Saturday, November 9, 2013

Playoff Implications - Week 10

Here is your weekly cheat sheet for NFL games with playoff implications. The purpose of this feature is to highlight games that have a big impact on the playoff picture (see last week's post for my first attempt). I can summarize the total impact a game is expected to have, as well as how it affects each particular team.

The main change from last week is that I am now filtering out results that aren't statistically significant. I am using the Fisher Exact Test for this purpose, with a significance level of 95%.

Ranking Week 10 Games by Leverage

The table below ranks the week 10 games by total leverage. Leverage in this context is a measure of both how uncertain a game's outcome is (games between evenly matched teams have higher leverage) and how much the playoff picture swings as a result of that outcome.

GameLeverage
PHI @ GB 40.4%
DET @ CHI 32.6%
CAR @ SF 26.9%
JAC @ TEN 19.9%
CIN @ BAL 19.0%
DAL @ NO 18.0%
MIA @ TB 17.4%
TEX @ ARZ 13.0%
DEN @ SD 12.0%
STL @ IND 5.7%
RAI @ NYG 3.9%
BUF @ PIT 2.8%
SEA @ ATL 0.6%

According to my modeling, the Eagles-Packers game is the clear top matchup, most likely because the outcome affects the NFC wild card race as well as two divisional races. One key callout here is that I am judging impact based solely on whether a team makes the playoffs. In other words, I ignore seeding implications. As we all know, seeding is important. For future weeks, I may simulate the playoffs as well to see which games have an impact on what we ultimately care about: Superbowl probability.

Beyond Week 10

I can also use my simulation dataset to look ahead to the remainder of the season and see which games (based on what we know now) will have significant playoff implications. The table below shows the top 10 week 11-17 games by total leverage. Think of this as a cheat sheet for whoever at NBC is in charge of flex scheduling.

WeekGameLeverage
17 PHI @ DAL 43.2%
15 GB @ DAL 39.9%
14 DET @ PHI 38.6%
16 CHI @ PHI 37.6%
12 CAR @ MIA 36.3%
14 DAL @ CHI 35.6%
17 GB @ CHI 33.1%
11 IND @ TEN 32.0%
13 GB @ DET 32.0%
15 NYJ @ CAR 31.7%

The NFC East and NFC North are wide open right now, so it's not too surprising that 3 of the top 4 games feature matchups between those two divisions. 
According to the table, Carolina-Miami has the highest leverage of the week 12 games, but there's virtually no chance that game is getting flexed as the default Sunday Night matchup that week is Broncos-Patriots. It's a low playoff leverage game, but potentially high leverage in the context of Superbowl probability, given its implications for seeding.

Week 10 Cheat Sheet

And here is your cheat sheet for week 10. You can filter by game or team below. As mentioned above, I have already weeded out any games that aren't statistically significant to a team's playoff chances.

Select Game: <-or-> Select Team:

