Wednesday, December 14, 2011

Betting Market Power Rankings – Week 15

by Michael Beuoy

Here are the week 15 betting market power rankings. As noted last week (link), I have modified the methodology to now incorporate the point spreads for the upcoming week, if they’re available (week 16, in this case).

Methodology
The goal I had was to generate a set of point-based rankings that would best predict how the betting market would set the point spread for the coming week’s matchups. The better I was able to match the point spread, the better the rankings were a reflection of the market’s estimate of team by team strength. Through trial and error experimentation, I found that using point spreads for the most recent five weeks (with higher weighting given to more recent weeks), combined with an adjustment that accounted for actual game outcomes, generated the best predictive accuracy. More detail here

The Rankings
Here are the Betting Market Power Rankings for Week 15:

Glossary:
LSTWK - The betting market rank as of the prior week (using the same methodology).
GPF - Stands for Generic Points Favored. It’s what you would expect a team to be favored by against a league average opponent at a neutral site.
GWP - Stands for Generic Win Probability. I converted the GPF into a generic win probability using the following formula: GWP = 1/(1+exp(-GPF/7)). This gives a more direct comparison to the ANS rankings.
ANS RNK - The Advanced NFL Stats Team Efficiency rankings for the same week.
ANS GWP - The Advanced NFL Stats Generic Win Probability for the same week.












































RANKTeamLSTWKGPFGWPANS RNKANS GWP
1 GB1.010.00.8130.73
2 NE2.08.00.7640.67
3 NO3.07.50.7450.65
4 BAL5.05.50.6970.64
5 PIT4.05.50.6920.76
6 SF6.04.50.65170.50
7 ATL7.03.50.62110.55
8 NYJ9.03.00.61120.54
9 DAL8.03.00.6080.62
10 DET12.02.00.58100.61
11 SD14.02.00.58140.52
12 PHI16.02.00.5790.61
13 NYG10.01.50.5660.64
14 TEX13.01.50.5510.77
15 DEN20.00.00.50240.39
16 TEN17.0-0.50.49180.48
17 CIN15.0-0.50.48130.52
18 MIA11.0-0.50.48200.45
19 CHI21.0-0.50.47150.51
20 SEA23.0-1.00.46250.38
21 RAI18.0-1.50.44160.50
22 ARZ22.0-2.00.43220.43
23 CAR19.0-2.50.41230.40
24 WAS26.0-2.50.41210.44
25 MIN25.0-3.00.40280.33
26 BUF24.0-3.50.37190.48
27 CLE29.0-5.50.31270.34
28 JAC30.0-6.00.30260.36
29 TB28.0-6.00.30300.30
30 KC27.0-6.50.29310.28
31 STL31.0-8.50.23290.32
32 IND32.0-9.00.21320.26



• Obligatory Tebow-Watch – Denver moved up 5 spots to number 15, and a GWP of 0.50.
• The ANS model still disagrees pretty significantly with the Betting Market on San Francisco.
• I’m surprised the Giants are ranked as low as they are.


Prediction of This Week’s Point Spreads

See below for how well the ranking methodology predicted this week’s point spreads:




























GamePred LineActLineDiff
 BAL @ SD -0.5-2.0-1.5
 CAR @ TEX 6.06.50.5
 CIN @ STL -5.5-6.0-0.5
 CLE @ ARZ 5.06.51.5
 DAL @ TB -5.5-6.5-1.0
 DET @ RAI -1.5-1.00.5
 GB @ KC -13.5-14.0-0.5
 JAC @ ATL 12.011.5-0.5
 MIA @ BUF -1.51.02.5
 NE @ DEN -5.0-6.0-1.0
 NO @ MIN -8.5-7.01.5
 NYJ @ PHI 0.53.02.5
 PIT @ SF 0.02.52.5
 SEA @ CHI 2.54.01.5
 TEN @ IND -6.0-6.5-0.5
 WAS @ NYG 7.07.00.0

4 comments:

Jeff Fogle said...

Those look to be right on the money MB for capturing how the market is ranking and rating teams. Any nitpicks would be very minor, and could easily be wrong anyway. Solid.

Only tweak I'd recommend again would be considering creating two versions for each team...one with their starting QB and one with the backup. That would yield 64 "teams" instead of 32..but would probably smooth out any sort of short term inconsistencies when there's a transition from a Cutler to a Hanie or a Schaub to a Yates. Your methodology cuts to the chase pretty quickly anyway...but, to the degree some tinkering might improve things a smidge, that would be one to consider for the future I'd think.

Hope you'll have a chance to continue posting these this year and in the future. And, seriously, something like this would be a great addition to the NBA field as well. Might be something to approach a place like Basketball Prospectus with, or even True Hoop with Henry Abbott. Did some stuff for HoopData last year...but that site tends to focus on in-game or player analytics rather than market material. Let me know if you're interested in pursuing an NBA version once the NFL has wrapped up and I can help ask around.

Michael Beuoy said...

Thanks Jeff. I think I'll try this out on the NBA once the season gets underway. I've been playing around with College Football, but the results were a bit erratic. I was thinking there wasn't enough interconnected-ness between the teams. That shouldn't be a problem with the NBA.

TBD said...

looking forward to this weeks. differences of 2.5 points or so are rather large (especially in games with only a small favorite) but it looks like a couple of these are from hard charging teams like Philly and SD, so your rankings may lag a bit in these cases

Anonymous said...

This is great stuff, Michael. I look at this every week before I make a play.

Post a Comment

Note: Only a member of this blog may post a comment.