by 19 at Ravens
29 (-)	Kansas City Chiefs	1-6	1.0702 (LW × 1.010)	lost by 4 at Jets
30 (-)	San Francisco 49ers	2-6	1.0303 (LW × 0.975)	lost by 21 vs Seahawks
31 (-)	Cincinnati Bengals	0-8	1.0285 (LW × 0.981)	lost by 29 at Texans
32 (-)	Detroit Lions	0-7	0.9913 (LW × 1.008)	lost by 8 vs Redskins

Comments:

• Parity: there are 25 teams with at least 3 losses. There are 25 teams with at least 3 wins (Obviously(?) not the same 25).

• The Buccaneers' loss drops their power index, but they maintain the #2 spot because the former #3 Steelers dropped out of the top 5, and the Bears are still a long way down.

• The Giants beat their first top-10 team since week 1. They've got the Cowboys and Eagles coming up, so they finally have their chance to prove (or disprove) they belong at the top of the rankings.

• The Browns have climbed to #14 based on the strength of a couple of big wins (Giants and Jaguars) and a close loss to the Redskins.

Teams ESPN likes more than APR:

Patriots (+12)

The Patriots are APR's lowest ranked 5-win team. But they haven't beat a team ranked higher than #23 (the Jets). They've got a bunch of higher-ranked teams yet to play (Colts, Bills, Dolphins, Steelers, Cardinals), so we'll find out if the Patriots are who APR thinks they are.

Broncos (+12)

ESPN calls the Broncos "free-falling", but move them up 3 spots on the bye. Their only road win was at the Raiders, and they gave the Chiefs their only win.

Teams APR likes more than ESPN:

Bears (-9)

The Bears' losses have come by a combined 8 points, and APR gives credit for close losses.

Browns (-8)

Their win over the Giants might have been a fluke, and the win Jaguars was pretty ugly, but they're still wins.

Dolphins (-8)

It would be easier to defend the Dolphins if they could get out of their "win one, lose one" pattern. Still, ESPN's 24th place seems kind of low for a team that's beat the Patriots, Chargers, and Bills.

NFL Week 8 Pick Results

2008 Navigation Links
Read this for a description of APR.
Read this for a description of SRS.
Line picks counted correct when the favored team wins; they do not have to beat the spread to count as a win.

Pick results for week 8.

Unanimous picks (8-4)

Correctly picked:

Redskins @ Lions

Raiders @ Ravens

Chiefs @ Jets

Bengals @ Texans

Falcons @ Eagles

Rams @ Patriots

Cardinals @ Panthers

Colts @ Titans

Not so much:

Bills @ Dolphins

The Dolphins remember how to play defense, and maybe the Bills have been playing over their head lately.

Giants @ Steelers

The Steelers offense just could not move the ball. Looking back at their schedule so far, the Steelers are 1-3 against teams with a reasonable defense (Giants, Eagles, Ravens). Combine with their numerous injuries, the Steelers first-half season successes could fade quickly.

Chargers @ Saints

Chargers played a sloppy game early, and the Saints jumped out to a big lead. The Chargers rallied late, but ran out of time.

Browns @ Jaguars

The Jags are another team that had trouble moving the ball. The Browns were held to just two field goals in the second half, but the Jaguar offense just couldn't get it done.

Split Picks (APR 1-1) (SRS 0-2) (Line 1-1)

Buccaneers (APR, SRS) @ Cowboys (Line)

Apparently the Cowboys remembered how to play defense. 14 points would've won the game for the Bucs, but they just couldn't get in the endzone.

Seahawks (APR) @ 49ers (SRS, Line)

Wow, that wasn't even close. Seahawks are a bad team this year, but apparently not as bad as the 49ers, at least not this week.

Standings:

	This week	Overall
APR:	9-5	62-38
SRS:	8-6	57-43
Line:	9-5	65-34-1
Unanimous:	8-4	48-21

New power rankings, and next weeks picks to come.

NCAA: APR's Top 30+, Week 9

2008 Navigation Links.
Read this for a description of APR.
Read this for some disclaimers about applying APR to the NCAA.

I'll update with poll information (and probably add a few more teams) once it's posted. Update: Done.

Rank	Team	W-L	Power Index	This week
1 (-)	Texas BCS=1	8-0	1.7584 (LW × 0.980)	won by 4 vs Oklahoma-St
2 (↑1)	Penn-St BCS=3	9-0	1.6911 (LW × 1.031)	won by 7 at Ohio-St
3 (↓1)	Florida BCS=8	6-1	1.6457 (LW × 0.988)	won by 58 vs Kentucky
4 (↑1)	USC BCS=5	6-1	1.6415 (LW × 1.025)	won by 7 at Arizona
5 (↑4)	Texas-Tech BCS=7	8-0	1.6274 (LW × 1.038)	won by 42 at Kansas
6 (↓2)	Oklahoma BCS=4	7-1	1.6171 (LW × 0.995)	won by 23 at Kansas-St
7 (↑1)	Oklahoma-St BCS=9	7-1	1.5934 (LW × 1.014)	lost by 4 at Texas
8 (↓1)	Alabama BCS=2	8-0	1.5931 (LW × 1.008)	won by 20 at Tennessee
9 (↑6)	North-Carolina BCS=22	6-2	1.5836 (LW × 1.077)	won by 21 vs Boston-Coll
10 (-)	Georgia BCS=6	7-1	1.5785 (LW × 1.010)	won by 14 at LSU
11 (-)	Missouri BCS=14	6-2	1.5516 (LW × 0.996)	won by 58 vs Colorado
12 (↓6)	Ohio-St BCS=12	7-2	1.5274 (LW × 0.965)	lost by 7 vs Penn-St
13 (↑1)	Boise-St BCS=11	7-0	1.5258 (LW × 1.038)	won by 17 at San-Jose-St
14 (↑2)	Ball-St BCS=16	8-0	1.4770 (LW × 1.008)	won by 22 vs East-Michigan
15 (↑27)	Connecticut BCS=25	6-2	1.4606 (LW × 1.094)	won by 24 vs Cincinnati
16 (↑5)	Michigan-St BCS=21	7-2	1.4527 (LW × 1.012)	won by 14 at Michigan
17 (↑3)	Utah BCS=10	8-0	1.4515 (LW × 1.011)	Bye
18 (↑22)	Virginia	5-3	1.4509 (LW × 1.086)	won by 7 at Georgia-Tech
19 (-)	Minnesota BCS=17	7-1	1.4497 (LW × 1.005)	won by 11 at Purdue
20 (↑9)	Florida-St BCS=15	6-1	1.4485 (LW × 1.044)	won by 10 vs Virginia-Tech
21 (↓4)	Tulsa BCS=18	8-0	1.4421 (LW × 0.994)	won by 30 vs UCF
22 (↓9)	TCU BCS=13	8-1	1.4332 (LW × 0.968)	won by 47 vs Wyoming
23 (↑1)	Iowa	5-3	1.4189 (LW × 0.997)	Bye
24 (↑13)	Miami-FL	5-3	1.4029 (LW × 1.037)	won by 6 vs Wake-Forest
25 (↓3)	Nebraska	5-3	1.3942 (LW × 0.973)	won by 12 vs Baylor
26 (↑6)	Oregon-St	4-3	1.3891 (LW × 1.009)	Bye
27 (↓9)	Virginia-Tech	5-3	1.3891 (LW × 0.962)	lost by 10 at Florida-St
28 (↑8)	Notre-Dame	5-2	1.3774 (LW × 1.011)	won by 26 at Washington
29 (↑2)	Mississippi	4-4	1.3755 (LW × 0.999)	won by 2 at Arkansas
30 (↓18)	Pittsburgh	5-2	1.3718 (LW × 0.917)	lost by 20 vs Rutgers
31 (↓8)	South-Florida BCS=23	6-2	1.3703 (LW × 0.962)	lost by 4 at Louisville
32 (↓5)	Boston-Coll	5-2	1.3621 (LW × 0.968)	lost by 21 at North-Carolina
33 (↑20)	Oregon BCS=24	6-2	1.3494 (LW × 1.044)	won by 34 at Arizona-St
34 (↓1)	Arizona	5-3	1.3488 (LW × 0.980)	lost by 7 vs USC
35 (↑3)	W-Michigan	6-2	1.3485 (LW × 1.005)	Bye
Other Notable Teams:
48 (↓9)	BYU BCS=20	7-1	1.3151 (LW × 0.983)	won by 7 vs UNLV
52 (↓24)	LSU BCS=19	5-2	1.2971 (LW × 0.927)	lost by 14 vs Georgia

Comments:

• Texas drops a bit, but they're so high, it's nearly inevitable unless they continue to blow out high-ranked teams. They are still way, way above #2 Penn State.