home team:
gamegame probteamplayoff probloseswinsswingleverage
DET @ CHI 46.1% CHI 36.1% 21.5% 53.1% 31.5% 15.7%
DET @ CHI 46.1% DET 60.9% 74.8% 44.5% -30.3% 15.1%
CAR @ SF 70.9% CAR 46.9% 69.1% 37.9% -31.2% 14.2%
PHI @ GB 53.4% GB 55.7% 41.8% 67.9% 26.1% 13.0%
PHI @ GB 53.4% PHI 30.4% 42.4% 19.9% -22.5% 11.2%
JAC @ TEN 84.7% TEN 43.8% 21.5% 47.8% 26.2% 9.4%
MIA @ TB 41.3% MIA 26.5% 34.2% 15.7% -18.5% 9.1%
CIN @ BAL 44.9% BAL 10.2% 3.5% 18.5% 15.0% 7.4%
DAL @ NO 71.9% DAL 68.2% 79.4% 63.8% -15.6% 7.0%
CAR @ SF 70.9% SF 92.2% 81.9% 96.4% 14.5% 6.6%
CIN @ BAL 44.9% CIN 90.4% 96.2% 83.3% -12.9% 6.4%
DEN @ SD 26.7% SD 12.4% 8.6% 22.8% 14.2% 6.3%
PHI @ GB 53.4% DAL 68.2% 61.9% 73.7% 11.7% 5.9%
DAL @ NO 71.9% NO 92.7% 85.3% 95.5% 10.2% 4.6%
TEX @ ARZ 59.2% ARZ 9.5% 4.3% 13.0% 8.7% 4.3%
STL @ IND 80.2% IND 89.7% 82.2% 91.6% 9.4% 3.7%
DAL @ NO 71.9% PHI 30.4% 24.8% 32.6% 7.8% 3.5%
MIA @ TB 41.3% TEN 43.8% 41.4% 47.0% 5.6% 2.8%
TEX @ ARZ 59.2% CAR 46.9% 50.2% 44.7% -5.4% 2.7%
PHI @ GB 53.4% CHI 36.1% 38.9% 33.6% -5.3% 2.7%
PHI @ GB 53.4% DET 60.9% 63.4% 58.7% -4.7% 2.3%
JAC @ TEN 84.7% MIA 26.5% 31.7% 25.6% -6.1% 2.2%
PHI @ GB 53.4% CAR 46.9% 49.0% 45.1% -3.9% 2.0%
CAR @ SF 70.9% DET 60.9% 57.8% 62.1% 4.3% 2.0%
JAC @ TEN 84.7% SD 12.4% 16.8% 11.6% -5.1% 1.8%
TEX @ ARZ 59.2% TEX 2.3% 4.6% 0.8% -3.7% 1.8%
CAR @ SF 70.9% GB 55.7% 52.9% 56.9% 3.9% 1.8%
CAR @ SF 70.9% CHI 36.1% 33.5% 37.1% 3.7% 1.7%
JAC @ TEN 84.7% IND 89.7% 93.6% 89.0% -4.6% 1.7%
DEN @ SD 26.7% MIA 26.5% 27.5% 23.9% -3.6% 1.6%
DAL @ NO 71.9% CAR 46.9% 49.1% 46.1% -3.0% 1.4%
RAI @ NYG 72.8% NYG 3.2% 1.1% 4.0% 3.0% 1.3%
MIA @ TB 41.3% NYJ 14.9% 13.9% 16.5% 2.6% 1.3%
MIA @ TB 41.3% BAL 10.2% 9.2% 11.7% 2.5% 1.2%
CIN @ BAL 44.9% CAR 46.9% 45.9% 48.3% 2.4% 1.2%
JAC @ TEN 84.7% CAR 46.9% 44.1% 47.4% 3.3% 1.2%
BUF @ PIT 60.8% BUF 1.3% 2.8% 0.4% -2.3% 1.1%
TEX @ ARZ 59.2% CHI 36.1% 37.4% 35.1% -2.3% 1.1%
RAI @ NYG 72.8% PHI 30.4% 32.2% 29.7% -2.5% 1.1%
CIN @ BAL 44.9% NYJ 14.9% 15.9% 13.7% -2.2% 1.1%
JAC @ TEN 84.7% BAL 10.2% 12.8% 9.7% -3.1% 1.1%
DEN @ SD 26.7% TEN 43.8% 44.4% 41.9% -2.5% 1.1%
PHI @ GB 53.4% SF 92.2% 93.4% 91.2% -2.2% 1.1%
STL @ IND 80.2% MIA 26.5% 28.6% 26.0% -2.6% 1.0%
DET @ CHI 46.1% GB 55.7% 56.7% 54.6% -2.1% 1.0%
