NFL Season Predictions as Percentages

My usual power rankings include a guess at how many games each team will win, and as such I can make some guesses as to what the playoffs will look like.  But, especially this far in advance, I have neither the time nor disposition to figure out all the tie-breakers and the like.  So I turned to to help me out on that front.  I used their java program and the predictions from my Luigi model to play out the rest of the season, much like Advanced NFL Stats does.  So consider this an update to the power rankings, and it allows for a little comparison between Luigi and Brian’s model.

First a quick disclaimer: because the java script uses a slider bar for game probabilities as opposed to directly inputting numbers, there are likely some tiny differences between my values and what actually ended up in the prediction.  But they should be small.  Also, because it uses a slider bar, I will not be doing this again for a while; there are simply too many games.  But if I remember I’ll try again at the end of the season when scenarios are more concrete and single games a bit more obviously important.

Looking at expected wins, they match up pretty well with what I have in my usual table (as they should).  A couple of potentially interesting notes though.  I give the Packers about a 3.5 or 4% chance of going undefeated this year, which is a bit better than Football Outsiders (insider required).  The 49ers are the only other team with even a measurable chance of going 15-1 (obviously, since they’re the only team left with 1 loss) at maybe 2%.  At the other end, Indianapolis and Miami are the only teams with a chance to go 0-16 (obviously, since they’re the only winless teams left), with the Colts ‘ahead’ on that count 2.5% to 1%.  So it’s pretty unlikely that we’ll see a repeat of the perfect Patriots season or the ‘perfect’ Lions season.

Moving to the playoff picture, I pinned the NFC division winners as the Packers, Giants, Saints, and 49ers.  Using the forecast program, I can tell you the probabilities for that are 91%, 51%, 79%, and 99.6%.  Even though the Packers are better than the 49ers, they have less of a lock on the division because they have to worry about the Lions whereas no one in the NFC West has a shot at San Francisco.  If you look at total playoff probability though, the Packers and 49ers are tied at 99.8%; even if they somehow fail to win the division, both are very likely to get a wild card.  The Giants aren’t so fortunate, only having a 64% chance at a playoff spot overall.  Despite having little chance of winning their division, the Lions look good for a wild card spot with a total playoff probability of 77%.  The other wild card is likely to be Philadelphia at 50%.  Other contenders include Atlanta (38%), Chicago (36%), Dallas (19%), and Tampa Bay (11%).  Everyone else has less than a 10% chance of making the playoffs.  All of that agrees pretty well with what I approximated from the estimated wins.

Moving to the AFC, I had division winners of the Pats/Bills, Steelers, Texans, and Chargers with wild card teams Bills/Pats and Bengals/Ravens.  With the probability forecasting it looks like the Bills have a slight edge over the Patriots, 54% to 44% for the division title.  Both are pretty likely to make the postseason, at 85 and 83%.  The Steelers are indeed the favorite in the AFC North but you would do better to take the field; I give them a 44.5% chance of winning the division.  They have a 80% chance of making the playoffs overall.  The Texans are looking pretty good at 94% for the playoffs.  The Chargers are actually below 50%, at 48.5% for the playoffs.  It’s fairly unlikely that two teams are going to make it out of the AFC West, so San Diego needs to bank on winning the division (46%).  The other playoff contenders are the Bengals (65%), Ravens (58%), Raiders (34%), Chiefs (24%), and Titans (17%).  Again, this all matches up with my win-based story.  The loser of Bills-Pats has a bit of a better hold on things than I would have guessed, so the Bengals/Ravens better hope they’re as good as I think they are.

Looking at the Advanced NFL Stats predictions, I appear to be more bullish on the Bills and less on the Steelers, with both having about 10% differences in their playoff probabilities.  I’m also much higher on the Bengals and less so on the Ravens.  I’m much lower on the Jets, but they don’t have great odds either way (21% for Brian, 8.5% for me).  In the NFC our scores are pretty similar, with me giving Atlanta more credit and Dallas less.  Overall, though, the broad strokes look about the same.

About the only other thing to look at is the prediction for the playoffs.  There are only two teams with more than a 10% chance to win the Super Bowl right now: the Packers and the Texans.  For the Packers the reason is obvious; they’re probably the best team in the league.  For the Texans it’s a little less clear.  I think they’re that high mostly because they have the best chance of simply making the playoffs in the AFC.  It might also be because for the playoffs it assumes Brian’s rankings, and his are pretty high on the Texans.  The next favorites in the AFC are Pittsburgh, Buffalo, and New England and no one else is above 5%.  In the NFC it’s New Orleans and San Francisco.

The most likely NFC projections for the playoffs are Packers and 49ers bye; Giants (4) versus Lions (5); and Saints (3) versus Eagles (6) (the Bears are actually the most likely 6 seed, but the Eagles are more likely to make the playoffs).  If that’s the case, Giants-Lions would be a toss-up with the Lions being a bit better but the Giants having home field and the Saints would be favored against either the Eagles or Bears, with Philly being the tougher match-up.  The Packers would be favored throughout.

In the AFC Buffalo and Pittsburgh are small favorites for the byes, leaving Houston (3) versus Cincinnati (6) and San Diego (4) versus New England (5) (New England has the highest likelihood of either the 5 or 6 seed, but is more likely to get the 5).  Houston would be favored, and I think the Patriots would as well despite being on the road.  I currently have Pittsburgh as the weakest of the four remaining teams, but they would have home field advantage over everyone except Buffalo.

There’s still plenty of time for the playoff picture to get clearer and more cloudy.  Last year I thought things were looking somewhat straightforward around this time and then a week later they were crazy again.  Games to watch this weekend include basically any game involving a team from the NFC East, AFC West, or AFC North as well as New England and Buffalo.  For those reasons I would look for Oakland-Denver (the Raiders need to not throw 6 interceptions again, and they’re good favorites to do so), Pats-Giants (a Giants win gives them as well as Buffalo an inside track), and Pittsburgh-Baltimore (a Ravens win would make things very interesting).  Game probabilities will be up Sunday morning as usual, or earlier if Bodog has enough numbers up on Saturday.



This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s