• It looks weird that Connecticut has suddenly leaped into the #15 spot. But they benefited from a lot of losses of teams formerly ahead of them, and the teams that they leap-frogged either lost, or beat much weaker teams. Connecticut has Pittsburgh, South Florida and West Virginia to play, so if they belong at the top of the rankings, they'll have a chance to prove it. Update: hey, Connecticut is ranked in the BCS!

• File this under 'I told you so': APR has never ranked Auburn higher than #41 (in week 4). As recently as week 6, AP had them ranked in the top 25. Auburn has gone 1-4 in their last 5, this week losing to lowly West Virginia. Worse, with Mississippi, Georgia, and Alabama remaining to play, they likely won't get 2 more wins to make a .500 season.

• And another for the 'I told you so' file: except for the first posting of the NCAA power rankings, APR has never had LSU higher than #28 (in week 8). This week, they take another big loss to a ranked team. Their situation isn't nearly as dire as Auburn, but it's clear LSU isn't a top 25 team.

• Every BCS-ranked team except 4 stragglers is now in APR's top 25!

NFL Week 8 Picks

2008 Navigation Links.

Picks for week 8, based on the week 7 rankings.
Read this for a description of APR.
Read this for a description of SRS.
Line values are taken from yahoo.com (except when noted otherwise), and are for comparison purposes only.

Game	APR	SRS	Line
Washington Redskins @ Detroit Lions	Redskins ρ=1.3793	Redskins δ=17.89	Redskins -7½
Oakland Raiders @ Baltimore Ravens	Ravens ρ=1.1978	Ravens δ=13.19	Ravens -6½
Indianapolis Colts @ Tennessee Titans	Titans ρ=1.1892	Titans δ=8.57	Titans -4
Kansas City Chiefs @ New York Jets	Jets ρ=1.1713	Jets δ=13.83	Jets -13
Buffalo Bills @ Miami Dolphins	Bills ρ=1.1445	Bills δ=2.61	Bills -1½
Tampa Bay Buccaneers @ Dallas Cowboys	Buccaneers ρ=1.1201	Buccaneers δ=6.93	Cowboys -2
Cincinnati Bengals @ Houston Texans	Texans ρ=1.1126	Texans δ=2.26	Texans -9½
New York Giants @ Pittsburgh Steelers	Steelers ρ=1.1063	Steelers δ=2.18	Steelers -2½
Atlanta Falcons @ Philadelphia Eagles	Eagles ρ=1.0743	Eagles δ=6.09	Eagles -9
St. Louis Rams @ New England Patriots	Patriots ρ=1.0556	Patriots δ=10.12	Patriots -7
Seattle Seahawks @ San Francisco 49ers	Seahawks ρ=1.0524	49ers δ=3.74	49ers -4½
San Diego Chargers @ New Orleans Saints	Chargers ρ=1.0514	Chargers δ=0.61	Chargers -3
Cleveland Browns @ Jacksonville Jaguars	Jaguars ρ=1.0488	Jaguars δ=0.91	Jaguars -7
Arizona Cardinals @ Carolina Panthers	Panthers ρ=1.0266	Panthers δ=2.46	Panthers -4½

Bye: Chicago Bears, Denver Broncos, Green Bay Packers, Minnesota Vikings

Comments:

• Disagreement on Bucs @ Cowboys and Seahawks @ 49ers.

• With the Giants facing their toughest opponent of the season, this has to be one of the Cowboys' best chances to gain some ground in the division. But they've gone 1-4 since their last convincing win Week 3 at the Packers. And their schedule doesn't get any easier after the Bucs, either.

• At 1-5, the Seahawks' season is pretty much done, even if Hasselbeck comes back immediately. As it is, this week at the 49ers might be one of their few remaining chances to win another game.

• The Cardinals are as close to a lock for the division as a team can be with 6 games played. The game @ Carolina is one they need to win if they want to be more than one-and-done in the playoffs.

• The Colts are already 3 games behind the division-leading Titans. If they lose, not only are they a long way out of first place in the AFC South, a 3-4 record will put them behind in the race for a wildcard spot, too.

NFL Week 7 Power Rankings

2008 Navigation Links
Read this for a description of APR.

Rank	Team	W-L	Power Index	This week
1 (-)	Tennessee Titans	6-0	1.5240 (LW * 0.997)	won by 24 at Chiefs
2 (-)	Tampa Bay Buccaneers	5-2	1.4674 (LW * 0.983)	won by 10 vs Seahawks
3 (-)	Pittsburgh Steelers	5-1	1.4477 (LW * 1.022)	won by 28 at Bengals
4 (↑1)	Chicago Bears	4-3	1.4106 (LW * 1.017)	won by 7 vs Vikings
5 (↑3)	Carolina Panthers	5-2	1.4047 (LW * 1.028)	won by 23 vs Saints
6 (↑10)	Buffalo Bills	5-1	1.4036 (LW * 1.087)	won by 9 vs Chargers
7 (↑2)	Philadelphia Eagles	3-3	1.3997 (LW * 1.025)	Bye
8 (↓2)	Arizona Cardinals	4-2	1.3683 (LW * 0.988)	Bye
9 (↑11)	Baltimore Ravens	3-3	1.3647 (LW * 1.075)	won by 14 at Dolphins
10 (-)	Washington Redskins	5-2	1.3568 (LW * 1.003)	won by 3 vs Browns
11 (↑1)	Jacksonville Jaguars	3-3	1.3277 (LW * 0.985)	Bye
12 (↑11)	Green Bay Packers	4-3	1.3231 (LW * 1.076)	won by 20 vs Colts
13 (↓6)	San Diego Chargers	3-4	1.3202 (LW * 0.959)	lost by 9 at Bills
14 (↓10)	Dallas Cowboys	4-3	1.3100 (LW * 0.936)	lost by 20 at Rams
15 (↑6)	New York Giants	5-1	1.3085 (LW * 1.040)	won by 12 vs 49ers
16 (↑3)	Atlanta Falcons	4-2	1.3029 (LW * 1.019)	Bye
17 (↓6)	Indianapolis Colts	3-3	1.2815 (LW * 0.949)	lost by 20 at Packers
18 (↑4)	Cleveland Browns	2-4	1.2658 (LW * 1.015)	lost by 3 at Redskins
19 (↓2)	Minnesota Vikings	3-4	1.2658 (LW * 0.989)	lost by 7 at Bears
20 (↑4)	New England Patriots	4-2	1.2587 (LW * 1.036)	won by 34 vs Broncos
21 (↓6)	New Orleans Saints	3-4	1.2556 (LW * 0.956)	lost by 23 at Panthers
22 (↓8)	New York Jets	3-3	1.2410 (LW * 0.938)	lost by 3 at Raiders
23 (↓10)	Miami Dolphins	2-4	1.2264 (LW * 0.916)	lost by 14 vs Ravens
24 (↑6)	St. Louis Rams	2-4	1.1924 (LW * 1.166)	won by 20 vs Cowboys
25 (↓7)	Denver Broncos	4-3	1.1911 (LW * 0.931)	lost by 34 at Patriots
26 (↓1)	Houston Texans	2-4	1.1668 (LW * 0.967)	won by 7 vs Lions
27 (↓1)	Oakland Raiders	2-4	1.1393 (LW * 1.030)	won by 3 vs Jets
28 (↑3)	Seattle Seahawks	1-5	1.1123 (LW * 1.120)	lost by 10 at Buccaneers
29 (↓1)	Kansas City Chiefs	1-5	1.0595 (LW * 0.977)	lost by 24 vs Titans
30 (↓1)	San Francisco 49ers	2-5	1.0569 (LW * 1.028)	lost by 12 at Giants
31 (↓4)	Cincinnati Bengals	0-7	1.0488 (LW * 0.966)	lost by 28 vs Steelers
32 (-)	Detroit Lions	0-6	0.9837 (LW * 1.018)	lost by 7 at Texans

Comments:

• For the first time this season, the top 3 teams stay the same one week to the next.

