Next-level Game Theory in Fantasy Football
Thanksgiving is right around the corner, and I know we are all awaiting the inevitable question from our relatives: “Do you have a job yet?” Now, there are many ways to get around this question (which may be the topic of another blog) but more and more lowly millennials are turning to Daily Fantasy Sports (DFS) websites like DraftKings for a “sustainable” income. I can imagine the awkward silence of your grandmother gives you when you tell them you manage a fantasy football team. While I don’t encourage gambling, there are ways for intelligent people to optimize their line-ups in order to maximize payout (or minimize your losses… if you’re a pessimist), and that’s game theory.
When you step back, it’s easy to see the application of game theory to DFS. There are players that have huge expected payoffs, like Aaron Rodgers. However, in this game, you’re goal isn’t just to maximize your payoff, but to maximize your payout compared to the rest of the “field”. So, if you choose Aaron Rodgers as your quarterback, you probably are guaranteed a solid amount of points, but you gain close to nothing on the field. This payout matrix encourages risk-taking. Although players do not know the true proportion of the field that is starting a given player, they can make an intelligent guess. Another variable in all of this is that each DFS player has a team budget. Players that are consistently good, like Rodgers, will cost more than high-risk players like Cam Newton. Therefore, along with maximizing your potential points-gained against the field, you have to maximize the value-per-dollar for each player you select. That being said, there is no “right” strategy for playing DFS. However, if you want to win any money, the system rewards intelligent high-risk moves. It involves researching under-valued players with juicy match-ups, and big names that are have potential to bust. Overall, this network behavior is almost a complement to what we talked about in class. Likely, there exists some threshold in which if the proportion of the field selects a given player is greater than the threshold, it is beneficial for you to abandon that player and select someone else. With some research and proper mathematical modeling, one could get a huge advantage on the field.
As I stated, I cannot encourage gambling, especially on a Cornell-facilitated blog… But if you are going to do it, make sure you bring your game theory skills to the pitch.
Article link: http://rotoviz.com/2015/07/next-level-game-theory-fantasy-football/