Embedded Files
Daily Fantasy Sports
Daily Fantasy Sports
As a lifelong athlete/martial artists myself, I'm a big fan of professional sports. After my final undergrad semester in October 2022, I tried applying my quant skills to practice via daily fantasy sports ("DFS "and why statistics is used in it explained in footnote) while studying for the GMAT and applying to graduate programs.
I have tried trading securities before, both with real capital and simulated trading, but there's a massive difference in the poise it takes to succeed when there are real financial consequences. As Buffett has consistently said, composure is far more important to succeeding in investing than intelligence. The stress to do well in DFS is relatable to finance: the absolute stakes may be lower ($1-10 per entry), but the relative loss is much higher (100% the entry fee), so the ability to compound returns that much more difficult.
I exported my results onto a bankrolling tool to summarize the net yield from my experience, shown below. I hope it demonstrates to you my basic abilities to apply quantitative tools to financial risk to yield profitable returns.
DFS explained:
On sites such as Draftkings or Fanduel, there are competitions everyday for limited pools of users to draft a unique team of different players from various teams playing on that day. Players get "fantasy points" for various types of accomplishments from that game (i.e. 1 point per game point, 1.2 points per assist, etc). Each pool has a set "salary ceiling/cap" and a specific number of players for specific positions for a given sport the user must select. Players with better average performances have higher salaries. The challenge is to maximize the points a user's draft team earns that day against other users given the salary limitation.
How statistics and data analysis apply:
The salaries are usually determined by average performance of the athlete for the season. Some users will export external data to apply measures of variance and standard deviation to their calculations. Multivariate correlation is also useful when some players perform statistically better or worse against certain other teams or players, playing alongside certain other teammates, playing after "n" number days of rest, etc. A player's salary usually fluctuates day to day weighted by his/her/their most recent performance so users try to exploit miscalculations by the house to optimize their daily team with the highest sum of players' talents.
Page updated
Google Sites
Report abuse