• I'm not really buying the Bears and Panthers as top-5 teams, or the Bills, Eagles, and Ravens as top-10. Though, looking at lower-ranked teams, I don't see any teams that are solidly better.

• Last week, I complained that ESPN ranked the Bills in their top 5, and now APR has them #6 after a 9 point win over the Chargers. Of course, their win over the Rams is starting to look a little more impressive...

• Speaking of the Rams—could they get back in the playoff hunt? They would need to finish 6-4 to make it to 8 wins. Even if they win their 3 remaining games against the 49ers and Seahawks, they'll need to win at least 3 from games against the Patriots, Cardinals (× 2), Bears, Dolphins, and Falcons. That will be an extraorinary turn-around, if they can do it.

Teams ESPN likes more than APR:

Giants (+13)

The Giants didn't lose this week, but that game against the 49ers sure wasn't pretty, either. This should start to be resolved next week, when the Giants go to Pittsburgh to play the Steelers.

Patriots (+11)

APR might be underating the patriots at #20, but but they haven't beat anyone higher-ranked than the Broncos (#25), and they've got a couple of ugly losses.

Teams APR likes more than ESPN:

Ravens (-11)

The Ravens' wins (Browns, Bengals, and Dolphins) sure don't seem that impressive. But they've got a couple of close losses against the high-ranked Steelers and Titans. They've got some easy teams coming up, which could help build some confidence in their rookie QB.

Bears (-9)

APR has the Bears overrated. Since when does the Bears' defense give up 41 points?

Historically Bad Teams Watch

With the Rams winning, and the Lions only losing by 7, none of this seasons' teams are in the "top 10" for lowest net points per game. The Lions and Chiefs are tied at #18 (along with the '71 Bills), and the Rams are now at #35.

Until some team makes it back into the top 10, I'm putting this feature on hiatus. Of course, that could be next week if the Chiefs lose by 23 or more to the Jets...

NFL Week 7 Pick Results

2008 Navigation Links
Read this for a description of APR.
Read this for a description of SRS.
Line picks counted correct when the favored team wins; they do not have to beat the spread to count as a win.

Pick results for week 7.

Unanimous picks (7-4)

Correctly picked:

Titans @ Chiefs
Steelers @ Bengals
Lions @ Texans
49ers @ Giants
Vikings @ Bears
Browns @ Redskins
Seahawks @ Buccaneers

Not so much:

• Cowboys @ Rams
  Brad Johnson will no doubt get his share of the blame for this, but Marion Barber lost a critical early fumble, and the defense only forced the Rams to punt once in the first half. This is just not the same Cowboys that got the #1 seed last year.

• Jets @ Raiders
  The Jets' offense did not have a good day. 8 drives went 30 yards or less. The Raiders were nearly as bad; they just (barely) managed to hang on to their win.

• Colts @ Packers
  The Colts still have major issues. Peyton Manning was just 50% on completions, with a couple of back-breaking interceptions. The Colts defense didn't force the Packers to punt until the 3rd quarter (and then special teams gave up the 1st down on a delay of game penalty). The Colts gave up 110 yards on 12 penalties.

  On the other side, Aaron Rodgers did a good job finding open recievers: 8 Packers caught the ball at least once, though none had more than 37 yards receiving.

• Chargers @ Bills
  

Split Picks (APR 1-2) (SRS 1-2) (Line 2-1)

• Ravens (SRS) @ Dolphins (APR, Line)
  Apparently the Ravens were not fooled by direct snaps to running backs. And the Dolphins' win over the Chargers is starting to seem like a long time ago.

• Broncos (APR, SRS) @ Patriots (Line)
  The only road game the Broncos have won this season was week 1 at the Raiders. Two interceptions and a non-existant run defense doesn't change that in Foxboro. Somehow, they're still at the top of the AFC West, but that can't last long.

• Saints (SRS) @ Panthers (APR, Line)
  The NFC South seems to have the football version of rock/scissors/paper going: Saints beat Bucs, Bucs beat Panthers, Panthers beat Bucs. Of course, for the Saints, that week 1 win over the Bucs is getting to be a long time ago.

Standings:

	This week	Overall
APR:	8-6	53-33
SRS:	8-6	49-37
Line:	9-5	56-29-1
Unanimous:	7-4	40-17

New power rankings, and next weeks picks to come.

NCAA: APR's Top 30+, Week 8

2008 Navigation Links.
Read this for a description of APR.
Read this for some disclaimers about applying APR to the NCAA.

I'll update with poll information (and probably add a few more teams) once it's posted. Update: done.

Rank	Team	W-L	Power Index	This week
1 (-)	Texas BCS=1	7-0	1.7940 (LW * 1.048)	won by 25 vs Missouri
2 (↑4)	Florida BCS=10	5-1	1.6658 (LW * 1.045)	Bye
3 (↓1)	Penn-St BCS=3	8-0	1.6396 (LW * 0.969)	won by 29 vs Michigan
4 (↑5)	Oklahoma BCS=4	6-1	1.6253 (LW * 1.042)	won by 14 vs Kansas
5 (↓1)	USC BCS=5	5-1	1.6013 (LW * 0.978)	won by 69 at Washington-St
6 (↑11)	Ohio-St BCS=9	7-1	1.5821 (LW * 1.084)	won by 38 at Michigan-St
7 (↓2)	Alabama BCS=2	7-0	1.5809 (LW * 0.977)	won by 4 vs Mississippi
8 (-)	Oklahoma-St BCS=6	7-0	1.5711 (LW * 0.998)	won by 28 vs Baylor
9 (↓2)	Texas-Tech BCS=8	7-0	1.5676 (LW * 0.990)	won by 18 at Texas-A&M
10 (↑1)	Georgia BCS=7	6-1	1.5622 (LW * 1.017)	won by 10 vs Vanderbilt
11 (↓8)	Missouri BCS=15	5-2	1.5583 (LW * 0.948)	lost by 25 at Texas
12 (↑15)	Pittsburgh BCS=17	5-1	1.4953 (LW * 1.050)	won by 21 at Navy
13 (↑8)	TCU BCS=14	7-1	1.4800 (LW * 1.030)	won by 25 vs BYU
14 (↑2)	Boise-St BCS=12	6-0	1.4706 (LW * 1.006)	won by 20 vs Hawaii
15 (↓3)	North-Carolina	5-2	1.4700 (LW * 0.965)	lost by 3 at Virginia
16 (↓3)	Ball-St BCS=20	7-0	1.4649 (LW * 0.969)	Bye
17 (↑15)	Tulsa BCS=19	7-0	1.4507 (LW * 1.034)	won by 42 vs UTEP
18 (↓4)	Virginia-Tech	5-2	1.4444 (LW * 0.957)	lost by 5 at Boston-Coll
19 (↑1)	Minnesota BCS=24	6-1	1.4431 (LW * 1.000)	Bye
20 (↑2)	Utah BCS=11	8-0	1.4353 (LW * 1.001)	won by 33 vs Colorado-St
21 (↓11)	Michigan-St	6-2	1.4353 (LW * 0.921)	lost by 38 vs Ohio-St
22 (↑6)	Nebraska	4-3	1.4327 (LW * 1.009)	won by 28 at Iowa-St
23 (↑1)	South-Florida BCS=16	6-1	1.4248 (LW * 0.996)	won by 32 vs Syracuse
24 (↑5)	Iowa	5-3	1.4226 (LW * 1.003)	won by 22 vs Wisconsin
25 (↓10)	Georgia-Tech BCS=18	6-1	1.4178 (LW * 0.956)	won by 4 at Clemson
26 (-)	Vanderbilt	5-2	1.4124 (LW * 0.991)	lost by 10 at Georgia
27 (↑11)	Boston-Coll	5-1	1.4072 (LW * 1.023)	won by 5 vs Virginia-Tech
28 (↑19)	LSU BCS=13	5-1	1.3991 (LW * 1.089)	won by 7 at South-Carolina
29 (↑11)	Florida-St BCS=25	5-1	1.3878 (LW * 1.014)	won by 9 at NC-State
30 (↓12)	Kansas BCS=23	5-2	1.3802 (LW * 0.950)	lost by 14 at Oklahoma
Other Notable Teams:
31 (↑8)	Mississippi	3-4	1.3767 (LW * 1.003)	lost by 4 at Alabama
32 (↑4)	Oregon-St	4-3	1.3767 (LW * 0.997)	won by 21 at Washington
33 (↑12)	Arizona	5-2	1.3766 (LW * 1.050)	won by 15 vs California
34 (↑8)	Northwestern	6-1	1.3662 (LW * 1.004)	won by 22 vs Purdue
35 (↓16)	South-Carolina	5-3	1.3630 (LW * 0.940)	lost by 7 vs LSU
36 (↓3)	Notre-Dame	4-2	1.3622 (LW * 0.975)	Bye
37 (↑28)	Miami-FL	4-3	1.3532 (LW * 1.103)	won by 18 at Duke
38 (↓3)	W-Michigan	6-2	1.3414 (LW * 0.969)	lost by 10 at Cent-Michigan
39 (↓8)	BYU BCS=21	6-1	1.3376 (LW * 0.950)	lost by 25 at TCU

