Blind Power Rankings Week 10: Changes on Top, Strength of Schedule and Average MOV

It's interesting that we keep finding ways to change the rankings, which goes to show how volatile the NFL really is. Some odd games this week ended up changing the rankings more than you'd normally expect heading into Week 10. A disappointing Seattle game against Tampa Bay coupled with a Kansas City game that saw them losing for longer than they were winning are somewhat notable, but the most bizarre outcome had to be the Indianapolis Colts' win over the Houston Texans.

On average, the Colts were behind in the game by -10.7 points. More interestingly, they were behind against a team that had just shuffled their starting quarterback and primary backup in favor of an undrafted free agent—local boy Case Keenum, who broke NCAA passing records by playing at the same school that produced another NCAA record-buster, Andre Ware.

While it is inaccurate to say that they played in the same system, it is easy to compare the two as system quarterbacks. Keenum plays under St. Paul, Minnesota native Tony Levine, who is playing with an Air Raid offense developed under Mike Price and Kevin Sumlin while Andre Ware played in a Run and Shoot offense, one that is both strikingly similar and the complete opposite of the Air Raid offense.

At any rate, neither of them provided the quarterbacks with the pedigree they needed (allegedly) to succeed at an NFL level.

Instead, an undrafted free agent in his second year nearly took on and upset one of the best teams in the NFL, led by the most ballyhooed quarterback prospect in twenty years.

And just to make advanced statisticians angry, Andrew Luck pulled off another come-from-behind win to "set things right". This has hurt the Colts' efficiency on both sides of the ball, but efficiency doesn't put you into the playoffs.

If this keeps up, Texas will have two teams led by amazingly talented undrafted free agent quarterbacks who carry the burden of not getting it done. But unlike Romo, Keenum has a lot of time to turn the narrative around. And in his hometown, no less.

Power Rankings

Ranks below, first organized without recency bias. Again, if you want to get a primer on what these shortcut words mean, take a look here.

Like always, I strongly prefer the efficiency ranks over anything else, even the average I have presented below.

Remember, all headers are sortable:

Power Rankings - No Bias
Team	Efficiency	Points	Wins	Average	Average Rank
Seattle Seahawks	2	4	2	2.67	1
Kansas City Chiefs	4	5	1	3.33	2
Denver Broncos	5	1	5	3.67	3
New Orleans Saints	3	6	6	5.00	4
San Francisco 49ers	8	3	4	5.00	4
Indianapolis Colts	9	7	3	6.33	6
Carolina Panthers	1	2	19	7.33	7
Green Bay Packers	12	8	8	9.33	8
Dallas Cowboys	6	10	14	10.00	9
New England Patriots	11	11	9	10.33	10
Cincinnati Bengals	15	9	7	10.33	10
Detroit Lions	10	12	10	10.67	12
Arizona Cardinals	7	14	13	11.33	13
Chicago Bears	14	13	15	14.00	14
Tennessee Titans	16	15	12	14.33	15
San Diego Chargers	27	18	11	18.67	16
Buffalo Bills	17	21	20	19.33	17
Miami Dolphins	25	16	17	19.33	17
Philadelphia Eagles	13	20	26	19.67	19
Cleveland Browns	22	19	22	21.00	20
St. Louis Rams	18	23	24	21.67	21
Washington Redskins	20	22	25	22.33	22
Baltimore Ravens	29	17	21	22.33	22
New York Jets	23	27	18	22.67	24
Houston Texans	21	25	23	23.00	25
Atlanta Falcons	24	24	27	25.00	26
Oakland Raiders	30	30	16	25.33	27
Tampa Bay Buccaneers	19	26	32	25.67	28
New York Giants	26	29	29	28.00	29
Minnesota Vikings	28	28	31	29.00	30
Pittsburgh Steelers	31	31	28	30.00	31
Jacksonville Jaguars	32	32	30	31.33	32

Whither the Panthers win total, who continue to lag behind despite the fact that they have been performing far better than average against the teams they have played, but have only gathered their wins against "weak teams" and have lost to the strong ones. Their early season foibles are being put in stark contrast to their late-season dominance.

