Here are the current SRS Ratings, weighted for the recency of each game, along with each team’s quarter-by-quarter Win Probability Added (WPA) so far this season:
Rank | Team | pfr_id | OSRS | OSOS | DSRS | DSOS | SRS | SOS | wpa_loc | wpa_vegas | wpa_1st | wpa_2nd | wpa_3rd | wpa_4th/ot |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Chicago Bears | chi | 7.9 | 1.0 | 8.0 | -1.4 | 15.9 | -0.4 | -0.068 | 0.476 | 0.128 | 0.017 | 0.729 | 0.219 |
| 2 | San Francisco 49ers | sfo | -0.2 | -0.1 | 11.4 | 2.2 | 11.2 | 2.0 | 0.068 | 0.956 | -0.066 | 0.059 | 0.566 | -0.083 |
| 3 | Houston Texans | htx | 6.9 | -1.0 | 4.2 | -0.2 | 11.1 | -1.2 | 0.068 | 1.419 | -0.038 | 0.943 | 0.127 | -0.020 |
| 4 | Green Bay Packers | gnb | 7.5 | 3.9 | 2.8 | 1.9 | 10.4 | 5.7 | -0.068 | 0.926 | 0.596 | -0.011 | -0.446 | -0.498 |
| 5 | New York Giants | nyg | 7.7 | 1.6 | 2.3 | -1.6 | 10.0 | -0.1 | 0.068 | 0.351 | -0.060 | 0.278 | 0.029 | 0.834 |
| 6 | Denver Broncos | den | 5.2 | -0.1 | 1.6 | 1.5 | 6.8 | 1.4 | 0.000 | -0.060 | -0.644 | -0.816 | 0.183 | 1.338 |
| 7 | New England Patriots | nwe | 7.4 | -0.4 | -0.7 | -0.3 | 6.7 | -0.7 | -0.068 | 1.209 | 0.737 | -0.017 | 0.363 | -1.724 |
| 8 | Seattle Seahawks | sea | -2.7 | 4.1 | 8.7 | 0.7 | 5.9 | 4.8 | -0.068 | -0.335 | 0.708 | -0.588 | -0.094 | 0.877 |
| 9 | Atlanta Falcons | atl | 2.5 | -2.5 | 3.2 | -1.2 | 5.7 | -3.6 | 0.000 | 0.632 | 0.525 | 0.199 | 0.404 | 1.240 |
| 10 | St Louis Rams | ram | -2.2 | 2.6 | 5.2 | 1.9 | 3.0 | 4.6 | 0.068 | -1.039 | -0.143 | 0.348 | -0.283 | 0.550 |
| 11 | Minnesota Vikings | min | -0.6 | -1.3 | 3.2 | -1.3 | 2.6 | -2.6 | 0.068 | 0.211 | 0.835 | -0.362 | 0.075 | 0.673 |
| 12 | Washington Redskins | was | 5.8 | 0.6 | -3.2 | 1.9 | 2.6 | 2.5 | -0.068 | -0.381 | 0.172 | 0.697 | 1.235 | -2.155 |
| 13 | Dallas Cowboys | dal | -0.7 | 3.3 | 2.5 | 1.5 | 1.7 | 4.9 | -0.136 | 0.365 | -0.216 | -0.359 | 0.051 | 0.295 |
| 14 | Arizona Cardinals | crd | -4.6 | 1.2 | 5.0 | -1.1 | 0.4 | 0.1 | 0.068 | -0.452 | -0.085 | 0.088 | -0.331 | 1.212 |
| 15 | Miami Dolphins | mia | -2.9 | 0.4 | 3.1 | -0.9 | 0.2 | -0.5 | 0.000 | -0.549 | -0.011 | 0.726 | 1.272 | -1.436 |
| 16 | Tampa Bay Buccaneers | tam | -0.6 | -2.7 | 0.8 | 0.5 | 0.1 | -2.2 | 0.136 | -0.531 | 0.583 | -0.276 | 0.276 | -1.188 |
| 17 | New York Jets | nyj | 0.0 | 0.6 | 0.1 | 1.0 | 0.1 | 1.5 | 0.068 | -0.605 | -0.080 | -0.084 | -0.495 | 0.695 |
| 18 | Baltimore Ravens | rav | 0.2 | -0.8 | -1.5 | -1.1 | -1.3 | -1.9 | 0.068 | 0.536 | -0.452 | 0.587 | 0.057 | 0.705 |
| 19 | New Orleans Saints | nor | 5.6 | -0.7 | -7.2 | -0.4 | -1.6 | -1.2 | 0.000 | 0.552 | -0.193 | -0.476 | -0.019 | -0.864 |
| 20 | Detroit Lions | det | 4.9 | 2.9 | -6.5 | -2.4 | -1.6 | 0.5 | -0.068 | 0.249 | -0.313 | -0.833 | -0.353 | 0.819 |
| 21 | Pittsburgh Steelers | pit | -2.5 | -2.6 | 0.1 | -1.3 | -2.4 | -3.9 | -0.136 | 0.700 | -0.027 | 0.290 | 0.163 | -0.991 |
| 22 | San Diego Chargers | sdg | -2.3 | -3.8 | -0.4 | -0.1 | -2.7 | -3.9 | 0.000 | 0.202 | 0.753 | 0.214 | 0.294 | -1.463 |
| 23 | Carolina Panthers | car | -4.1 | 1.5 | 0.7 | 1.4 | -3.4 | 2.9 | 0.136 | -0.246 | -0.833 | 0.124 | -0.212 | -0.968 |
| 24 | Philadelphia Eagles | phi | -6.6 | -0.6 | 2.5 | 0.3 | -4.0 | -0.4 | 0.000 | 0.466 | -0.675 | -0.296 | 0.729 | -0.224 |
| 25 | Indianapolis Colts | clt | -3.2 | 0.7 | -3.1 | -0.4 | -6.3 | 0.3 | 0.136 | -0.592 | 0.446 | -0.291 | -0.066 | 0.367 |
| 26 | Cincinnati Bengals | cin | -0.9 | -1.2 | -5.7 | -2.6 | -6.7 | -3.8 | -0.068 | 0.009 | 0.762 | 0.304 | -1.197 | -0.311 |
| 27 | Cleveland Browns | cle | -5.1 | -3.0 | -2.6 | -0.1 | -7.7 | -3.1 | -0.068 | -1.153 | -0.411 | -0.800 | 0.161 | -0.229 |
| 28 | Oakland Raiders | rai | -3.4 | 0.6 | -6.6 | -1.5 | -10.0 | -0.9 | 0.000 | -0.368 | -0.390 | 0.014 | -1.529 | 1.272 |
| 29 | Buffalo Bills | buf | 1.1 | 0.0 | -11.1 | -1.9 | -10.0 | -1.9 | -0.068 | -0.303 | 0.313 | -0.034 | 0.180 | -0.589 |
