On Monday, I posited a way to predict future NFL team performance while ignoring things like points scored, points allowed and win-loss record. Instead, I argued that we should break each team into four parts: passing efficiency, rushing efficiency, defensive passing efficiency and defensive rushing efficiency.
Passing efficiency is defined as simply net passing yards per attempt ((passing yards minus sack yards lost), divided by (pass attempts plus sacks)). Rushing efficiency is just yards per carry; the defensive statistics follow the same method. After comparing each team to the league average in each of the four categories, we can see how many yards per attempt better than average each team is. We add up the four yards per attempt measures relative to league average to come up with a total score, multiply by five, and viola! Team ratings!
Does this sound kooky? Sure. But my research indicates that it may be more accurate than common perception. Let's see what the numbers say through ten weeks. For both offensive and defensive ratings, positive numbers are good and negative numbers are bad. In addition to the four categories (passing offense, rushing offensive, passing defense, rushing defense), I've included total offense and total defense grades, for you to sort as you desire.
|1||New Orleans Saints||1.88||0.52||0.71||-0.32||2.40||0.39||2.79||13.9|
|6||Green Bay Packers||0.47||0.22||0.38||0.58||0.69||0.97||1.65||8.3|
|8||New York Giants||0.99||0.05||0.60||-0.11||1.04||0.49||1.53||7.7|
|9||New York Jets||-0.31||0.43||0.85||0.28||0.12||1.13||1.25||6.3|
|10||New England Patriots||1.11||-0.23||0.55||-0.21||0.88||0.35||1.23||6.1|
|13||San Diego Chargers||1.16||-0.77||0.37||0.01||0.40||0.38||0.78||3.9|
|20||San Francisco 49ers||-0.87||0.24||-0.21||0.73||-0.63||0.52||-0.11||-0.5|
|27||St. Louis Rams||-0.89||0.35||-1.03||-0.44||-0.54||-1.47||-2.01||-10.0|
|30||Tampa Bay Buccaneers||-1.29||-0.06||-0.98||-0.67||-1.36||-1.66||-3.01||-15.1|
|31||Kansas City Chiefs||-1.25||-0.64||-0.99||-0.23||-1.89||-1.22||-3.11||-15.6|
Remember, these rankings are pretty basic. No adjustment was made for strength of schedule, which means this undervalues teams like the Dolphins, Titans and Jets, and overvalues teams like the Vikings, Saints and Packers. Teams who have suffered serious injuries will be overvalued; teams who are seeing new players emerge are undervalued. Special teams are totally ignored in this analysis. All that said, this basic system has had some very good historical success against the spread, especially when the difference between the projected and actual point spreads are large.
Here are the week 12 projected point spreads:
Green Bay Packers @ Detroit Lions (+19.2)
Oakland Raiders @ Dallas Cowboys (-28.1)
New York Giants @ Denver Broncos (+0.6)
Tampa Bay Buccaneers @ Atlanta Falcons (-14)
Miami Dolphins @ Buffalo Bills (-2.9)
Cleveland Browns @ Cincinnati Bengals (-26.5)
Indianapolis Colts @ Houston Texans (+8.8)
Carolina Panthers @ New York Jets (-9.3)
Washington Redskins @ Philadelphia Eagles (-10.8)
Seattle Seahawks @ St. Louis Rams (+0.7)
Kansas City Chiefs @ San Diego Chargers (-22.4)
Jacksonville Jaguars @ San Francisco 49ers (-3.2)
Arizona Cardinals @ Tennessee Titans (-1.6)
Chicago Bears @ Minnesota Vikings (-11.6)
Pittsburgh Steelers @ Baltimore Ravens (+5.3)
New England Patriots @ New Orleans Saints (-10.8)
And here are the week 12 actual point spreads:
Green Bay Packers @ Detroit Lions (+10.5)
Oakland Raiders @ Dallas Cowboys (-13.5)
New York Giants @ Denver Broncos (+6.5)
Tampa Bay Buccaneers @ Atlanta Falcons (-11.5)
Miami Dolphins @ Buffalo Bills (+3)
Cleveland Browns @ Cincinnati Bengals (-14)
Indianapolis Colts @ Houston Texans (+3.5)
Carolina Panthers @ New York Jets (-3)
Washington Redskins @ Philadelphia Eagles (-9)
Seattle Seahawks @ St. Louis Rams (+3)
Kansas City Chiefs @ San Diego Chargers (-13.5)
Jacksonville Jaguars @ San Francisco 49ers (-3)
Arizona Cardinals @ Tennessee Titans (OFF)
Chicago Bears @ Minnesota Vikings (-10.5)
Pittsburgh Steelers @ Baltimore Ravens (OFF)
New England Patriots @ New Orleans Saints (-3)
Going game by game, we see:
Green Bay is undervalued by nearly 9 points (Detroit is about 19 points worse than GB; the Packers are only being given 10 points)
Dallas is undervalued by nearly 15 points
Denver is undervalued by about 6 points
Atlanta is undervalued by 2.5 points
Buffalo should be favored by 3; instead, they're 3-point dogs.