Comments:

• Texas' power index is head-and-shoulders above everyone else. That's the highest power index (beyond early-season anomolies) I've ever seen APR generate (last year the Patriots maxed out at 1.7340 after week 11).

• Note that a lot of teams are clustered pretty close together if you go by power index. Boise State is just a whisker above North Carolina, and Utah and Michigan State are in a virtual tie.

  That's one of the reasons why there's so much shuffling around week to week: it doesn't take much change in the power index to move a team up or down.

• Penn State gets no love for their 29-point beat-down of the very weak Michigan squad. It's not helping that Wisconsin's power index has fallen off a cliff the last two weeks, either.

• On the other hand, Florida moves up in part because the Gators' win over LSU last week looks better, thanks to LSU's win over South Carolina.

• Congradulations to the Wildcats for their very solid win over the Cal Bears. For you sleepy-heads on the east coast who didn't watch this game, you'll want to look for a highlight package. Arizona running back Keola Antolin had 149 yards on 21 carries for 3 touchdowns. I'm pretty sure I heard the announcers say he's a freshman. If he can keep up that kind of production, Arizona is going to be a real threat the rest of the season.

NFL Week 7 Game Picks

2008 Navigation Links.

Picks for week 7, based on the week 6 rankings.
Read this for a description of APR.
Read this for a description of SRS.
Line values are taken from yahoo.com (except when noted otherwise) and are for comparison purposes only.

Game	APR	SRS	Line
Seattle Seahawks @ Tampa Bay Buccaneers	Buccaneers ρ=1.5030	Buccaneers δ=27.07	Buccaneers -10½
Tennessee Titans @ Kansas City Chiefs	Titans ρ=1.4107	Titans δ=26.13	Titans -7½
Dallas Cowboys @ St. Louis Rams	Cowboys ρ=1.3690	Cowboys δ=29.30	Cowboys -6½
Pittsburgh Steelers @ Cincinnati Bengals	Steelers ρ=1.3041	Steelers δ=11.06	Steelers -9½
Detroit Lions @ Houston Texans	Texans ρ=1.2498	Texans δ=11.30	Texans -9
San Francisco 49ers @ New York Giants	Giants ρ=1.2244	Giants δ=11.24	Giants -10
New York Jets @ Oakland Raiders	Jets ρ=1.1959	Jets δ=14.56	Jets -3
Indianapolis Colts @ Green Bay Packers	Colts ρ=1.0988	Colts δ=4.33	Colts -1½
Minnesota Vikings @ Chicago Bears	Bears ρ=1.0843	Bears δ=7.88	Bears -3
Cleveland Browns @ Washington Redskins	Redskins ρ=1.0838	Redskins δ=2.55	Redskins -7
San Diego Chargers @ Buffalo Bills	Chargers ρ=1.0661	Chargers δ=8.83	Chargers -1
Baltimore Ravens @ Miami Dolphins	Dolphins ρ=1.0541	Ravens δ=0.03	Dolphins -3
Denver Broncos @ New England Patriots	Broncos ρ=1.0522	Broncos δ=8.33	Patriots -3
New Orleans Saints @ Carolina Panthers	Panthers ρ=1.0403	Saints δ=0.76	Panthers -3

Bye: Philadelphia Eagles, Atlanta Falcons, Arizona Cardinals, Jacksonville Jaguars

Comments:

• It bears repeating: APR and SRS do not take into account injuries. Which is why they think the Cowboys should have an easy win, while the line only gives them 6½ points.

• If 49ers @ Giants sounds familiar, it's because they played last year too (coincidentally also in week 7). Final score then: 49ers 15, Giants 33.

• Saints @ Panthers is being played in England, which means it is only technically a home game for the Panthers. It also means besides the usual practices and game planning, both teams have long plane rides ahead of them, plus jet lag to deal with (New Orleans is in the Central time zone, so they get an extra hour of lag to deal with). How on earth did the NFL get these teams to agree to do this with a critical divisional game?

  As bob p points out in the comments, it's next week's Chargers @ Saints that will be played in London. <Emily-Litella>Never mind</Emily-Litella>

• Only one of the sites yahoo is getting odds from wants to give a line on Chargers @ Bills. Not exactly a lot of confidence out there. But the Chargers have faced a tougher schedule, and done reasonably well. I think it will qualify as a genuine upset if the Bills get the win.

• Only the bottom three games are not unanimous. APR only differs from the Line on Broncos @ Patriots. Note that for Ravens @ Dolphins, SRS has δ=0.03, which is virtually a push. The δ=0.76 for Saints @ Panthers isn't much stronger.

• Colts are only getting 1½ points? It's only been one game, but last week's Ravens @ Colts sure makes it seem like the Colts have worked through their early-season issues. If they can play like that again, it will be tough for the Packers to fare much better.

