Best Ball, Basketball and Other Thoughts

Some Introductory Thoughts for a Hot Best Ball Summer

Hello folks! Thank you for allowing me a short post-rookie draft sabbatical and staying with me. The reality is the content drop immediately post-NFL draft was probably in line with about 6 weeks worth of regularly scheduled content on this newsletter, but with rookie drafted being concentrated in such a short time span I wanted to expedite everything I could for you.

Unfortunately, every year rookie draft content leaves me completely exhausted by the end and I need a couple weeks to recharge the creative battery.

This marks our newsletter’s official transition to Best Ball SZN. Same Newsletter, Same style, new game. (For the most part - I will make sure to churn out periodic dynasty content along the way)

---

For those who follow me on the Elon App, you probably know that this is the time of year I tweet 4X more about my assorted anxieties regarding the Boston Celtics than fantasy football advice.

Perhaps in an effort to justify my stream of consciousness fandom tweets, I want to start my intro column to Best Ball SZN on Thinking About Thinking with some comparisons between the way we analyze basketball and the way we discuss best ball.

It’s my belief that a concerningly high percentage of analysis in either fantasy sports or sports generally is simply assigning causal reasoning to highly variant events. This is most prevalent when discussing small sample events.

Perhaps no sport is as plagued by this type of analysis as basketball. Every seven-game series is painted to be a referendum on the viability of a coach, player, team or even its “will to win.” However, the uncomfortable reality is that a primary factor in said referendum is simply whether shots fall. It’s un-interesting and unsatisfying to spend a morning talk show simply reading off a team’s three-point shooting percentage vs. their seasonal average, and therefore we often wind up assigning credit and blame for the result of a series to a mix of vague platitudes and legitimate - if over-weighted - micro-strategies deployed by each team.

To be clear, I am not suggesting that every shot has an equal chance of going in based on where it is taken. However, the long-run effects of shot quality are substantially more marginal than game-to-game swings in shooting variance. For example, taking pull-up 3-point shots instead of catch-and-shoot attempts, ‘above-the-break’ vs. ‘corner 3s,’ or contested vs. open 3s each reduce the expected points of the shot by between 10-15-percent.

Defenses contesting shot attempts and forcing shooters out of favourable positions, and offenses demanding a higher quality shot diet through ball movement and penetration absolutely effect the expected outcome of a game. However, this effect is only partially responsible for the outcome of a game. We often over-credit successful teams for the quality of shots taken (or lack their of) and under-credit teams making or missing a higher than expected percentage of the shots taken.

Rather than identifying systemic, predictive takeaways, we wind up “describing the variance” i.e. pointing out the reasons why a particular outcome may have taken place without adding significant predictive value moving forward.

From a basketball analysis perspective, this error - such that it exists - isn’t particularly harmful. However, if coaches alter their game plans, or front offices alter their roster in reaction to variant outcomes which don’t predict future ones, it can have extremely harmful effects.

Describing the Variance in Best Ball Tournaments

This is where the best ball connection comes in. Spike Week founder Erik Beimfohr constantly stresses the minimal samples nature of the game, describing the entirety of our data sample as equivalent to three weeks of DFS results. Despite the wealth of content available on tournament best ball, all of it is primarily drawing conclusions from three slates.

Therefore, when drawing conclusions from any data relating to the three seasons since Underdog launched Best Ball Mania we need to be extremely cautious about whether our takeaways have predictive value moving forward or whether they are just describing the variance.

A perfect example of this is the dreaded ZeroRB discourse. Hayden Winks has posted the below charts each of the past two years titled “When it was Optimal to Draft RBs.”

As you will see below, in 2021 teams that drafted zero RBs through two rounds were at a deep disadvantage to teams drafted one or two backs to open their teams. However, in 2022, ZeroRB teams remained optimal through round 6.

The reasons for this discrepancy are largely identifiable but have limited application moving forward. In 2022 the elite WR scorers were largely isolated to the first few rounds while several mid-round RBs had apex outcomes; Josh Jacobs, Rhamondre Stevenson, Tony Pollard and Miles Sanders. In 2021 conversely there were several early-round WR busts (A.J. Brown, Calvin Ridley, Deandre Hopkins) and multiple explosions from the mid-rounds (Ja’Marr Chase, Cooper Kupp).

