The past couple of weeks, I've argued that future performance can better be predicted by breaking down each team into four parts and ignoring things like win-loss records, points differential, and betting lines. By looking at a team's offensive and defensive passing and rushing efficiencies, I think you get a better sense of how good each team really is. The more granulated the data, the better. In a game, you only get one win or loss and 50 or so points scored, but there are over 150 plays in most games. Understanding how each team does when they pass or run, and try to stop the pass and run, are key tools to understanding how good each team is.
|1||New Orleans Saints||2.28||0.36||2.64||0.43||-0.25||0.18||2.82||14.1|
|4||Green Bay Packers||0.56||0.05||0.60||0.83||0.62||1.45||2.06||10.3|
|7||New York Jets||-0.43||0.47||0.04||1.28||0.24||1.52||1.56||7.8|
|9||San Diego Chargers||1.89||-0.88||1.02||0.52||-0.15||0.37||1.39||6.9|
|10||New York Giants||0.84||0.05||0.89||0.31||0.15||0.46||1.35||6.8|
|12||New England Patriots||1.23||-0.19||1.05||-0.04||-0.13||-0.17||0.88||4.4|
|19||San Francisco 49ers||-0.74||0.12||-0.62||-0.01||0.59||0.58||-0.04||-0.2|
|27||St. Louis Rams||-1.16||0.32||-0.84||-0.70||-0.39||-1.09||-1.92||-9.6|
|28||Tampa Bay Buccaneers||-0.90||0.00||-0.90||-0.57||-0.56||-1.14||-2.04||-10.2|
|31||Kansas City Chiefs||-1.50||-0.51||-2.02||-1.19||-0.23||-1.42||-3.43||-17.2|
I've also calculated each team's SRS grade, using the same formula that I did with my college ranking system. So big wins are slightly reduced, close wins are given a bit more credit, and road teams are given three points, compared to using the SRS with just inputs of points scored and points allowed.
|1||New Orleans Saints||15.4||-2.0||13.4||12-0|
|3||New England Patriots||7.0||2.5||9.4||7-5|
|5||San Diego Chargers||8.1||-2.0||6.1||9-3|
|8||Green Bay Packers||7.4||-2.7||4.7||8-4|
|11||New York Jets||2.5||0.9||3.3||6-6|
|15||New York Giants||1.5||0.6||2.1||7-5|
|18||San Francisco 49ers||0.9||0.0||1.0||5-7|
|28||Tampa Bay Buccaneers||-11.2||2.0||-9.2||1-11|
|29||Kansas City Chiefs||-10.0||0.6||-9.5||3-9|
|32||St. Louis Rams||-13.5||0.9||-12.7||1-11|
To predict the games for week 14, I've decided to combine the SOS component from the SRS system above with my points score from my yardage-based predictive system. So the Giants move from +6.8 to +7.4 while the Saints move from +14.1 to +12.1. While this wasn't officially part of my system before, this just feels right, and it's something I'd been doing when making the final picks, anyway. Listed in the table below are the names of the road team, home team, the road team's rating, the home team's rating, the projected point spread (with 3 points given to the home team), the actual point spread from the road team's point of view, and the differential. Any differential larger than +/- 5 is a game we want to examine.
