Week 6 NFL Daily Fantasy Football Results - Some Successes, But Mostly Losses (-$46.00)

It is hard to put an optimistic spin on repeated losses, but I still am learning a lot.

I set out at the beginning of this year to apply mathematical modeling to the process of picking the best possible players in fantasy football each week. I anticipated some success, some losses, and a fun ride along the way.

What I am learning, and learning very quickly, is that this process is far beyond the scope of mathematical modeling and statistical analysis. It is far beyond my "win some, lose some" mentality. And, it is not for the feint of heart. Each week I am toiling over important factors and algorithmic optimization. However, I have so much to learn still about optimal lineup structuring and game theory when it comes to DFS.

While the money I am wagering in this experiment is not life changing, and is just about my risk tolerance for something so speculative, I quite sure I would be equally as embarrassed by my losses even if I was wagering more. Point blank, my current reality is that I am a fish in the Draft Kings world, and am currently donating my money to all comers!

This past week I lost $46 on $74 of entries for a 62% loss on the week. This puts me overall down about $236, or 39% for the year.

My Best Lineup - 132.16 Points

All negative thoughts aside, I am hitting on some of my lineup structures. This lineup above was from a GPP that included the concept the Chiefs would tear up a historically terrible Raiders defense, with lots of love to spread around to guys like Chris Conley for a relatively cheap salary. Many concepts hit here, but many did not. Hard not to be disappointed by Antonio Brown and what I thought was a sneaky Lions DST play.

Needless to say, 132.16 is never going to hit anything in any iteration of any game on Draft Kings. I am embarrassed it was my best. However, from this, I have learned I need to trust my instincts on higher priced guys and pay closer attention on game day to conditions that might effect lineups. I did not admittedly see the weather report out of Oakland on Sunday that lent it self more to a ground attack.

I am also playing way too much chalk and need to focus on guys that are not so overwhelmingly owned, even if they are huge algorithm hits. These mistakes are causing me to bleed money.

My Worst Lineup - 81.36 Points

Every one of my yearlong fantasy league lineups scored higher than this. I was back and forth about whether to even share this one publicly, but in the efforts of being completely transparent, here it is in all its glory. This was for a H2H matchup that I obviously lost. I could chalk this up to lots of bad breaks, but I had no business putting a high variance guy like Richard in a H2H.

Across all lineups, I was heavy on Delanie Walker, but I didn't work my way through the fact Joe Haden was out on Sunday for the Browns. The Titans receivers were thus sucking up all the receptions and leaving my lineups missing critical points. I am learning fast to pay more attention to details like this in the future.

Inside the Data

The majority of my projections exist between a 0.9 and 4.8 point overestimate. Below are examples of players that fell with in each range of differences between predicted and actual performances.

My bell curve this week is just about where it was last week and the week before. I am projecting about 68% of my fantasy scores on the interval [-8.78, 7.43], where a negative means an underestimate, positive means overestimate. I would like to squeeze this range down a lot more in the weeks to come.

Player by Player Results - Week 6

As I always tell my students, the greatest successes I have ever experienced in my life have usually followed a pretty long string of mistakes. While I do not know how long this current string will last, I do know I am learning quite a lot and am drawing quite a bit on the inspiration from the DFS community here. I hope to have a big bounce back week next week and chip away at some of my losses!

---