I look forward to their game against the 49ers, followed by another important (for national perception, at least) game against the Patriots.

Speaking of the Patriots, it didn't take long for their play to catch up to their wins after putting up an impressive performance against the Steelers, even when accounting for the Steelers' terrible play this year. You'd think New England's first four games were an extended preseason, but their recency-adjusted scores are not entirely impressive (Reminder, all headers are sortable):

Power Rankings - Recency Bias
Team	Efficiency	Points	Wins	Average	Average Rank
Seattle Seahawks	1	4	2	2.33	1
Kansas City Chiefs	4	6	1	3.67	2
Denver Broncos	6	1	6	4.33	3
San Francisco 49ers	8	3	3	4.67	4
Indianapolis Colts	5	7	4	5.33	5
Carolina Panthers	2	2	15	6.33	6
New Orleans Saints	3	8	8	6.33	6
Green Bay Packers	10	5	7	7.33	8
Cincinnati Bengals	12	9	5	8.67	9
Dallas Cowboys	7	10	13	10.00	10
Detroit Lions	11	12	10	11.00	11
Arizona Cardinals	9	13	12	11.33	12
New England Patriots	17	11	9	12.33	13
Tennessee Titans	13	15	16	14.67	14
Chicago Bears	14	14	17	15.00	15
San Diego Chargers	22	17	11	16.67	16
Philadelphia Eagles	16	16	26	19.33	17
Cleveland Browns	20	20	19	19.67	18
St. Louis Rams	15	21	24	20.00	19
Buffalo Bills	19	22	21	20.67	20
Washington Redskins	18	23	22	21.00	21
Oakland Raiders	24	30	14	22.67	22
Baltimore Ravens	30	18	20	22.67	22
Houston Texans	21	26	23	23.33	24
Miami Dolphins	28	19	25	24.00	25
New York Giants	23	24	27	24.67	26
New York Jets	27	31	18	25.33	27
Atlanta Falcons	26	25	29	26.67	28
Tampa Bay Buccaneers	25	27	32	28.00	29
Minnesota Vikings	29	28	31	29.33	30
Pittsburgh Steelers	31	29	28	29.33	30
Jacksonville Jaguars	32	32	30	31.33	32

Ye Gods, the Jaguars are bad.

Two teams that could surprise us by performing much better than expectations (aside from the Panthers, who I've been saying this about all season) are the Dallas Cowboys, the Arizona Cardinals, and crazily enough the Philadelphia Eagles. On the other end of that spectrum are the San Diego Chargers, Indianapolis Colts, the Cincinnati Bengals and the Miami Dolphins.

Of course, no one would be surprised by the Dolphins given what's happening over there. The Jaguars, despite the recent suspension of Justin Blackmon, are far and away the most functional locker room in the state of Florida (among professional NFL teams).

I won't hold out hope for the Eagles given the state of their quarterback situation (I really doubt Nick Foles is as good as that game would even begin to imply) and defense, but I do think the Cowboys and Cardinals are worth watching.

Similarly, it will be easy to point to Leon Hall and Geno Atkins as the reason for the Bengals decline, although their offensive numbers may be ignored in light of Giovani Bernard's excellent performances. The truth seems to be that they've had some issues all year, and those two players' absences will only magnify the issues already there.

The data are folding in the loss of Aaron Rodgers (Green Bay has a two-point rank differential between their overall efficiency score and their recency-biased score) and Eli Manning's quiet improvement, but not quickly enough. So proceed with caution, as always.

Against the Spread

Week 9 Record ATS

Efficiency: 10-3
Efficiency ("What Have You Done For Me Lately - Recency Bias"): 10-3
Point Differential: 7-6
Point Differential (WHYD4ML): 7-6

Weeks 6-9 Record ATS

Efficiency: 31-24
Efficiency ("What Have You Done For Me Lately - Recency Bias"): 28-27
Point Differential: 31-24
Point Differential (WHYD4ML): 27-28

The numbers without bias seems to be doing well recently, and it'll be interesting to see how they do as the season progresses, especially in an unusually injury-heavy year.