NFL Week 6 Power Rankings

2008 Navigation Links
Read this for a description of APR

Power Rankings For Week 6

Rank	Team	W-L	Power Index	This week
1 (-)	Tennessee Titans	5-0	1.5291 (LW * 1.031)	Bye
2 (↑6)	Tampa Bay Buccaneers	4-2	1.4931 (LW * 1.072)	won by 24 vs Panthers
3 (↑13)	Pittsburgh Steelers	4-1	1.4160 (LW * 1.084)	Bye
4 (↑5)	Dallas Cowboys	4-2	1.3996 (LW * 1.012)	lost by 6 at Cardinals
5 (↑6)	Chicago Bears	3-3	1.3872 (LW * 1.030)	lost by 2 at Falcons
6 (↓1)	Arizona Cardinals	4-2	1.3853 (LW * 0.982)	won by 6 vs Cowboys
7 (↓1)	San Diego Chargers	3-3	1.3769 (LW * 0.979)	won by 20 vs Patriots
8 (↓4)	Carolina Panthers	4-2	1.3664 (LW * 0.952)	lost by 24 at Buccaneers
9 (↑8)	Philadelphia Eagles	3-3	1.3661 (LW * 1.048)	won by 14 at 49ers
10 (↓8)	Washington Redskins	4-2	1.3521 (LW * 0.914)	lost by 2 vs Rams
11 (↑10)	Indianapolis Colts	3-2	1.3505 (LW * 1.149)	won by 28 vs Ravens
12 (↑12)	Jacksonville Jaguars	3-3	1.3476 (LW * 1.161)	won by 7 at Broncos
13 (↓10)	Miami Dolphins	2-3	1.3387 (LW * 0.912)	lost by 1 at Texans
14 (↓7)	New York Jets	3-2	1.3224 (LW * 0.945)	won by 12 vs Bengals
15 (-)	New Orleans Saints	3-3	1.3134 (LW * 0.996)	won by 31 vs Raiders
16 (↑3)	Buffalo Bills	4-1	1.2915 (LW * 1.062)	Bye
17 (↓3)	Minnesota Vikings	3-3	1.2794 (LW * 0.965)	won by 2 vs Lions
18 (↓8)	Denver Broncos	4-2	1.2788 (LW * 0.946)	lost by 7 vs Jaguars
19 (↑1)	Atlanta Falcons	4-2	1.2786 (LW * 1.064)	won by 2 vs Bears
20 (↓2)	Baltimore Ravens	2-3	1.2700 (LW * 0.985)	lost by 28 at Colts
21 (↓8)	New York Giants	4-1	1.2588 (LW * 0.941)	lost by 21 at Browns
22 (↑4)	Cleveland Browns	2-3	1.2475 (LW * 1.119)	won by 21 vs Giants
23 (-)	Green Bay Packers	3-3	1.2291 (LW * 1.059)	won by 10 at Seahawks
24 (↓12)	New England Patriots	3-2	1.2153 (LW * 0.903)	lost by 20 at Chargers
25 (↑4)	Houston Texans	1-4	1.2073 (LW * 1.162)	won by 1 vs Dolphins
26 (↓4)	Oakland Raiders	1-4	1.1057 (LW * 0.948)	lost by 31 at Saints
27 (-)	Cincinnati Bengals	0-6	1.0858 (LW * 1.017)	lost by 12 at Jets
28 (↓3)	Kansas City Chiefs	1-4	1.0840 (LW * 0.961)	Bye
29 (↓1)	San Francisco 49ers	2-4	1.0281 (LW * 0.971)	lost by 14 vs Eagles
30 (↑2)	St. Louis Rams	1-4	1.0223 (LW * 1.231)	won by 2 at Redskins
31 (↓1)	Seattle Seahawks	1-4	0.9934 (LW * 1.026)	lost by 10 vs Packers
32 (↓1)	Detroit Lions	0-5	0.9660 (LW * 1.129)	lost by 2 at Vikings

Comments

• This week is just completely nuts. 3 of last week's top 5 teams lost. The Cowboys, Broncos, Bears, Patriots and Giants (formerly ranked 8, 9, 10, 11, 12, and 13) all lost. The Titans are now the only remaining undefeated team; the Bengals and Lions are the only remaining 0-fers.

• The Browns' big win over the Giants is causing all kinds of churn. The Steelers were at #8 pending MNF (thanks to wins by the Jaguars, Eagles, and Texans). The Browns' win was enough to push them up to #3.

  A good example of previous games playing a big factor in a team's power index, which seems to be the theme this week.

• The Cowboys also move up 3 spots thanks to the Browns' win (plus 2 spots from wins by the Eagles and Packers).

• The Bears move up on their loss thanks to wins by the Eagles, Buccaneers, and Colts (and the Lions moving up because of a close loss).

• The Cardinals move down, thanks to losses by the 49ers, Dolphins, and Redskins.

• The Chargers lose ground because of losses by the Dolphins, Raiders, Broncos, and Panthers.

• The Rams' power index actually moved up quite a bit (23%). But they were in such a deep hole last week, their win was only enough to move them up 2 spots.

• In spite of the big upsets suffered by the Redskins and Giants, it is the Patriots that suffer the most in terms of rank (moving down 12 spots) and power index (losing nearly 10%).

Teams ESPN likes more than APR:

Giants (+19)

OK, ESPN's power rankings are mostly retrodictive (i.e. describe the teams accomplishments, rather than predict what they'll do in the future). But to only drop them one spot on a crushing 21-point loss to a (formerly) 1-3 team? Wow.

Bills (+12)

Three of the Bill's four wins are against the Raiders (#26), Seahawks (#31) and Rams (#30). Their win against the Jaguars (#12) looks better this week, but how is that good enough for a top 5 ranking?

Patriots (+7)

Another team with a big loss, and ESPN drops them 8 spots, but no match for the 12 spots they drop in the APR rankings.

Teams APR likes more than ESPN:

Bears (-15)

The Bears are a team that seems to run hot and cold. APR sees mostly hot; ESPN sees mostly cold.

Dolphins (-10)

Maybe a team that APR has ranked too high. But surely they deserve better than the #24 spot.

Historically Bad Teams Watch

And just like that, the '76 Bucs have "first" place all to themselves. We see teams that play as bad or worse than them for stretches. But to play that bad for a whole season—that takes a real dedication to sucking that is very hard to maintain.

Rank	Team	Net Pts/Game	W-L	Next Week
1	1976 Buccaneers	-20.50	0-14	n/a
2	2008 Rams	-20.40	1-4	vs Cowboys
3	1972 Patriots	-18.4	3-11	n/a
4	1973 Oilers	-17.71	1-13	n/a
5	1967 Falcons	-17.64	1-12	n/a
6	1981 Colts	-17.12	2-14	n/a
7	1966 Giants	-17.00	1-12	n/a
8	1966 Falcons	-16.64	3-11	n/a
9	2008 Lions	-16.60	0-5	@ Texans
10	1990 Patriots	-16.56	1-15	n/a
...	...	...	...	...
39	2008 Chiefs	-13.20	1-4	vs Titans

NFL Week 6 Pick Results

2008 Navigation Links
Read this for a description of APR.
Read this for a description of SRS.
Line picks are counted correct when the favored team wins; they do not have to beat the spread to count as a win.

Not a good week for the favored teams...

Unanimous picks (5-4)

Correctly picked:

Lions @ Vikings
Bengals @ Jets
Raiders @ Saints
Eagles @ 49ers
Patriots @ Chargers

Not so much:

• Rams @ Redskins
  Biggest upset of the season? Even if the Redskins have been playing a little over their heads lately, the Rams have been awful. But the Rams did show a little life two weeks ago against the Bills... Did Scott Linehan get things turned around just a little too late to save his job?

• Bears @ Falcons
  Matt Ryan is starting to look like the genius pick of the 2008 draft. And the Bears find out that a 1-point lead with 17 seconds to play is not as safe as they would like.

• Jaguars @ Broncos
  The Broncos lost 2 fumbles, Jay Cutler threw an interception, and the Colts aren't the only AFC South team shaking off a little rust this week.

• Giants @ Browns
  Two words: bad Eli. OK, that's not entirely fair. There's some blame that belongs to the defense, too. On the Brown's last offensive score, they had 3 false starts, delay of game, and a holding penalty. And still drove 87 yards (not counting the 30 penalty yards) for a touchdown.

Split Picks (APR 2-3) (SRS 1-4) (Line 3-2)

• Dolphins (APR, SRS) @ Texans (Line)
  The Dolphins' defense has looked improved this season, and if it could've held the Texans just one more time, it would've been enough. On the other hand, the Texans could easily be 3-2 at this point, so maybe they are very underated at 1-4.

• Packers (APR, SRS) @ Seahawks (Line)
  The Seahawks made this more interesting than they had any right to, with the Packers only getting a two-score lead well into the fourth quarter. If the Seahawks had a QB that could get the ball to his receivers, they would be winning more games.