| 30 | Tennessee Titans | oti | -3.0 | -1.6 | -8.4 | 2.1 | -11.4 | 0.5 | -0.068 | -1.194 | -0.596 | 0.449 | -0.667 | 1.576 |
| 31 | Jacksonville Jaguars | jax | -8.8 | -0.3 | -3.3 | 1.0 | -12.1 | 0.8 | 0.000 | -0.647 | -0.229 | -0.144 | -0.227 | -0.754 |
| 32 | Kansas City Chiefs | kan | -8.9 | -2.6 | -6.1 | 1.0 | -15.0 | -1.6 | 0.000 | -0.802 | -1.094 | 0.049 | -0.977 | 0.823 |
{ 7 comments… read them below or add one }
Riffing off of Chase’s NYT post about conference balance a few weeks ago, the bottom 8 teams are all from the AFC, while 4 of the top 5 teams are from the NFC. The average NFC team is +3.8, while the avg AFC team is -3.7 (the difference being due to the weighting of games and teams not all playing the same # of games).
Wow, the Dolphins are the 4th-best AFC team?
Looks like it. Fwiw, PFR’s straight-up SRS (unweighted) also has them 4th, and Sagarin’s “pure points” rating has them 5th in the AFC behind Baltimore.
But I’m not really sure what’s driving those results, since SRS doesn’t think all that highly of their opponents… It thinks St. Louis is good, and they beat them, but only by 3 at home. It thinks Houston is really good, but Miami lost to them by 20 (albeit on the road). It thinks the Jets and Cards are average, and the Phins lost to both of them by 3, while Cincy and Oakland are bottom-feeders. Somehow that adds up to a slightly-above-average SRS.
Miami is getting +8.7 for beating Oakland by 19, +2.8 for beating STL, +1.3 for losing to ARI, +0.3 for beating CIN, and -5.9 and -6.3 for the Houston and Jets games. So the Raiders game is what’s really helping Miami, although that explains why they’re at +0.1 in the SRS.
Why they’re the 4th best team in the AFC is because the conference is gross.
Just taking a casual look at next weeks lines (having done least squares – unweighted for last weeks numbers) there are a few that I’m not getting (in terms of why are these teams getting points as it doesn’t make sense) and I can spot the SRS numbers above would back this up:
Indianapolis getting +3.5 (least squares has them winning by 2.7, the SRS would have them winning as well)
Atlanta +2.5 (?!?) – is this the Dome team playing outdoors issue (I expect you could run a regression on this with enough data – e.g. do dome teams under-perform outdoors?). Least squares and SRS both have them winning (by 7ish – biggest mismatch in relation to the line where there isn’t a potential health issue).
Seattle +3.0 – least squares and SRS both have them winning (but for some reason I don’t feel comfortable about this one)
Washington +5 – least squares and SRS both have them winning outright and I like this one – as well the AFC isn’t good right now
Oakland +2 – least squares and SRS both have them winning but this looks like a coin flip with two weak AFC teams.
Mathematically I would be keen to see how this goes as I just can’t see the logic right now (I might change my views on one or two once I have done the weighted numbers but these just don’t look right). It would be interesting to see if there is an obvious bias or prejudice against these teams.
George,
What is the Least Squares method?
It’s the basic Least Squares method Wayne Winston refers to in his book Mathletics (which has been referred to on here on a couple of occasions). Basically I do that system as I can’t get the SRS to work properly (I got the Massey system Neil referred to in a post a couple of weeks ago working today – haven’t factored in home advantage or capping in it yet and have slightly off numbers at the moment but I have an offense/defense and overall rating). The disadvantage of the Least Squares system is that the easiest way to run it is using Solver in Microsoft Excel but the variable limit within the Solver essentially means you can’t use it for College Football practically (limit of 200 variables).