Picks

Efficiency:

Washington over MINNESOTA (+3)
Seattle over ATLANTA (+6.5)
Detroit over CHICAGO (+2.5)
Philadelphia over GREEN BAY (-7)
TENNESSEE (-13) over Jacksonville
INDIANAPOLIS (-10) over St. Louis
Oakland over NEW YORK GIANTS (-7.5)
Buffalo over PITTSBURGH (-3.5)
Cincinnati over BALTIMORE (+1)
Carolina over SAN FRANCISCO (-6)
ARIZONA (-2.5) over Houston
Denver over SAN DIEGO (+7)
Dallas over NEW ORLEANS (-7)
TAMPA BAY (+3) over Miami

Efficiency (WHYD4ML):

Washington over MINNESOTA (+3)
Seattle over ATLANTA (+6.5)
Detroit over CHICAGO (+2.5)
Philadelphia over GREEN BAY (-7)
Jacksonville over TENNESSEE (-13)
St. Louis over INDIANAPOLIS (-10)
Oakland over NEW YORK GIANTS (-7.5)
Buffalo over PITTSBURGH (-3.5)
Cincinnati over BALTIMORE (+1)
Carolina over SAN FRANCISCO (-6)
ARIZONA (-2.5) over Houston
Denver over SAN DIEGO (+7)
Dallas over NEW ORLEANS (-7)
TAMPA BAY (+3) over Miami

The New York Giants line seems dumb, but what do I know. The efficiency score is not very confidence about the Chicago line and the recency-adjusted efficiency scores are iffy on the Tampa Bay game, but for the most part they are confident.

Point Differential:

MINNESOTA (+3) over Washington
Seattle over ATLANTA (+6.5)
CHICAGO (+2.5) over Detroit
GREEN BAY (-7) over Philadelphia
TENNESSEE (-13) over Jacksonville
INDIANAPOLIS (-10) over St. Louis
Oakland over NEW YORK GIANTS (-7.5)
Buffalo over PITTSBURGH (-3.5)
Cincinnati over BALTIMORE (+1)
Carolina over SAN FRANCISCO (-6)
ARIZONA (-2.5) over Houston
Denver over SAN DIEGO (+7)
Dallas over NEW ORLEANS (-7)
Miami over TAMPA BAY (+3)

Point Differential (WHYD4ML):

Washington over MINNESOTA (+3)
Seattle over ATLANTA (+6.5)
CHICAGO (+2.5) over Detroit
GREEN BAY (-7) over Philadelphia
TENNESSEE (-13) over Jacksonville
INDIANAPOLIS (-10) over St. Louis
NEW YORK GIANTS (-7.5) over Oakland 
Buffalo over PITTSBURGH (-3.5)
Cincinnati over BALTIMORE (+1)
Carolina over SAN FRANCISCO (-6)
ARIZONA (-2.5) over Houston
Denver over SAN DIEGO (+7)
Dallas over NEW ORLEANS (-7)
Miami over TAMPA BAY (+3)

I am glad the points model disagrees with the efficiency model so much. It really allows me to test some of the things I've looked at over the weeks.

Also, it seems like there's at least a small clue as to why the line for the Oakland-New York game was set the way it was—perhaps a slight overreaction to the Giants' recent improvement and the stinker the Raiders put up against the Chargers. Also, I imagine markets determine a lot of the line, and the New York betting market might be decidedly larger than San Francisco's.

Strength Schedule

This year, more than most years, it seems, the discussion of strength of schedule has played a big role. Naturally, for these types of rankings, I have a strength of schedule data table I use to adjust everyone's scores. Given the need for people to know these things in light of the Chiefs' and Bronco's respective records, we can take a look below.

But I want to stress that arguments about strength of schedule will mostly already be coopted by the data above. Saying that the "Chiefs rank highly but did it against poor teams" will make no sense at all, because they have performed better against those teams than other teams did—and by a significant margin. So on down the list.

Again, everything is nicely sortable. The ranked order is from toughest to easiest.