• Ravens (APR, SRS) @ Colts (Line)
  Before this week, the Colts looking really bad, needing come-backs to win against the Vikings and Texans. But Peyton Manning looks like he's finally shaking off the rust of missing the preseason, and Joe Flacco looks like a rookie QB.

• Panthers (APR, SRS) @ Buccaneers (Line)
  The Panthers' special teams has a punt blocked, Jake Delhomme throws three interceptions, and the Panthers' offense just can't do anything on a day where the offense mounts just two drives longer than 17 yards before garbage time.

• Cowboys (SRS, Line) @ Cardinals (APR)
  The Cardinals seemed to be the best team in this game, with the Cowboys hanging around thanks to a number of breaks going their way (including an offsides penalty on an injured player, then a do-over on a blocked punt nullified by a time out). The Cowboys' breaks ran out before the game clock, to their detriment.

I've noticed on these split picks, it's often APR and SRS taking the road team, and the Line taking the home team. When I originally wrote APR, it didn't look like it was worth giving any weight to home field advantage. Maybe that deserves renewed consideration.

Standings

	This week	Overall
APR:	7-7	45-27
SRS:	6-8	41-31
Line:	8-6	47-24-1
Unanimous:	5-4	33-13

New power rankings, and next weeks' picks to come.

2008 NFL & NCAA Links

2009

This is a post of links to the NFL- and NCAA-related posts made during the 2008 seasons. Entries will be linkified as new posts are added.

Preseason

NFL Preseason Power, Week 3
NFL Week 1 Picks (based on preseason data)
NFL Week 1 Pick Results. Not impressive.

NFL, 2008 Season

Week 1			Power Rankings
Week 2	Picks	Pick Results	Power Rankings
Week 3	Picks	Pick Results	Power Rankings
Week 4	Picks	Pick Results	Power Rankings
Week 5	Picks	Pick Results	Power Rankings
Week 6	Picks	Pick Results	Power Rankings
Week 7	Picks	Pick Results	Power Rankings
Week 8	Picks	Pick Results	Power Rankings
Week 9	Picks	Pick Results	Power Rankings
Week 10	Picks	Pick Results	Power Rankings
Week 11	Picks	Pick Results	Power Rankings
Week 12	Picks	Pick Results	Power Rankings
Week 13	Picks	Pick Results	Power Rankings
Week 14	Picks	Pick Results	Power Rankings
Week 15	Picks	Pick Results	Power Rankings
Week 16	Picks	Pick Results	Power Rankings
Week 17	Picks	Pick Results	Power Rankings
Wildcard	Picks	Pick Results	Power Rankings
Divisional	Picks	Pick Results	Power Rankings
Championship	Picks	Pick Results	Power Rankings
Superbowl	Picks	Pick Results	Power Rankings

NCAA

Week 4 Top 30+
Week 5 Top 30+
Week 6 Top 30+
Week 7 Top 30+
Week 8 Top 30+
Week 9 Top 30+
Week 10 Top 30+
Week 11 Top 30+
Week 12 Top 30+
Week 13 Top 30+
Week 14 Top 30+
Week 15 Top 30+
Bowl game picks
Final Top 30+

Wrapup

• APR vs ESPN
• APR's Upsets
• APR Picks by Team
• APR Strength of Schedule
• APR Pick Reliability

NCAA: APR's top 30+, Week 7

Rankings for week 7, 12 October 2008.

Week 6 rankings here.
Read this for a description of APR.
Read this for some disclaimers about applying APR to the NCAA.

Rank	Team	W-L	Power Index	This week
1 (LW-4)	Texas AP=1	6-0	1.7125 (LW * 1.060)	won by 10 at Oklahoma
2 (LW)	Penn-St AP=3	7-0	1.6920 (LW * 1.011)	won by 41 at Wisconsin
3 (LW+2)	Missouri AP=11	5-1	1.6444 (LW * 0.924)	lost by 5 vs Oklahoma-St
4 (LW)	USC AP=6	4-1	1.6371 (LW * 0.993)	won by 28 vs Arizona-St
5 (LW-1)	Alabama AP=2	6-0	1.6177 (LW * 1.004)	Bye
6 (LW-3)	Florida AP=5	5-1	1.5943 (LW * 1.032)	won by 30 vs LSU
7 (LW)	Texas-Tech AP=7	6-0	1.5841 (LW * 1.008)	won by 6 vs Nebraska
8 (LW-8)	Oklahoma-St AP=8	6-0	1.5748 (LW * 1.054)	won by 5 at Missouri
9 (LW+6)	Oklahoma AP=4	5-1	1.5598 (LW * 0.944)	lost by 10 vs Texas
10 (LW-7)	Michigan-St AP=20	6-1	1.5580 (LW * 1.057)	won by 17 at Northwestern
11 (LW-2)	Georgia AP=10	5-1	1.5358 (LW * 1.010)	won by 12 vs Tennessee
12 (LW)	North-Carolina AP=18	5-1	1.5235 (LW * 0.997)	won by 5 vs Notre-Dame
13 (LW+3)	Ball-St AP=24	7-0	1.5123 (LW * 0.985)	won by 17 at West-Kentucky
14 (LW+3)	Virginia-Tech AP=17	5-1	1.5086 (LW * 0.984)	Bye
15 (LW+7)	Georgia-Tech AP=rv(25)	5-1	1.4824 (LW * 0.943)	won by 3 vs Gardner-Webb
16 (LW-15)	Boise-St AP=15	5-0	1.4617 (LW * 1.046)	won by 17 at Southern-Miss
17 (LW-2)	Ohio-St AP=12	6-1	1.4590 (LW * 0.993)	won by 13 vs Purdue
18 (LW-6)	Kansas AP=16	5-1	1.4529 (LW * 1.020)	won by 16 vs Colorado
19 (LW-16)	South-Carolina AP=rv(9)	5-2	1.4493 (LW * 1.047)	won by 7 at Kentucky
20 (LW-12)	Minnesota AP=rv(48)	6-1	1.4429 (LW * 1.035)	won by 7 at Illinois
21 (LW-1)	TCU AP=rv(61)	6-1	1.4368 (LW * 0.993)	won by 6 at Colorado-St
22 (LW+1)	Utah AP=14	7-0	1.4344 (LW * 0.991)	won by 33 at Wyoming
23 (LW-10)	California AP=25	4-1	1.4313 (LW * 1.027)	Bye
24 (LW-2)	South-Florida AP=19	5-1	1.4303 (LW * 1.014)	Bye
25 (LW-4)	Navy	4-2	1.4248 (LW * 1.014)	Bye
26 (LW+12)	Vanderbilt AP=22	5-1	1.4248 (LW * 0.941)	lost by 3 at Mississippi-St
27 (LW+2)	Pittsburgh AP=23	4-1	1.4241 (LW * 1.006)	Bye
28 (LW-2)	Nebraska	3-3	1.4205 (LW * 1.016)	lost by 6 at Texas-Tech
29 (LW-12)	Iowa	4-3	1.4185 (LW * 1.048)	won by 36 at Indiana
30 (LW+7)	Wake-Forest AP=21	4-1	1.4105 (LW * 0.976)	won by 5 vs Clemson
31 (LW-11)	BYU AP=9	6-0	1.4087 (LW * 1.043)	won by 18 vs New-Mexico
32 (LW+17)	Tulsa AP=rv(84)	6-0	1.4031 (LW * 0.927)	won by 6 at SMU
33 (LW-3)	Notre-Dame AP=rv(1)	4-2	1.3972 (LW * 1.019)	lost by 5 at North-Carolina
Other notable teams
37 (LW+19)	Kentucky AP=rv(7)	4-2	1.3797 (LW * 0.938)	lost by 7 vs South-Carolina
40 (LW+12)	Florida-St AP=rv(38)	4-1	1.3683 (LW * 0.974)	Bye
42 (LW+22)	Northwestern AP=rv(3)	5-1	1.3606 (LW * 0.935)	lost by 17 vs Michigan-St
46 (LW-4)	Oregon AP=rv(4)	5-2	1.2946 (LW * 1.011)	won by 7 vs UCLA
47 (LW+10)	LSU AP=13	4-1	1.2853 (LW * 0.937)	lost by 30 at Florida
50 (LW+7)	Cincinnati AP=rv(1)	5-1	1.2811 (LW * 0.952)	won by 3 vs Rutgers

Comments:

• Penn State did increase in power this week; just not enough to stay ahead of Texas.

