Saturday, November 12, 2016

Guess the Lines - Week 11 (plus Week 10 results)

The results are in. Who was right, the pundits or the models?

Last week, I launched a weekly feature that pits my Vegas NFL rankings against the instincts of Bill Simmons and Cousin Sal, as featured in The Ringer's "Guess the Lines" podcast. The podcast aired Monday, and I had results ready to go Tuesday, but other events interceded that night.

As a reminder, I am using my Vegas rankings for the NFL to predict the point spreads for future games. The rankings use point spreads from this season's games to reverse engineer an implied power ranking.

Opening lines tend to come out a week and a half in advance, and my rankings automatically update when they do. In order to make the test properly out of sample, I need to make my predictions almost a week in advance of The Ringer's Monday podcast. The predictions below are based on my rankings as November 1. The aGPF and hGPF columns are the away and home "Generic Points Favored" numbers from my rankings. The predicted point spread is just the difference between the two numbers plus a 2.5 point adjustment for home field advantage.

I added one additional prediction to the results table below. ESPN Chalk publishes their own Vegas-based rankings on a weekly basis. Using the rankings posted November 1 (paywall), I added a column for the predicted point spread according to ESPN Chalk's handicappers.

Here are the results. Predictions in bold signify those that were closest to the actual point spread. Note that for the "actual", I am using the point spread as reported by Cousin Sal on the podcast.

Week 10 Resultspoint spread
gameaGPFhGPFmeespnbillsalactual
CLE @ BAL -7.2 -0.3 -9.5 -8.5 -8.0 -10.0 -10.0
CHI @ TB -4.1 -2.5 -4.0 -4.5 -3.0 -4.5 -1.5
SF @ ARZ -6.6 3.1 -12.0 -13.5 -8.0 -8.0 -13.0
MIA @ SD -2.3 0.8 -5.5 -4.0 -4.0 -3.0 -3.5
ATL @ PHI 2.7 1.1 -1.0 -1.0 +3.0 +1.5 -1.0
HOU @ JAC -1.9 -4.0 -0.5 +2.0 +3.5 +4.0 +1.0
LA @ NYJ -2.9 -3.0 -2.5 -4.0 -3.0 -4.0 -2.5
KC @ CAR 2.7 2.8 -2.5 -2.0 -1.0 -1.0 -3.0
SEA @ NE 4.3 8.8 -7.0 -7.0 -7.0 -6.0 -7.0
MIN @ WAS 2.4 -1.1 +1.0 +3.0 -3.0 -3.0 -2.5
DEN @ NO 3.5 -0.5 +1.5 +4.5 +3.0 +2.0 -1.0
DAL @ PIT 2.5 1.6 -1.5 -1.0 +1.5 -3.0 -2.5
GB @ TEN 2.3 -2.2 +2.0 +3.5 +3.0 +3.0 +2.5
CIN @ NYG 1.5 0.8 -2.0 0.0 -3.0 -3.0 -2.5
Mean Absolute Error 1.29 1.86 1.89 1.68

To judge accuracy, I am using Mean Absolute Error, which is just the average of the absolute value of the miss for each game. By that metric, the rankings published here were the most accurate in predicting the week 10 lines, with a mean absolute error of 1.29 points. Cousin Sal comes in second at 1.68 points, and Bill and ESPN are clustered together at 1.89 and 1.86 points respectively.

However, if you just wanted to count winners and losers, Bill Simmons did the best, having come closest in 7 games (including ties). Myself, ESPN, and Cousin Sal tie for second with 5 games.

Week 11 Predictions

Here are the week 11 predictions for both my model, as well as the November 8 rankings from ESPN Chalk. As mentioned above, opening lines for Week 11 are already out, and my rankings have adjusted accordingly. The predictions below, however, are from the rankings archived on November 8 (you'll just have to trust me).

Week 11 Predictionspoint spread
gameaGPFhGPFmeespnbillsalactual
NO @ CAR 1.3 2.8 -4.0 -5.0
BUF @ CIN 1.5 1.3 -2.5 -4.0
PIT @ CLE 3.2 -8.3 +9.0 +8.0
BAL @ DAL -0.8 3.5 -7.0 -7.0
JAC @ DET -4.8 -1.5 -6.0 -4.5
TEN @ IND -2.9 -2.3 -3.0 -3.5
TB @ KC -3.7 2.5 -8.5 -7.5
ARZ @ MIN 3.0 0.7 0.0 -3.0
CHI @ NYG -4.4 1.0 -8.0 -5.5
MIA @ LA -1.5 -2.8 -1.0 +0.5
NE @ SF 9.2 -8.0 +14.5 +14.0
PHI @ SEA 1.4 4.2 -5.5 -5.5
GB @ WAS 2.0 0.3 -1.0 +1.5
HOU @ OAK -1.4 1.2 -2.5 -2.0

For most games, my rankings and ESPN are within a point of each other. The biggest disagreement is on the Minnesota-Arizona game, where I have it as a pickem, but ESPN would project Minnesota as a 3 point favorite at home. The early line has Minnesota as a 1 point favorite.

No comments:

Post a Comment