We largely know the conditions that will bring about a great or terrible ZeroRB year but it is very difficult to predict whether those will manifest in a given season. What charts like Winks’ can help with is give us an idea of HOW to build successful teams of several different constructions in the years those teams hit. But even so, the conclusions we would draw remain highly subject to the small-sample that created them.

For example, in 2022 ZeroRB performed exceptionally overall but tanked if you deferred RB until Round 7 or later. Does this tell us that Rounds 7-8 are the point of diminishing returns for ZeroRB even in its most favourable year? That seems plausible. Pushing beyond six rounds sans running back would mean drafting at least one player in your first seven picks who cannot enter your starting lineup in a given week prior to your first running back. Having drafted at least four WRs in the first six rounds, you really ought to be drafting under the assumption that most if not all your early receiver selections are paying off consistently. This is one of the reasons pairing ZeroRB builds with early QB and TE selections has been consistently strong.

However, what if this chart is merely showing us Josh Jacobs’ ADP? (He was largely drafted in Round 6 or 7 last year)

Beyond any strategic min/maxing, the most important feature of successful ZeroRB teams in 2022 was drafting Josh Jacobs, and/or a combination of Pollard, Sanders and Stevenson, all of whom were largely selected in Rounds 6-8. If Jacobs’ ADP was in round 4, or if Pollard’s was in round 10, the chart would likely have displayed very different results.

Finding a way to draw meaningful conclusions from the data available, while recognizing the inherit limits in its predictive value is crucial to determining your best ball strategy in 2023 and beyond. But articulating that balance is much easier said than done.

Luckily, we play in a market-based game. We don’t have to be “correct” to be “right.” When dealing with uncertainty in any market, we don’t know the correct answer; rather we can merely combine data and intuition to wager which answers are more likely to be correct in a majority of scenarios. However, we could easily be wrong in two ways: (1) we are actually incorrect about the most likely outcome; (2) we are correct about the most likely outcome, but an unlikely outcome happens anyhow

Our ability to determine the most optimal approach to drafting a particular construction or player is only as important as the market allows it to be. If the market begins to over-estimate the likelihood of a certain scenario, we are best served to bet against it. A great example I’ve discussed before in this news letter is the self fulfilling prophecy of the RB dead zone.

Once an indicator of an inefficiency in a naive market, the RBDZ morphed into a prescription which changed how we draft, and thus reduced its own universality of application. I’ve compared this phenomenon before to the Bicameral mind theory of consciousness in the TV-series Westworld.

We - as a fantasy market - often point out trends flowing from systemic misevaluations, and apply the trend as a new rule rather than critically examining its root causes. This can result in players reducing or rising in cost for abstract or inapplicable reasons, which can open up major value in betting against an overly certain market.

It is impossible for me or anyone to tell you with certainty how to balance all the available data available with general intuition and the market, however we will certainly try our best to apply a critical eye and an open-mind to the best ball economy in the year 2023 to find our edges in the ever-evolving market.

Our Hot Best Ball Summer

This is just a short introduction to Thinking About Thinking’s transition to primarily best ball tournament content for the duration of the summer. If you’ve been with me since launch, expect to see a similar mix and style to what we’ve been doing for dynasty.

I will mix bigger picture, macro-strategy pieces with micro analysis on players or strategies I find especially intriguing.

Every 2-4 weeks I will be authoring a recurring piece called “We Just Meta” in which I will detail cover current market inefficiencies at the macro (position, construction) and micro (player) level I perceive in drafts and how I’m approaching it for the time being.

Interspersed between those I will be publishing some deep dives into particular strategies for roster construction, and/or some players and teams I am targeting as key stands or fades.

The last two years at PlayerProfiler I have written extensively on holistic fundamental of best ball tournament strategy - much of which I would still stand by in 2023. My hope is to dive more into the weeds this year by actioning several of these principles into specific strategic levers and player stands.

You can check out my three-part best ball strategy guide from 2022 here. My work from the 2021 season is linked throughout.

All the while, I will continue with some periodic dynasty content as well. On that note I have exciting news that I will be dropping Dynasty Ranks on this site starting in June as a joint project with Pat Kerrane of Legendary Upside and Davis Mattek of AutoMattek Absolutes and SportsGrid. So keep your eyes peeled to your inbox Dynasty heads!

Talk soon!