• Missouri had a big power lead last week; their close loss to a good team wasn't enough for APR to drop them very far.

• BYU and LSU are the only ranked teams out of APR's top 30. LSU had their chance to make the case that they were underrated, and came away with a 30-point beat-down for their trouble.

More info on APR

I want to expand and emphasize a statement I made in my earlier post on how APR works. I wrote:

a simple sort of teams from largest to smallest power index is used to generate a power rankings

This is important. Each team's position in the power ranking is based only on the relative values of their respective power index values generated by APR. In particular:

• No consideration is given to the ranking a team had in the previous week. It is possible (and in fact, not unusual) for teams to fall even though they won, or rise even though they lost.

• No consideration is given to win-loss records, and it's very possible for, say, a 4-2 team to be ranked below a 1-5 team.

An example

I know it's possible for a 4-2 team to be ranked below a 1-5 team, because it happened during the 2007 season.

After the week 6 games of the 2007 season, the Baltimore Ravens were 4-2. The ESPN power rankings had them in the #7 spot. But APR ranked them #25. Below the 1-5 Falcons, the 1-4 Bengals, and the 1-4 Saints. In all, APR had 12 teams with .500 or worse records ranked above the Ravens that week.

Why should a team with a winning record be ranked so low? As usual with APR, the Ravens had played a some weak teams, including losses to the Browns and Bengals, and a 2-point victory over a very weak 49ers squad.

APR is meant to be a predictive measure of power (e.g. how well teams perform the following week). In this case, the low ranking was clearly justified, as the Ravens lost their next 9 games (including giving the Dolphins their one win of the season). The only game the Ravens won after week 6 was in week 17, against a Steelers team playing Charlie Batch at QB.

Conclusion:

This is exactly the kind of thing APR is meant to show: the power of each team as it continues to play games in the season. If you read this description of the SRS ranking algorithm, the discussion makes a distinction between predictive systems (which team is more likely to win their next game) and retrodictive systems (which team accomplished more in the past). APR (like SRS) is a predictive system, and should be approached as such.

NFL Week 6 Game Picks

Picks for week 6, based on week 5 rankings.

Read this for a description of SRS.
Read this for a description of APR.
Line values are taken from yahoo.com, except when noted otherwise. They are in all cases subject to change, and for comparison purposes only.

Update: I see yahoo has changed their format; they are now showing a matrix of information from 5 different odds makers. Oh, well...

Game	APR	SRS	Line
St. Louis Rams @ Washington Redskins	Redskins ρ=1.7799	Redskins δ=37.2	Redskins -13.5
Detroit Lions @ Minnesota Vikings	Vikings ρ=1.5492	Vikings δ=23.1	Vikings -13.5
Miami Dolphins @ Houston Texans	Dolphins ρ=1.4124	Dolphins δ=14.0	Texans -3
Cincinnati Bengals @ New York Jets	Jets ρ=1.3098	Jets δ= 5.1	Jets -6
Philadelphia Eagles @ San Francisco 49ers	Eagles ρ=1.2305	Eagles δ=15.7	Eagles -5
New York Giants @ Cleveland Browns	Giants ρ=1.1997	Giants δ=13.7	Giants -7
Green Bay Packers @ Seattle Seahawks	Packers ρ=1.1988	Packers δ=18.5	Seahawks -3
Jacksonville Jaguars @ Denver Broncos	Broncos ρ=1.1645	Broncos δ= 6.1	Broncos -3.5
Oakland Raiders @ New Orleans Saints	Saints ρ=1.1302	Saints δ=12.6	Saints -7.5
Chicago Bears @ Atlanta Falcons	Bears ρ=1.1201	Bears δ= 9.5	Bears -2.5
Baltimore Ravens @ Indianapolis Colts	Ravens ρ=1.0964	Ravens δ=10.1	Colts -5.5
New England Patriots @ San Diego Chargers	Chargers ρ=1.0452	Chargers δ= 9.7	Chargers -6
Carolina Panthers @ Tampa Bay Buccaneers	Panthers ρ=1.0302	Panthers δ= 2.4	Buccaneers -2
Dallas Cowboys @ Arizona Cardinals	Cardinals ρ=1.0198	Cowboys δ= 4.3	Cowboys -5.5

Bye: Buffalo Bills, Pittsburgh Steelers, Tennessee Titans, Kansas City Chiefs

Comments:

• The Rams and Lions are both drawing lines of +13.5. The Lions have lost by fewer than 14 points just once (week 1 at the Falcons), and the Rams have never lost by fewer than 17 points (week 4 vs the Bills).

• The line likes the Texans over the Dolphins? On the one hand, I don't think the Texans are going to give up two late fumbles again. On the other hand, if the Dolphins are anywhere near as good, and the Texans anywhere near as bad as APR thinks they are, Dolphins should be able to take this one.

• The line likes the Seahawks over the Packers? Neither team is anywhere as good as last year. But the Seahawks have been just awful. If the Packers can't get a win against them, it's going to be a very long season for Cheese Heads.

• Cowboys @ Cardinals is the only disagreement between APR and SRS.

• I'm not so sure APR is wrong about Cowboys @ Cardinals, either. Coming off a loss to the Redskins, the Cowboys needed a late rally to hold off the 0-4 (now 0-5) Bengals. The Cowboys defense seems suspect, and I think it favors the Cardinals if this turns into an offensive shoot-out like the Cowboys' week 2 game against the Eagles.

Algorithms: How APR works, and what it's trying to do

For those of you coming here from footballoutsiders.com, let me say right away that APR is not supposed to be a replacement or alternative to DVOA, DPAR, or any of their other metrics.

The APR algorithm:

1. Initially, every team is assigned a power index of 1.0

2. For each team, a power index detailed below) is assigned to it for each game played.

3. For each team, the game power values for that team are then combined together using a weighted average (old games weighted less than new games) to give a new power index for the team.

4. Steps 2 and 3 are repeated a number of times to create a feedback loop.

The power for a played game is computed according to the following formula:

gamePower = resultPower × marginPower × opponentPower

Where:

• resultPower is a constant based on whether the team won or lost, and whether they were the visiting or home team, with the following constraints:

  road win > home win > tie > road loss > home loss

• marginPower is a constant based on the margin of victory (or loss). Right now, margin power is assigned in the following ranges:

  • won by 14 or more
  • won by 7 or more
  • won (or lost) by 6 or fewer points
  • lost by 7 or more
  • lost by 14 or more

• opponentPower is the power index of the opponent team.

Using the power index values

Once power index values have been generated, a simple sort of teams from largest to smallest power index is used to generate a power rankings. Similarly, picks for the following week's games are made by chosing the team with the largest power index value for games played to that point.

Evaluating design choices

The efficacy of the algorithm (as well as choices for the various constants) is judged strictly on how well it does picking games on the historical data set of NFL games played. I have made attempts to give more power for wider margins of victory (or loss); such attempts have lead to fewer games picked correctly and were discarded. This means teams can generate a only limited amount of power playing very weak teams, even with blow-out wins.

The design of the APR algorithm has also lead to a related phenomenon I call "power by association" (or PBA, if you like TLAs): When a weak team plays a strong team close (especially with the weak team on the road), the weak team will increase in power, even if they lose (and the strong team will decrease in power, even if they win).

