bscully27 Posted October 20, 2015 Share Posted October 20, 2015 To make the best model, I want to make sure I am posing the right question to be solved. So what are a few of the ways to describe the daily fantasy contest problem? Should I simply predict points with more accuracy, predict the best lineup, something else? I have NFL data from 2009 and already beat DK's projections 67% of the time but this doesn't guarantee me success. Ideally I'd like to have all past contests with winning scores to know if a team I set is good enough to make money. Since I don't have that data, I'm wondering what others have done. What have you guys done or have heard done? Quote Link to comment Share on other sites More sharing options...
Shorttynaz Posted October 20, 2015 Share Posted October 20, 2015 It's simple - pick the players who will score the most points. Problem solved.. Quote Link to comment Share on other sites More sharing options...
BA Baracus Posted October 20, 2015 Share Posted October 20, 2015 (edited) I think that by taking a purely analytical approach, you are entering into a nuclear war armed with a BB gun (unless you have access to some supercomputers and are a data scientist by profession). I think that a large number of DFS entries are generated by computer models and the guys making the models do it professionally. I have a monster spreadsheet that weighs various projections (including my own, which are based heavily on individual matchup analysis) and I update it weekly to hopefully improve its accuracy. Doing all of this, I am able to get all of my millionaire maker entries to cash, but I have never been higher than a 4th (from the bottom) tier payout. I am starting to come to the conclusion that either a) I suck at this and it will never be worth my time or b ) the deck is hopelessly stacked against me and it will never be worth my time. That said, good luck! Edited October 20, 2015 by BA Baracus Quote Link to comment Share on other sites More sharing options...
bscully27 Posted October 20, 2015 Author Share Posted October 20, 2015 I am a data scientist and my current projection model is all coded in Python. Takes 2 minutes to download and run everything. I'm just wondering what is the best relative benchmark to compare my team to. For example, what if I take a lineup optimizer find the best team given a site's salary and projected points. Then I use my projected points and come up with optimal team 2. If team 2 is projected to outscore by 20%, I could be comfortable. Does that make sense? Also, there are different strategies for head-to-head vs large contest. Large contest you need players with the highest upside. Head to head, consistency is more important. Quote Link to comment Share on other sites More sharing options...
Shorttynaz Posted October 20, 2015 Share Posted October 20, 2015 Bye Felicia.. Quote Link to comment Share on other sites More sharing options...
Sunday Couch Potatoe Posted October 20, 2015 Share Posted October 20, 2015 Quote Link to comment Share on other sites More sharing options...
keggerz Posted October 20, 2015 Share Posted October 20, 2015 I am a data scientist and my current projection model is all coded in Python. Nice! Quote Link to comment Share on other sites More sharing options...
stevegrab Posted October 20, 2015 Share Posted October 20, 2015 I was told there'd be no math 1 Quote Link to comment Share on other sites More sharing options...
BA Baracus Posted October 21, 2015 Share Posted October 21, 2015 I was told there'd be no math It's the universal language Quote Link to comment Share on other sites More sharing options...
Macksimus Posted October 21, 2015 Share Posted October 21, 2015 Go back to fantasy baseball, NERD!!! J/K ...let us know when you hit it big in DFS. Quote Link to comment Share on other sites More sharing options...
doobwaa Posted October 21, 2015 Share Posted October 21, 2015 7. Quote Link to comment Share on other sites More sharing options...
squirrelmastr21 Posted October 22, 2015 Share Posted October 22, 2015 I don't play DFS. But if I did, I would look at the lineups that have won previous weeks and try to work backwards. What were the indicators that correlated to those players. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.