More

    The Art of Dynasty | Chapter 10 (Bayesian Inference)

    Dynasty fantasy football is much like war. You need to know your opponent, understand how to prepare and create a battle plan, how to construct a balanced roster with a defined vision, and know when to go for the title or (re)build for a future run. A deep and intimate understanding of both strategy and football is required to be consistently successful in a dynasty league. This treatise illustrates the fine art of building a true dynasty and maintain a top position in your leagues for a considerable time.

    These are the 10 core tenets I believe are critical to success in building a dynasty behemoth that will yield results year over year.

    1. Startup Drafts
    2. Roster Construction
    3. Ideal Archetypes
    4. Ship Chasing
    5. Iron Bank
    6. Rookie Pick Values
    7. Rookie Drafts
    8. Trading
    9. Roster Management
    10. Bayesian Inference

    Take lock and stubbornness are a death sentence in dynasty. Bayesian Inference is all about changing your opinions and thoughts around a player given new information.

    As fellow team member @bpoFSU puts it, Bayesian Inference is all about processing as much data as you can and not being stuck in “take lock”. Brian notes that you should process new information, see how it compares to what we already know, and use that to create a new opinion. The only wrong take is one that is not open to change based on new information.

    The NFL landscape is constantly changing and so are the outlooks of players in terms of fantasy football. This is why many smart fantasy gamers will tell you to “buy the dip” on players who are expected to bounce back. As I mentioned in Chapter 8, you should look to trade for players who you expect to return to form. A player who has a proven track record over his career should be bet on to get back on track. But applying Bayesian principles to fantasy football goes both ways. As you notice a downward trend on certain players, consider moving off of them before it’s too late.

    Avoid Anchoring

    With that said, Bayesian Inference also tells us that we need to constantly be changing the way we view players who have “burned” us in the past. Players who started off their careers in a lackluster fashion, but eventually break out should be viewed without bias. Anchoring is the idea that we use pre-existing data as a reference point for all subsequent data, which can skew our decision-making processes. This causes us to rely too heavily on past information. By utilizing Bayesian principles in our fantasy processes, we can avoid falling prey to anchoring bias.

    A prime example is Ryan Tannehill whose career started off in disappointing fashion (under Adam Gase). But as he switched teams and was afforded another opportunity, he has broken out and performed at a different level. Now Tannehill is in an extremely efficient offense surrounded by playmakers like AJ Brown and Julio Jones. If you were stubborn and refused to change your perspective then you missed out on an opportunity to add a solid QB to your roster at a heavily discounted cost.

    Another example to consider is Josh Allen who had a career year in 2020. Before last season, Allen’s career numbers included a completion percentage of 56% with 30 total passing touchdowns. However, with the addition of Stefon Diggs, Allen took a major leap forward in his development and posted a 69.2% completion percentage and 37 passing touchdowns. This could always be an outlier, but Bayesian Inference encourages us to consider making adjustments on how we view the fourth year QB. If we dig into his numbers from 2018 to 2019, Allen improved in every statistical category in terms of accuracy, efficiency, and production. From his rookie season to second year, Allen’s completion percentage increased over 6%. Allen also threw for more touchdowns, less interceptions, and posted higher yards (and air yards) per attempt. Then came the massive leap after being paired with an elite talent at wide receiver. The Bills franchise QB has improved every year so far so his track record is indicating a player who is continuing to develop and become a better player. Using Bayesian principles, we should be adjusting our perceptions of Allen if we aren’t believers of his talent.

    Be flexible, read nuances, identify trends, and change your perspective and adjust.

    Become a Patron!

    Stay in the Loop

    Get the daily email from CryptoNews that makes reading the news actually enjoyable. Join our mailing list to stay in the loop to stay informed, for free.

    Latest stories

    - Advertisement - spot_img

    You might also like...