Bales Goes Deep: It’s All Mathematics
Here’s a short list of things that interest me.
- Sports Analytics
- Theoretical Physics
-
Cooking at HomeGrubHub - Philosophy
- Making Up and Playing Games
- Rap Music
- Shark Tank
One thing I really like to do in my articles, and in life, is something I learned from James Alutcher that he calls “idea sex”—taking concepts from two completely separate areas that you know a lot about and combining them. It’s fun and I think it forces you to think about problems in creative ways.
That’s especially important for me in daily fantasy sports because I spend so much time on this stuff that I think it’s easy to fall into the same patterns of thinking. Idea sex snaps me out of that. Peter Thiel emphasized a similar idea in his book Zero to One—that many of today’s best ideas are reincarnations of past ways of thinking.
So let’s have some idea sex using one of my favorite songs from Dante Terrell Smith, aka Mos Def.
Editor's Note: 4for4 has an ALL-NEW DFS SUBSCRIPTION! It includes our revolutionary Lineup Generator, Stack Value Reports, Ceiling & Floor Projections, GPP-Specific Articles, Ownership Outlooks, DFS Theory Articles, Lineup Construction Articles, a Downloadable .csv with Projections and Salary Data, and much more! DETAILS HERE! Previous subscribers don't worry, you'll get the same DFS content you got last season, but we think you'll want to UPGRADE to our DFS subscription because it takes our DFS content to the next level!
I’m going to concentrate on the last few lines:
“Numbers is hard to feel and they never have feelings, but you push too hard, even numbers got limits. Why did one straw break the camel’s back? Here’s the secret: the million other straws underneath it. It’s all mathematics.”
What can Mos Def teach us about daily fantasy sports?
Numbers is hard to feel and they never have feelings.
Why do we use analytics and create models to help us make decisions? Because numbers aren’t emotional and they have no biases. We might interpret stats in ways that are more or less useful than others—we might use the “wrong” stats or draw conclusions that aren’t “real”—but the numbers themselves are always objective.
When I watch a football game, for example, I’ll probably be able to tell if a quarterback played relatively well or not. But my views are still likely to be skewed by various biases—the earliest and latest plays in the game are most likely to stick out in my mind, for example—and those are favoritisms that statistics can overcome.
Further, I can’t possibly capture and quantify every play in my mind. I can’t watch a game and know if a quarterback threw for 8.8 YPA or 7.4 YPA, for example—which is actually the difference between the top quarterback and the league-average for starters. This is why we use stats—to better quantify past events and predict future ones in an objective way.
But you push too hard, even numbers got limits.
I use models to help me predict sports outcomes, but I don’t just blindly trust what the models spit out. Every model has blind spots—gaps in what it is measuring or factors that just can’t be properly quantified.
This is especially problematic in football because the sport is so holistic and event-based. Maybe a running back’s YPC has been thrown off by two fluky long runs. Maybe his left tackle is pivotal to his success and just got injured. A model might not understand these factors, so there are limits to how hard you can push.
Why did one straw break the camel’s back? Here’s the secret: the million other straws underneath it.
There’s a decent amount of variance in daily fantasy sports, but only in the short-term. Over the long run, the luck evens out and skill dominates. A great player who wins a tournament had to experience some luck to come out victorious, but those who consistently finish near the top of the leaderboards are no luckier than a sharp entrepreneur who hits on his sixth business idea after the first five failed or the poker player who sucks out on the river but was getting the right odds to call.
Over a large enough sample size, there is no such thing as luck. That means that, if you’re a profitable player, much of your strategy should revolve around just staying in the game and maximizing opportunities to exploit your edge; the camel’s broken back is inevitable as you add more straws.
It’s all mathematics.
Stats are useful because they help us overcome biases and compute complex events. They have limits, but a lot of that has to do with our inability to adequately quantify everything that happens on a football field or basketball court. In theory, though, we should be able to do that as we develop more sophisticated techniques for collecting and processing data.
It really is all mathematics.
