Gut vs Numbers: How To Win $1 Million by Drafting A Winning DraftKings team
It’s no secret that I’m a data-driven guy—someone who uses math and science as a foundation for all of my fantasy football decisions. 1-week fantasy football is arguably even more mathematical than season-long, and a nice foundation of (simple) analytical thinking can really help take you over the edge to become a profitable player (and even get a shot at $1 million!)
If there’s one sport in which the numbers aren’t always totally indicative of what’s going on, however, it’s football. Whereas a sport like baseball is very standardized—it’s always batter versus pitcher in more or less the same situations—football is not. You have all different sorts of game situations, offensive systems, personnel groupings, play-calling tendencies, and so on. The sport doesn’t set itself up to be analyzed as analytically as baseball or basketball, which allows for the use of your “gut” more so than in other sports.
Nonetheless, I still think data should be the backbone of your weekly fantasy football decision-making. Here’s why.
Testability and Evolution
The main reason that we can and should place a lot of importance in certain numbers is because a mathematical, scientific approach to fantasy football allows us to make testable (and falsifiable) predictions. We can test our hunches, figure out what’s right and what isn’t, and ultimately improve as fantasy owners.
In that way, “fantasy football Sabermetrics” is evolutionary because it’s self-correcting. It can improve. It can evolve. How do we get better with our gut decisions, though? How can we make better instinctive decisions? We can’t really—that process doesn’t really evolve—unless we add analytics to the mix.
Let me give you an example. There’s pretty strong evidence to believe that playing a tight end in the flex on DraftKings isn’t a wise move in head-to-head leagues. Historically, lineups with a tight end in the flex have performed really poorly.
This data is the result of a scientific, falsifiable approach to weekly fantasy football. It could very well be the case that you’ve played tight ends in the flex and found success in your cash games. Maybe you had a quality performance or two that clouded your memory such that you now play a tight end in the flex in all of your head-to-head matchups.
The only way to know if those gut feelings are accurate is to make testable predictions. When it comes down to it, selecting players based on your gut feelings might work some of the time, but it isn’t highly useful unless we add data. Without the numbers, we can’t test our beliefs or start the evolutionary process toward becoming a smarter, more profitable player.
That doesn’t at all mean that subjective decision-making is useless in weekly fantasy football. There are lots of times when it makes sense to go against the numbers and make a decision based on perhaps non-quantifiable factors.
However, we should always strive to have numbers and hard data as the foundation of our decision-making. Again, the reason is because it allows us to evolve as weekly fantasy football players. We can use a scientific approach to improve in every facet of our games—player evaluation, projections, lineup selection, bankroll management—in a way that making choices based on hunches doesn’t allow.
It’s important to note that, even with a data-driven approach to weekly fantasy football, you’ll still need to eventually use subjective insights to make the best decisions. If you blindly follow the data, you’ll be forcing yourself into a situation in which you need an unprecedented level of accuracy to be correct.
Take weekly player projections, for example. Many weekly fantasy football players project players each week and then determine player values by comparing those projections to the player salaries. That’s fine, but the values can change pretty drastically with small changes in the player projections. To trust the values 100 percent, the projections would need to be basically flawless. That’s too stringent of a requirement for anyone.
The value in projecting players is that it creates a nice base with which to work. You can immediately identify which players are good values, creating a pool of potential players from which to choose. Then, you can use more subjective factors to determine which of those options is best for you; you’re using a foundation of analytics to improve your subjective decision-making.
Actually, the process of doing research and testing theories is valuable in and of itself because it enhances your ability to think critically and make smart decisions, even if subjective. In that way, we should be using analytics not as a standalone tool for weekly fantasy football decision, but as a means to improve the objective framework through which we should try to make all of our less objective decisions.