The APR ranking system is by no means perfect. It doesn't take into account injuries, break-out players, fluke wins, how different teams match up, the strength of individual units within a team, or the strength of teams in different game situations. At the end of the season, it doesn't take into account that some teams will clinch their post-season fate early, and elect to rest key starters for one or more games.

Still, it does (to my way of thinking, anyway) a remarkably good job at picking games, and often reveals over- and under-rated teams before ESPN or other subjective-based power ranking systems notices.

Update: and, if you've made it this far, be sure to read this post, which expands on the consequences of APR's design.

NFL Week 5 Power Rankings

Disclaimers:

• The Secretary disavows all knowledge of these rankings.

• These rankings are algorithmically determined based on the final scores of games played this season. This means the power index is a measure of "Demonstrated Power"—and good teams are intrinsically limited on how much power they can demonstrate over very bad teams.

• The power index algorithm is intended to be predictive; as far as I am concerned, the final arbiter of what is a good or bad ranking algorithm is how well it does picking games.

• If you bet actual money on games, you are taking your own chances.

• Yes, I realize that it's not exactly conventional wisdom to put the Dolphins at #3. But that's where APR puts them.

• If it makes you feel any better, I am willing to stipulate in advance that the team of your choice is unfairly ranked. (Giants fans? I'm lookin' at you.)

Rank	Team	W-L	Power Index	This week
1 (LW-2)	Tennessee Titans	5-0	1.4831 (LW * 1.015)	won by 3 at Ravens
2 (LW+1)	Washington Redskins	4-1	1.4787 (LW * 1.001)	won by 6 at Eagles
3 (LW-12)	Miami Dolphins	2-2	1.4680 (LW * 1.119)	won by 7 vs Chargers
4 (LW-2)	Carolina Panthers	4-1	1.4351 (LW * 1.011)	won by 34 vs Chiefs
5 (LW-9)	Arizona Cardinals	3-2	1.4110 (LW * 1.075)	won by 24 vs Bills
6 (LW+2)	San Diego Chargers	2-3	1.4067 (LW * 0.966)	lost by 7 at Dolphins
7 (LW-6)	New York Jets	2-2	1.3988 (LW * 1.065)	Bye
8 (LW+6)	Tampa Bay Buccaneers	3-2	1.3931 (LW * 0.946)	lost by 3 at Broncos
9 (LW+4)	Dallas Cowboys	4-1	1.3836 (LW * 0.962)	won by 9 vs Bengals
10 (LW-1)	Denver Broncos	4-1	1.3516 (LW * 0.989)	won by 3 vs Buccaneers
11 (LW+3)	Chicago Bears	3-2	1.3465 (LW * 0.969)	won by 27 at Lions
12 (LW-6)	New England Patriots	3-1	1.3459 (LW * 1.053)	won by 9 at 49ers
13 (LW+3)	New York Giants	4-0	1.3377 (LW * 0.977)	won by 38 vs Seahawks
14 (LW-5)	Minnesota Vikings	2-3	1.3255 (LW * 1.064)	won by 3 at Saints
15 (LW+8)	New Orleans Saints	2-3	1.3186 (LW * 0.933)	lost by 3 vs Vikings
16 (LW-1)	Pittsburgh Steelers	4-1	1.3060 (LW * 1.011)	won by 5 at Jaguars
17 (LW+8)	Philadelphia Eagles	2-3	1.3029 (LW * 0.952)	lost by 6 vs Redskins
18 (LW+2)	Baltimore Ravens	2-2	1.2887 (LW * 0.986)	lost by 3 vs Titans
19 (LW+7)	Buffalo Bills	4-1	1.2161 (LW * 0.915)	lost by 24 at Cardinals
20 (LW-3)	Atlanta Falcons	3-2	1.2021 (LW * 1.027)	won by 3 at Packers
21 (LW-3)	Indianapolis Colts	2-2	1.1754 (LW * 1.021)	won by 4 at Texans
22 (LW+1)	Oakland Raiders	1-3	1.1667 (LW * 0.963)	Bye
23 (LW+3)	Green Bay Packers	2-3	1.1607 (LW * 0.946)	lost by 3 vs Falcons
24 (LW+2)	Jacksonville Jaguars	2-3	1.1607 (LW * 0.976)	lost by 5 vs Steelers
25 (LW)	Kansas City Chiefs	1-4	1.1277 (LW * 0.983)	lost by 34 at Panthers
26 (LW)	Cleveland Browns	1-3	1.1150 (LW * 0.994)	Bye
27 (LW-3)	Cincinnati Bengals	0-5	1.0679 (LW * 1.044)	lost by 9 at Cowboys
28 (LW+1)	San Francisco 49ers	2-3	1.0588 (LW * 0.971)	lost by 9 vs Patriots
29 (LW+1)	Houston Texans	0-4	1.0394 (LW * 0.971)	lost by 4 vs Colts
30 (LW+1)	Seattle Seahawks	1-3	0.9683 (LW * 0.939)	lost by 38 at Giants
31 (LW-1)	Detroit Lions	0-4	0.8557 (LW * 0.997)	lost by 27 vs Bears
32 (LW+1)	St. Louis Rams	0-4	0.8308 (LW * 0.963)	Bye

Comments:

• The Redskins were still #1 before the results of the Monday Night game. The Vikings win made the Titans win over the Vikings look better, and pushed them to the top spot.

• Yeah, the Dolphins have somehow found their way to the #3 spot. I realize this looks pretty wacky. Here's some internals of their power index calculation:

  Game:	vs SD	@ NE	@ ARI	vs NYJ
  Power:	1.630	1.907	1.050	1.169

  They are getting huge amounts of power from their wins, and it's a weighted average, so the early losses count for less every week.

• The Chargers are at #6 largely on the strength of their wins over the Jets and Raiders:

  Game:	@ MIA	@ OAK	vs NYJ	@ DEN	vs CAR
  Power:	1.236	1.510	1.781	1.275	1.209

  Note that the Jets' wins over the Dolphins and Cardinals both look better this week.

• Yeah, the Giants are not only #13, they fall 3 spots in spite of a blow-out win. Here's their internals:

  Game:	vs SEA	vs CIN	@STL	vs WAS
  Power:	1.300	1.154	1.217	1.749

  Even a blow-out win over a team as weak as the Seahawks isn't enough to sustain a very high power index. Also, as indicated above, older games are weighted less, so the Giants' win over the Redskins is discounted more than the Dolphins' and Chargers' more recent wins.

• More bad news for the Giants. They have the Browns (APR #26) and 49ers (APR #28) coming up, so they're probably going to continue to drift down. Then they get to the brutal part of their schedule: Steelers, Cowboys, Eagles, Ravens, Cardinals, Redskins... it may take a while, but if the Giants are the real deal, the will start climbing.

• Also check out the Bears dropping 3 spots for the trouble of blowing out the hapless Lions. APR's just not that impressed with blow out wins against the likes of the Seahawks, Lions and Rams.

• Teams ESPN likes more than APR:
  Giants (+12) covered above. Steelers (+11) haven't beat a team APR ranks higher than the Ravens (#18). Bills(+10) haven't beat a team APR ranks higher than the Raiders (#22). Colts (+10) have two skin-of-their-teeth wins against the Vikings (#14) and Texans (#29).

• Teams APR likes more than ESPN:
  Dolphins (-20) covered above. Jets (-11) beat the Dolphins and Cardinals, which actually looks pretty good, at least this week. Chargers (-10) covered above.

Historically Bad Teams Watch

The Rams had a Bye this week, so they don't move. The Lions had another blow-out loss, and could actually push the 76 Bucs down to 3rd place with another big loss. The Chiefs are included as an example of a more ordinary "bad team".

Rank	Team	Net Pts/Game	W-L	Next Week
1	2008 Rams	-26.00	0-4	@ Redskins
2	1976 Buccaneers	-20.50	0-14	n/a
3	2008 Lions	-20.25	0-4	@ Vikings
. . .	. . .	. . .	. . .	. . .
39	2008 Chiefs	-13.20	1-4	Bye