|
|
DFS Statistical Model Up and Running |
|
1
|
Gadget   United States. Feb 25 2014 18:27. Posts 295 | | |
So I've been working on a statistical model for projecting fantasy points and I finally have it working to give me some basic numbers. I have a lot of stuff I still plan on adding, but for now it is using all the info I've been using by hand for the past few months. I don't want to give out all the specifics of it, but it takes the fantasy points per minute of each player and multiplies that by a number of factors based on the matchup that day. I manually input the number of minutes above/below the players season average of minutes I think they will play that day. This gives me the ability to account for injuries on a daily basis. The best part is it takes my ~2 hour research process daily down to about 15 minutes, giving me time to look at things a bit more in depth. Here are my top picks at each position for today with my projection:
PG:
Lowry(8300): 41
Lillard(7900): 36
Brooks(3600): 23
SG:
Harden(9200): 43
Stephenson(7600): 35
Green(5900): 27
SF:
Ariza(6600): 32
Batum(6200): 29
Parsons(7200): 34
PF:
Booker(3500): 25
Robinson(4300): 29
Love(11600): 51
C:
Noah(9100): 41
Howard(9300): 41
Hibbert(5400): 29
I haven't gotten a Full Tilt email yet. Has anyone else not received an email that is expecting one?
|
|
| 1
|
jdmk2114   United States. Feb 26 2014 08:16. Posts 25 | | |
How many hours did it take you to build the model?
Whatever it is, I'd say that's an efficient use of your time. |
|
| 1
|
Gadget   United States. Feb 26 2014 18:13. Posts 295 | | |
It took me and a friend about 20 hours of work. |
|
| |
|
|
Poker Streams | |
|