Strength of Schedule
Team	Efficiency	Points	Wins	Average	Average Rank
Jacksonville Jaguars	6	1	1	2.67	1
Houston Texans	8	2	2	4.00	2
Arizona Cardinals	1	4	12	5.67	3
Buffalo Bills	7	5	10	7.33	4
Washington Redskins	12	6	5	7.67	5
New York Giants	3	3	18	8.00	6
Tampa Bay Buccaneers	2	7	21	10.00	7
St. Louis Rams	5	14	13	10.67	8
Miami Dolphins	14	8	15	12.33	9
Tennessee Titans	16	18	13	12.33	10
San Francisco 49ers	10	22	7	13.00	11
Minnesota Vikings	11	10	20	13.67	12
Detroit Lions	13	11	17	13.67	12
Dallas Cowboys	17	17	9	14.33	14
Cleveland Browns	26	9	8	14.33	14
Atlanta Falcons	4	13	28	15.00	16
Seattle Seahawks	9	25	11	15.00	16
Green Bay Packers	22	12	14	16.00	18
Indianapolis Colts	24	20	4	16.00	18
Oakland Raiders	27	23	6	18.67	20
New York Jets	15	15	27	19.00	21
Cincinnati Bengals	23	19	16	19.33	22
Baltimore Ravens	28	16	19	21.00	23
Philadelphia Eagles	20	24	22	22.00	24
Chicago Bears	21	21	25	23.33	25
New Orleans Saints	19	27	31	25.67	26
San Diego Chargers	25	30	23	26.00	27
Carolina Panthers	18	29	32	26.33	28
Pittsburgh Steelers	31	26	24	27.00	29
New England Patriots	29	28	29	28.67	30
Denver Broncos	32	31	26	29.67	31
Kansas City Chiefs	30	32	30	30.67	32

No, the Jaguars do not have the toughest strength of schedule because playing them boosts all the scores. Removing games that teams have played in would produce the exact same results, because that is the point of the iterative function. The Jaguars have played the Chiefs, Broncos, Colts, Seahawks and 49ers, all of whom are top ten in every ranking system.

Their schedule gets easier from here on out (teams ranked 16th, 7th, 21st, 22nd, 21st, 17th, 16th and 9th in efficiency, respectively) but not as easy as many have advertised. More importantly, their performance against those top-ranked teams still kills their chances and emphasizes how bad they really are.

In all likelihood, the Vikings are fighting with the Bucs, Steelers and Falcons (and Giants and Texans) for spots #2, #3, or #4 (or $5 or #6) in the draft. The Bucs have the talent to win, but not the coaching staff, while the Steelers have an aging/injury problem as well as some pointed questions of their own coach they have to answer. The Falcons have significant injuries to deal with while the Giants and Texans are just a mess to deal with.

Everything, naturally, is subject to change. But I wouldn't hold out hope that the Jaguars won't get the top pick.

Perhaps that table does underscore, however, how bad the Steelers really are this year. Putting up those numbers against the teams that they did is truly abysmal.

Perhaps the best takeaway, however, is that half the teams at the top of most power rankings have had a weak strength of schedule. Isolating one doesn't seem right to me.

Of course, we'll want to look at remaining strength of schedule (headers sortable):

Future Strength of Schedule
Team	Efficiency	Points	Wins	Average	Average Rank
St. Louis Rams	3	1	3	2.33	1
San Diego Chargers	9	2	1	4.00	2
Atlanta Falcons	2	3	8	4.33	3
Minnesota Vikings	7	6	5	6.00	4
New Orleans Saints	1	4	21	8.67	5
Tampa Bay Buccaneers	5	7	15	9.00	6
New York Giants	11	11	6	9.33	7
Carolina Panthers	6	10	14	10.00	8
Kansas City Chiefs	21	5	4	10.00	8
Oakland Raiders	13	14	7	11.33	10
Denver Broncos	19	15	2	12.00	11
Arizona Cardinals	10	20	9	13.00	12
San Francisco 49ers	4	13	23	13.33	13
Philadelphia Eagles	8	12	20	13.33	13
New England Patriots	14	8	18	13.33	13
Pittsburgh Steelers	23	9	10	14.00	16
Jacksonville Jaguars	12	22	17	17.00	17
Dallas Cowboys	15	19	19	17.67	18
Indianapolis Colts	17	25	11	17.67	18
Baltimore Ravens	24	18	16	19.33	20
Chicago Bears	20	16	26	20.67	21
Seattle Seahawks	16	23	25	21.33	22
New York Giants	26	17	22	21.67	23
Washington Redskins	18	21	28	22.33	24
Houston Texans	27	28	13	22.67	25
Tennessee Titans	28	30	12	23.33	26
Miami Dolphins	22	26	29	25.67	27
Green Bay Packers	25	27	30	27.33	28
Cincinnati Bengals	32	24	27	27.67	29
Cleveland Browns	30	31	24	28.33	30
Detroit Lions	29	29	32	30.00	31
Buffalo Bills	31	32	31	31.33	32

