Playoff Projections - Week 8

Just like past years, all of the numbers below come from Chris Cox at NFL-forecast.com. His app uses the win probabilities from the ANS team efficiency model to run a Monte Carlo simulation of the remaining NFL games thousands of times. Based on current records, our estimates of team strength, and knowledge of the NFL's tie breaking procedures we can come up with some pretty interesting predictions of how each team will fare come the end of the season. If you want to use a different model or just fiddle with the numbers by hand, go ahead and download the app yourself.

What is the deal with these numbers?

This is our first look at playoff projections this year so it seems like a good time to talk about how good these predictions really are. Like all models, this one has limitations. Here are some of the issues that will come up.
The only inputs into the ANS model are box scores and play-by-play information from this season’s games. The model does not know how good your coach is at choosing when to challenge a play or how well your QB runs a 2 minute offense. And not all statistical categories that might have some impact are considered, e.g. special teams performance.

For simplicity it is assumed that the underlying team strength is constant throughout the year but obscured by random variation. The model doesn’t know that you benched your RB or that your DE lost a limb to MRSA.

The estimates of team strength are based on the limited data set that we have so far. These estimates will get more accurate as the season progresses. In the meantime, these numbers may be over- or underconfident for any given team or in any given game. Having said all that, the model does a pretty good job. We should take its outputs as a good first pass at the problem to be modified by common sense on occasion.

AFC playoff teams already becoming clear, less so in the NFC

Even with only 8 weeks in the books, we have a very good idea who will be making the playoffs in the AFC. Cincinnati and Denver are almost guaranteed spots (99+%) and Indianapolis, Kansas City, and New England are very likely to get in. 5 or 6 different teams have at lost some chance of securing that last slot. Cincinnati just about has the North locked up and Indianapolis is a 90% favorite in the South.

Where the top of the AFC has distinguished itself, the bottom of the NFC is most obvious. The Falcons, Rams, Vikings, and Bucs can give in to the despair they are feeling, because each has less than a 0.1% chance of playing in January according to the simulations. Just about every other NFC team can hold out hope of getting in.

High leverage game of the week: CHI @ GB

The 187th Bears-Packers matchup is of major importance to both teams. A Chicago loss would make it nearly impossible for them to win the division and a 4-4 record would give them only a 3.5% chance of sneaking in as a wildcard. A Chicago victory would make it much more likely that the Bears or Lions would steal the division from Green Bay (a jump from 28% likely with a GB win to 55%). Both teams face about a 20% swing in playoff probability depending on the outcome of this game.

Some random observations

Despite their better record, the Chiefs are the underdog in the AFC West according to our model. Denver has performed much better in the categories that contribute to the team efficiency rankings (especially offensive pass numbers) and is thus the 70% favorite in the division. It is even more extreme in the postseason, where Denver is given a 21% chance of winning it all compared to KC’s 3%.

The Bengals and Broncos both have about a 40% chance of making the Super Bowl. Indianapolis and Kansas City are the only other AFC teams with more than a 2% chance. There seems to be much greater parity at the top of the NFC.

There is a greater than 70% chance that Seattle and/or San Francisco will make it to the NFC championship game. Other than the AFC West, no other division comes close to this level of certainty of having a representative in the conference championship game. For comparison, the NFC East has less than a 15% chance.

In the NFC East, you pretty much have to win the division or stay home because none of them has more than a 3% chance of grabbing a wildcard slot.

The Numbers

AFC EAST
Team	1st	2nd	3rd	4th
NE	76	19	5	1
NYJ	16	47	25	12
BUF	6	22	45	28
MIA	3	12	26	59
AFC NORTH
Team	1st	2nd	3rd	4th
CIN	99	1	0	0
CLE	1	58	28	12
PIT	0	28	39	33
BAL	0	13	33	54
AFC SOUTH
Team	1st	2nd	3rd	4th
IND	91	7	1	0
TEN	5	46	47	2
HOU	4	46	47	3
JAC	0	1	4	95
AFC WEST
Team	1st	2nd	3rd	4th
DEN	70	28	2	0
KC	29	65	6	0
SD	1	7	81	12
OAK	0	0	12	88
NFC EAST
Team	1st	2nd	3rd	4th
DAL	66	23	9	1
PHI	27	47	21	5
NYG	5	22	45	28
WAS	2	8	24	66
NFC NORTH
Team	1st	2nd	3rd	4th
GB	66	29	6	0
DET	31	58	11	0
CHI	4	14	81	1
MIN	0	0	1	99
NFC SOUTH
Team	1st	2nd	3rd	4th
NO	84	16	0	0
CAR	16	81	3	0
ATL	0	3	90	7
TB	0	0	7	93
NFC WEST
Team	1st	2nd	3rd	4th
SEA	75	25	0	0
SF	25	73	2	0
ARI	0	2	86	12
STL	0	0	12	88

AFC Percent Probability Playoff Seeding
Team	1st	2nd	3rd	4th	5th	6th	Total
DEN	41	24	4	1	28	2	100
CIN	37	47	13	2	0	1	100
KC	17	10	2	0	64	6	98
IND	4	14	44	29	0	2	93
NE	1	4	27	45	1	6	83
SD	0	0	0	0	6	22	28
NYJ	0	0	4	12	1	15	31
TEN	0	0	2	3	0	9	14
CLE	0	0	0	0	0	15	17
HOU	0	0	1	3	0	6	10
MIA	0	0	1	2	0	4	7
BUF	0	0	1	4	0	5	12
PIT	0	0	0	0	0	5	5
BAL	0	0	0	0	0	2	2
OAK	0	0	0	0	0	2	2
JAC	0	0	0	0	0	0	0
NFC Percent Probability Playoff Seeding
Team	1st	2nd	3rd	4th	5th	6th	Total
SEA	57	14	4	0	22	2	99
NO	19	38	23	4	3	7	93
SF	18	6	1	0	59	11	94
GB	5	23	33	5	3	15	83
DET	1	10	18	2	7	28	66
CAR	0	6	8	2	5	20	41
DAL	0	2	7	57	0	1	67
PHI	0	1	3	23	0	3	30
CHI	0	0	3	1	0	5	9
ARI	0	0	0	0	1	8	10
NYG	0	0	0	5	0	0	6
WAS	0	0	0	2	0	0	2
ATL	0	0	0	0	0	0	0
STL	0	0	0	0	0	0	0
MIN	0	0	0	0	0	0	0
TB	0	0	0	0	0	0	0