|Pittsburgh Steelers||@||Cleveland Browns||9.9||-18.2||-25.1||-10||-15.1|
|New Orleans Saints||@||Atlanta Falcons||12.1||-6.3||-15.4||-10||-5.4|
|Green Bay Packers||@||Chicago Bears||7.6||-2.3||-6.9||-3||-3.9|
|Denver Broncos||@||Indianapolis Colts||7.3||9.0||4.7||+7||-2.3|
|San Diego Chargers||@||Dallas Cowboys||4.9||8.4||6.5||+3||3.5|
|Seattle Seahawks||@||Houston Texans||-6.6||-0.5||9.1||+6||3.1|
|Miami Dolphins||@||Jacksonville Jaguars||-4.6||-0.4||7.2||+3||4.2|
|Buffalo Bills||@||Kansas City Chiefs||-4.0||-16.6||-9.6||+1||-10.6|
|Cincinnati Bengals||@||Minnesota Vikings||1.7||3.5||4.8||+6.5||-1.7|
|Carolina Panthers||@||New England Patriots||-2.0||6.9||11.9||+13.5||-1.6|
|Philadelphia Eagles||@||New York Giants||9.2||7.4||1.2||+1||0.2|
|St. Louis Rams||@||Tennessee Titans||-8.7||4.4||16.2||+13||3.2|
|Washington Redskins||@||Oakland Raiders||-0.2||-14.6||-11.3||-1||-10.3|
|Detroit Lions||@||Baltimore Ravens||-12.5||4.5||19.9||+13||6.9|
|New York Jets||@||Tampa Bay Buccaneers||8.7||-8.2||-13.9||-3||-10.9|
|Arizona Cardinals||@||San Francisco 49ers||0.4||-0.2||2.4||-3.5||5.9|
The strongest pick of the week? Pittsburgh @ Cleveland. The Steelers are likely overrated in this system because they're a much worse team without Troy Polamalu (who won't play tonight), they've got terrible special teams, and they've managed to underachieve all season long. The Browns have now foiled my system twice, by covering in closer-than-expected losses to the Bengals and Chargers. Pittsburgh was the big stinker in my system last week, losing at home to the lowly Raiders. The weather is expected to be awful tonight, in what is sure to be another miserable NFL Network telecast. All the subjective signs are saying to stay away from Pittsburgh covering this 10-point spread, but the objective signs (PIT -25.1) say to take it. Take the Steelers and hope history stops repeating itself.
The next best bet is the Bills over the Chiefs. Kansas City is favored to win. And the game isn't against Cleveland, Detroit or Oakland. When this starts making sense to you, let me know. Last week, I thought the Kansas City line was way off in the favor of the Chiefs being a competitive team. I wrote "Do people forget how bad Kansas City is? If no one has watched Kansas City play poorly, did it really happen?" The Chiefs then lost at home by 31 points. Buffalo isn't great, but they're much better than Kansas City. Take the Bills.
Another team that is 10 points undervalued is the Redskins, who are only one-point favorites in Oakland. I understand that Bruce Gradkowski is playing a lot better than JaMarcus Russell, so maybe the Raiders are undervalued in my system. But Oakland seems to play well a few times a season and miserable for the rest of the year. Washington is coming off a heartbreaking loss, traveling across the country, and the Raiders are hot. Those factors drove down this spread, but they don't change the fact that Washington is a much better team than the Raiders. Don't forget, the Thanksgiving embarassment against the Cowboys was just one game before last for Oakland. Take Washington.
There's one other team that's 10 points undervalued -- the Jets. Initially, the spread opened at New York -4; after news came out that Mark Sanchez wasn't playing, the line dropped just one point. Kellen Clemens may not be very good, but he won't have to be. The Jets rank 4th in rushing efficiency, 1st in defensive pass efficiency, and 1st in total defensive efficiency. The Jets have struggled this year because of the passing offense; while Clemens looked very rusty against the Bills, I doubt he can be significantly worse. While the Jets have a great running game and defense, Tampa isn't good at anything. The only area where the Bucs aren't in the bottom ten is rushing offense, where they are league average. Tampa ranks 31st against the run, making this a terrible matchup for the Bucs. Take New York.
There's one other road team that looks undervalued; New Orleans. The Saints are only 10 point favorites in Atlanta, and New Orleans is obviously a team that can run up the score on almost anyone. With Matt Ryan and Michael Turner both questionable for Atlanta, this game seems pretty risky. Considering that the SRS indicates that the spread should be right around 10, I say we should avoid this game.
There are two home teams that look undervalued. San Francisco should be about a two- or three-point favorite against the Cardinals, but instead are 3.5 point underdogs. I'm not sure if this is a great matchup for the 49ers; they're terrific at stopping the run but Arizona is a pass-happy team. Still, I don't see a strong reason to object to what the numbers say. San Francisco is likely undervalued here, because the passing game has improved significantly with Alex Smith running the spread offense the past few weeks. Take the 49ers.
Finally, again, bet against the Lions. The Ravens have a great run defense, the Lions don't have Matthew Stafford, and Baltimore is at home. This has the feel of a game where the Lions get held out of the end zone by the Ravens defense, while Baltimore kicks in a defensive or special teams touchdown of their own. Take Baltimore.
To recap, I like: Pittsurgh -10, Buffalo +1, Washington -1, the Jets -3, San Francisco +3.5, and Baltimore -13.
This entry was posted on Thursday, December 10th, 2009 at 7:00 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.