What strikes out at me is that there is a significant difference between strength of schedule and the strength of schedule of the other two measures. I'll be bold enough to say that this means most analyses of "remaining strength of schedule" are not that great because they rely on wins and do not concentrate on performance.

UPDATE: I read this second set of SOS data incorrectly when talking about it in the original posting of this entry. Ignore the things and read it like you read the one above it—lower rank number (1, 2, 3) means harder, higher rank number means easier (30, 31, 32). Go Bills!

Average Margin of Victory

I've liked this measure for some time as it does the best job of parsing out certain types of data and emphasizes who has been in control of a ball game. Unfortunately, it likes to de-emphasize the fourth quarter a bit too much, but I think it's the best points-related measure of team quality out there today.

To recap, it's simply the average point differential a team has at any point in the game—if two teams win 31-3, but accumulate their points at different times, they'll have a different average point differential. If the first team scored 10 points off the bat and steadily accumulated points over the game, they'd end up with a higher average point differential than the second team if that second team did almost all of its scoring in the fourth quarter.

Again, I stole this straight from Chase Stuart at Football Perspective, and he calls them game scripts and uses them for a purpose they are much better suited to: pass/run ratios. All I've done is adjust for strength-of-schedule and attach importance to it.

Still, it's demonstrative of something in some way.

Rank	Team	Average MOV
1	Carolina Panthers	6.9
2	Green Bay Packers	5.9
3	San Francisco 49ers	5.4
4	New Orleans Saints	3.9
5	Denver Broncos	3.7
6	Cincinnati Bengals	3.5
7	Dallas Cowboys	3.4
8	New England Patriots	2.6
9	Kansas City Chiefs	2.4
10	Seattle Seahawks	2.8
11	Detroit Lions	1.2
12	Atlanta Falcons	1.1
13	Indianapolis Colts	0.3
14	Arizona Cardinals	0.3
15	Buffalo Bills	0.1
16	Miami Dolphins	-0.1
17	San Diego Chargers	-0.2
18	Chicago Bears	-0.3
19	Cleveland Browns	-0.5
20	Philadelphia Eagles	-0.6
21	Tampa Bay Buccaneers	-0.8
22	Tennessee Titans	-1.1
23	Minnesota Vikings	-1.8
24	New York Giants	-2.0
25	Oakland Raiders	-2.1
26	Baltimore Ravens	-2.7
27	Houston Texans	-3.3
28	St. Louis Rams	-3.4
29	New York Jets	-3.4
30	Washington Redskins	-3.9
31	Pittsburgh Steelers	-5.3
32	Jacksonville Jaguars	-12.2

They're beginning to flatten out in general, but the Packers took a big hit with Aaron Rodgers out. For the most part, there's been little change.

Also taking a large tumble are the Seattle Seahawks, which are really no surprise given their last two games.

Jacksonville will keep on doing it's thing. Normally dropping 1.5 points in this measure would move a team about 6 ranks. The Jaguars, however, stay at 32. The difference between the 32nd ranked team and the 31st ranked team is the same as the distance between 31st and 10th. Ouch.

For historical perspective, the Jaguars' non-SOS adjusted average MOV is -13.2. The 0-16 Lions and the 2-14 Colts were at -7.6 and -7.1 at the end of their seasons.

This is currently one of the worst teams in history. If their average MOV for the rest of the season is simply -1.0, they'll end with a season total at -7.1. And that will require consistent close play.