Cincinnati is undervalued by 12.5 points
Indianapolis is undervalued by 5.3 points
The Jets are undervalued by 6.3 points
Philadelphia is undervalued by 1.8 points
The Rams are undervalued by about 2 points
San Diego is undervalued by about 9 points
SF/JAX game is properly valued
Arizona game is off due to Warner's injury; I suspect with a healthy Warner, the line would be close to the projected line
Minnesota is undervalued by about 1 point
Pittsburgh game is off due to Roethlisberger's injury; I suspect Baltimore would probably be getting fewer than 5 points if Ben was healthy.
The Saints are undervalued by about 8 points
So now what? What does this all mean? If I was a betting man -- which I am not -- here's what I would do:
Dallas -14 (vs. the Raiders)
- Has Dallas suffered any major injuries? No.
Has Dallas played a really easy schedule? They've had the 28th hardest (or 5th easiest) SOS according to Sagarin.
Has Oakland improve their team? I suppose benching Russell and playing Gradkowski could be considered a good thing.
Has Oakland played a really difficult schedule? They've got the 5th toughest SOS according to Sagarin.
Is Oakland particularly capable of playing in Dallas due to proximity or past history? No.
Add up all of those things, and maybe Dallas shouldn't really be a 28-point favorite. But I still don't think I can justify this line. I'd confidently take the Cowboys.
Cincinnati -14 (vs. the Browns)
- Have the Bengals suffered any major injuries? Benson's hurt, but is likely to play against Cleveland.
How's their SOS? 20th
What about Cleveland? Brady Quinn could be on the verge of breaking out. Or something like that.
How's their SOS? 8th.
Is Cleveland particularly capable of playing well in Cleveland? They play there once a year and they're in the same state. On the other hand, since 1978, the road team has won only 41% of games in this Ohio rivalry.
San Diego -13.5 (vs. the Chiefs)
- Any injury issues for SD? Nothing new.
KC? Dwayne Bowe is out of the lineup, although Chris Chambers was the hero last week.
HFA? KC plays in SD once a year, but otherwise this is not an easy trip to make.
I don't see anything to make me feel uncomfortable about the Chargers giving 13.5 points to Kansas City.
Green Bay -10.5 (@ Detroit)
Injuries: Not only is Green Bay relatively healthy, but Matthew Stafford is not. What's up with this line?
SOS: Green Bay has played cupcakes so far, evidenced by their third weakest SOS. Of course, Detroit's just another cupcake, and they've had a pretty easy schedule (#24), too.
HFA? Green Bay hasn't always fared well indoors at Detroit, even against some bad Lions teams. That's particularly true on Thanksgiving, where Mike McMahon nearly beat them in '01 and Joey Harrington actually did beat them in '03. But GB blew out the Lions in Detroit last season, and I think we'll see a repeat of that again this year.
Take both Thanksgiving Day favorites (more on that Thanksgiving Night game, later).
New Orleans -3
- Injuries: the Saints have been hit by the injury bug of late, and (last weekend notwithstanding) have struggled on the field because of that in close wins over Miami, Atlanta, Carolina and St. Louis. The Patriots are pretty healthy.
SOS: New Orleans has faced a super easy schedule (#31) while the Patriots have faced a slightly tougher than average slate.
HFA? This would be the spot where an unprofessional journalist would make some crack about how the Pats are intimiately aware of how to tape practices before games at the Superdome. I won't be doing that; instead I'll say that generally, long trips to distinctly different climates from a team's home climate in stadiums in which such teams are unfamiliar with are generally tough road games. The Patriots have not played in New Orleans since Super Bowl XXXVI.
I don't like taking what looks like a slumping NO team against a streaking NE team. But I'm sure there have been many times where teams had injuries/slumped and played hot teams in my dataset, and that's why the line was lower than projected. So I'm not going to use that as a reason to back off what the data said on Monday, so I'll take the Saints.
Four other games are close -- the Jets, Broncos, Bills and Colts are all undervalued by 5 or 6 points. The remainder of the games are either properly valued, or "off." I'd avoid the Jets game because of the big injuries to Leon Washington and Kris Jenkins, and the fact that my analysis ignores turnovers (they're generally unpredictable, but I'm going to make an exception for Mark Sanchez).
Denver is a 6.5 point dog to the Giants, when they really are just about even with New York as long as they're at home. The Giants have faced an easier schedule than the Broncos so far and playing in Denver is never easy. No, I don't feel excited about it, but I'll back the numbers here and say to take Denver.
The Bills/Dolphins game worries me because Miami has faced the #1 SOS so far this year while the Bills have faced the 21st hardest one; even without Ronnie Brown, I don't feel confident taking the worse team in my rankings that is overvalued due to SOS and only getting three points.
The Colts are better by about a half a yard per play in each of the four categories; they've faced a slightly easier schedule than Houston, but I don't think the HFA is significant when the Colts play in a dome and Indy plays in Houston's dome once a year. The numbers say to be cautious -- the Colts are only undervalued by 5.3 points -- but that passes my threshold.
So to recap, my system (tweaked by my subjective thoughts based on things the system ignores) says to take Dallas, Cincinnati, San Diego, Green Bay, the Broncos and the Colts. Whatever the results, I'll keep them to the comments, and then run this again next Wednesday. Anyone else care to make some predictions?
This entry was posted on Wednesday, November 25th, 2009 at 8:38 am and is filed under Statgeekery. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.