DAL @ NO 71.9% WAS 4.5% 2.9% 5.1% 2.2% 1.0%
CIN @ BAL 44.9% CHI 36.1% 36.9% 35.0% -2.0% 1.0%
JAC @ TEN 84.7% NYJ 14.9% 17.2% 14.5% -2.7% 1.0%
DEN @ SD 26.7% NYJ 14.9% 15.5% 13.3% -2.2% 1.0%
CIN @ BAL 44.9% CLE 11.0% 10.1% 12.0% 1.9% 1.0%
RAI @ NYG 72.8% NYJ 14.9% 13.4% 15.5% 2.1% 0.9%
DEN @ SD 26.7% BAL 10.2% 10.8% 8.7% -2.1% 0.9%
MIA @ TB 41.3% NE 96.0% 95.2% 97.1% 1.8% 0.9%
MIA @ TB 41.3% IND 89.7% 89.0% 90.8% 1.7% 0.9%
JAC @ TEN 84.7% CLE 11.0% 13.0% 10.6% -2.4% 0.9%
TEX @ ARZ 59.2% CIN 90.4% 91.5% 89.7% -1.7% 0.8%
PHI @ GB 53.4% NYG 3.2% 2.4% 4.0% 1.6% 0.8%
BUF @ PIT 60.8% PIT 1.4% 0.4% 2.0% 1.7% 0.8%
DET @ CHI 46.1% NO 92.7% 91.9% 93.5% 1.5% 0.8%
DEN @ SD 26.7% IND 89.7% 90.2% 88.5% -1.7% 0.7%
CAR @ SF 70.9% ARZ 9.5% 8.3% 10.0% 1.6% 0.7%
PHI @ GB 53.4% WAS 4.5% 3.7% 5.2% 1.5% 0.7%
TEX @ ARZ 59.2% BAL 10.2% 9.4% 10.8% 1.5% 0.7%
MIA @ TB 41.3% SD 12.4% 11.8% 13.2% 1.4% 0.7%
PHI @ GB 53.4% NO 92.7% 93.4% 92.0% -1.3% 0.7%
BUF @ PIT 60.8% SF 92.2% 93.0% 91.7% -1.4% 0.7%
MIA @ TB 41.3% CIN 90.4% 89.9% 91.2% 1.3% 0.6%
JAC @ TEN 84.7% NE 96.0% 97.4% 95.7% -1.7% 0.6%
STL @ IND 80.2% TEX 2.3% 3.5% 2.0% -1.5% 0.6%
TEX @ ARZ 59.2% SF 92.2% 92.9% 91.7% -1.2% 0.6%
CIN @ BAL 44.9% WAS 4.5% 5.0% 3.8% -1.2% 0.6%
DAL @ NO 71.9% NYG 3.2% 2.4% 3.6% 1.2% 0.5%
TEX @ ARZ 59.2% NE 96.0% 96.6% 95.6% -1.0% 0.5%
RAI @ NYG 72.8% RAI 0.5% 1.3% 0.2% -1.1% 0.5%
SEA @ ATL 30.2% SEA 99.6% 99.9% 98.9% -0.9% 0.4%
DEN @ SD 26.7% DEN 99.6% 99.9% 99.0% -0.9% 0.4%
CIN @ BAL 44.9% PIT 1.4% 1.1% 1.8% 0.7% 0.4%
STL @ IND 80.2% PIT 1.4% 2.0% 1.2% -0.7% 0.3%
TEX @ ARZ 59.2% PIT 1.4% 1.1% 1.6% 0.5% 0.3%
SEA @ ATL 30.2% ATL 0.2% 0.0% 0.4% 0.4% 0.2%
TEX @ ARZ 59.2% SEA 99.6% 99.8% 99.5% -0.3% 0.1%
BUF @ PIT 60.8% ATL 0.2% 0.3% 0.0% -0.3% 0.1%
JAC @ TEN 84.7% STL 0.0% 0.1% 0.0% -0.1% 0.0%

Other Takes

Advanced NFL Stats publishes playoff projections each week, using their efficiency model and Chris Cox's NFL Forecast Tool.

This past week, I also stumbled upon a very cool site that in many ways is a far more robust version of what I'm attempting to do in this post. Check out Sports Club Stats playoff projections (warning: for the stat-minded, that site can be a time vampire, threat level: tvtropes.org).

No comments:

Post a Comment