Playoff Probabilities: Week 17

With the final week of the regular season upon us, it's time to see where things stand for the playoff contenders. As always, these numbers come courtesy of Chris Cox at NFL-forecast.com and are generated with the help of his NFL-Forecast software app, which uses the win probabilities generated by the team efficiency model to simulate the NFL season thousands of times. And if you don't buy the game probabilities from Advanced NFL Stats, you can tweak them as much as you like to generate your own playoff projections. I encourage everyone to download the app and test out your own scenarios.

The AFC West & Wild Card

All sorts of advanced degrees are required to parse the melange of tiebreakers and scenarios that surround the final two playoff spots in the AFC. The Bengals, Jets, and Titans are all still eligible for the final wild card spot, in addition to the Raiders, who also could beat out the Broncos for the AFC West. There are plenty of places to get the full breakdown of the various clinching scenarios, but let's skip all the "strength-of-victory" and "if-then-and-unless" chatter and cut straight to the numbers:


Playoff Prob. With WinWith Loss
DEN 10047
CIN 10042
OAK 680
NYJ 290
TEN 240

The table above lists each team's probability of making the postseason—first column if that team wins on Sunday, second column if they lose. As you can see, the Broncos and Bengals both control their respective destinies (in the playoff-berth sense at least) and can clinch a spot with a win, while the Raiders, Jets, and Titans will need various degrees of help.

As far as if-then mumbo-jumbo goes, my personal favorite is the tale of the Tennessee Titans. To get the wild card spot, first Tennessee needs to win and have the Bengals lose. Simple enough. Then, if the Jets win, the Titans need at least one of the Broncos and Raiders to lose. But if the Jets lose, they need both of those teams to win. Got that? Good, let's move on.

High Leverage Games of the Week

Dallas at New York Giants | Sunday, January 1 | 8:20 pm ET

Playoff Prob. NYG WinDAL Win
DAL 0100
NYG 1000

If you didn't see this one coming, you haven't been paying attention. As anticipated, the battle for the NFC East has come down to the final week. Whoever wins this game wins the division; whoever loses, goes home. Simple as that. (And a tie goes to the Giants.) Despite their mediocre record, I wouldn't underestimate the Giants in the playoffs. If they can make it past Dallas on Sunday, they'll have a 13% probability to win the NFC Championship—the third highest probability in the conference behind only the Packers and Saints.

Baltimore at Cincinnati | Sunday, January 1 | 4:15 pm ET

Playoff Prob. CIN WinBAL Win
BAL 100100
CIN 10042

The other big game is yet again out of the AFC North, where Cincinnati will look to clinch a wild card berth, and, for only the second time this season, Bengals fans will actually get to watch their team play a home game on local TV.

For Baltimore, it's all about seeding. A win clinches a first-round bye (and gives them a 25% probability of claiming the #1 seed). A loss eliminates the possibility of the #1 seed and reduces the Ravens to the wild card if the Steelers win over the Browns. As we've seen, playoff seeding is no small matter. Overall, losing this game would cut the Ravens' estimated probability of a Super Bowl appearance by more than half, from 20% to just 8%.

News & Notes

  • Having gone on an ill-advised miniature winning streak, the Colts have reduced their probability of the number one pick to 74%. They'll close the season against a slowly disintegrating Jaguars team while the Rams will have to face a 49ers team fighting for a first-round bye. A Colts win and a Rams loss means the #1-pick would go to STL next year.
  • Our newly-crowned #1-ranked Pittsburgh Steelers, who we have as an 18% shot to win the Super Bowl, can't get no respect from the betting markets, which have placed their odds of winning the Super Bowl at 14/1. Presumably this reflects concern over Roethlisberger's ankle, though the Steelers showed last week they can compete without him (even if just against the Rams) and they have a 40% probability of securing a first-round bye and an extra week of recovery time for their quarterback.

The probabilities below are the result of simulating the season 50,000 times using the game-win probabilities from the team efficiency model. They may not add up to 100 (in percent form) due to rounding. Enjoy.


AFC EAST
TeamRecWins1st2nd3rd4th
NE12-312.7100000
NYJ8-78.4010000
BUF6-96.3005347
MIA5-105.6004753
AFC NORTH
TeamRecWins1st2nd3rd4th
BAL11-411.5613900
PIT11-411.8396100
CIN9-69.5001000
CLE4-114.2000100
AFC SOUTH
TeamRecWins1st2nd3rd4th
HOU10-510.8100000
TEN8-78.2010000
JAC4-114.6001000
IND2-132.4000100
AFC WEST
TeamRecWins1st2nd3rd4th
DEN8-78.6792100
OAK8-78.52133460
SD7-87.50463321
KC6-96.4002179
NFC EAST
TeamRecWins1st2nd3rd4th
NYG8-78.6648290
DAL8-78.43613500
PHI7-87.8079210
WAS5-105.2000100
NFC NORTH
TeamRecWins1st2nd3rd4th
GB14-114.6100000
DET10-510.4010000
CHI7-87.6001000
MIN3-123.4000100
NFC SOUTH
TeamRecWins1st2nd3rd4th
NO12-312.8100000
ATL9-69.8010000
CAR6-96.2001000
TB4-114.2000100
NFC WEST
TeamRecWins1st2nd3rd4th
SF12-312.6100000
ARI7-87.6061390
SEA7-87.4039610
STL2-132.4000100


AFC Percent Probability Playoff Seeding
Team1st2nd3rd4th5th6thTotal
NE78220000100
BAL124800390100
HOU00100000100
DEN000790079
PIT103000610100
CIN000007070
OAK0002101536
NYJ000001010
TEN0000044
JAC0000000
SD0000000
CLE0000000
MIA0000000
KC0000000
BUF0000000
IND0000000

NFC Percent Probability Playoff Seeding
Team1st2nd3rd4th5th6thTotal
GB10000000100
SF07030000100
NO03070000100
NYG000640064
ATL00005446100
DET00004654100
DAL000360036
CHI0000000
ARI0000000
MIN0000000
WAS0000000
STL0000000
TB0000000
CAR0000000
PHI0000000
SEA0000000


AFC Percent Probability to Advance
TeamDivision RoundConference GameSuper BowlSup Bowl Champion
PIT86522918
NE100603115
HOU81412314
BAL8536147
CIN14410
OAK11310
DEN21410
NYJ2000
TEN1000
BUF0000
CLE0000
IND0000
JAC0000
KC0000
MIA0000
SD0000

NFC Percent Probability to Advance
TeamDivision RoundConference GameSuper BowlSup Bowl Champion
GB100654220
NO77482412
SF8536134
NYG411773
DET391462
ATL361252
DAL21831
ARI0000
CAR0000
CHI0000
MIN0000
PHI0000
SEA0000
STL0000
TB0000
WAS0000


  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

6 Responses to “Playoff Probabilities: Week 17”

  1. Jonathan says:

    "Overall, losing this game would cut the Ravens' estimated probability of a Super Bowl appearance by more than half, from 20% to just 8%."

    And this assumes the Ravens would likely face a powerful Houston Texans squad as a 2 seed--when in reality, we know they're not all that good anymores. So the actual reward for winning that game is even higher. Baltimore's game is arguably more meaningful than even the NYG/DAL game.

    It is hard to overstate the importance of the battle for playoff positioning. The playoffs have already begun for BAL/PIT/NE/SF/NO.

  2. Josh Katz says:

    Jonathan--Very true... Although, if the Texans are bad enough that they lose in the first round, then Baltimore (as the #2-seed) would have to face the winner of AFC West vs. Steelers, which would probably be even tougher than how tough the model thinks the game against the Texans would be.

    But I'd agree that the result of BAL/CIN has an even greater impact on the playoffs than Giants-Cowboys, especially when you consider the effect it has on Pitt's chances as well.

    A CIN win gives Pitt a 38% prob. of appearing in the Super Bowl; a BAL win would cut this almost in half, to 22%. Really, apart from those already eliminated, every team in the AFC has something riding on the outcome of this game.

  3. Michael Beuoy says:

    I've really enjoyed these write ups this season. The high-leverage game of the week and the standings charts are fantastic tools for understanding the dynamics of playoff and seeding positioning. Hope this is back next year.

  4. Anonymous says:

    This is well done, fun reading; that said, I have trouble relying on some of the analytics with the knowledge gained through actual observation. The models do not, as far as I can tell, take into account injuries, match-ups, game changing penalties, prevent defenses, run the clock out offenses, etc. Most significantly, the probabilities key off of basically subjective predictions as to the outcome of a given game. Yes, the computer is not emotional and does not have an opinion but the metrics fed it are highly subjective. Of course the better teams predict to do better because, well, they're the better teams! Looked at your analysis of the importance of playoff seedings and noticed the number one seed does about twice as good as the number two seed. I think that is entirely attributable to the fact that the number one seed is the better team, as both get the first round bye and play similiar talent in the later rounds. What prompted this rebuke is your statements and numbers regarding the Giants and Texans. The QB injury in Houston should take their numbers down dramatically; prior stats are not meaningful with this cast. How can the Giants have a 2 out of 3 probability to beat Dallas, and then a probability better than Atlanta or Detroit to move on in the playoffs? Either a case of classic NY hype or statistics run amuck. My eyes tell me Giants were a blocked FG from losing 6 in a row and being out of the playoffs, and are only in a position to win because Dallas also stinks. Giants win against Jets was, to me at least, a classic match-up advantage as Giants strength (pass rush and manning)played into Jets weakness (offensive line and frazzeled QB). Giants are no better than 50-50 in this game with Dallas, and have proven they cannot beat quality teams at home or on the road. Does the model take into account that the Giants lost to Seattle and Washington at home?

    Anyway, love your work and keep it coming.

  5. Anonymous says:

    Anonymous:

    "The models do not...take into account injuries"

    This one statement from your comment is actually correct.

  6. Anonymous says:

    "I think that is entirely attributable to the fact that the number one seed is the better team, as both get the first round bye and play similiar talent in the later rounds"

    So in your opinion home field advantage doesn't exist?

    "The QB injury in Houston should take their numbers down dramatically"

    What? You understand this is a *computer* model right? Once you start adding a human element (like arbitrarily accounting for injuries), you start inching more and more towards a useless ESPN like intuition driven power rankings. You can make injury adjustments in your own head, that's not what the rankings are for.

Leave a Reply