# Approximating Curves II: Assimilation of the Jump Shot Process

If you were to ask one-hundred shooting coaches “What’s the most important aspect to making a jump shot?” you will probably get at least fifty different responses. Answers may range from detailed such as the finger mechanics of the release or “shooting axis,” to broad, holistic responses such as “Find your repeatable comfort zone at…

# Approximating Curves I: Mechanical Process

Now that the 2019-2020 season has ended, let’s take a quick look at something almost every data scientist knows: polynomial projection. Now, if you’re a data scientist and find yourself mumbling, “I’ve never heard of that,” don’t worry: You have. Over the next few posts, we are going to discuss a larger problem of approximating…

# Analytic Breakdown: 1963 Finals Game 6

On April 24th, 1963, the Boston Celtics took the floor at the Los Angeles Sports Arena holding onto a 3-2 advantage in the NBA Finals against the Los Angeles Lakers. For the only time in the history of the league, a Finals team trotted out only  (later to be named) Hall of Fame players: Bob…

# Extending Possessions: Geometric Distribution

In a recent post, we took a brief glimpse at offensive rebounding and discussed some pro’s and con’s about crashing for rebounds and provided an illustration about where rebounds go after missed attempts. In one such instance, you would have seen this plot: For the sake of argument, let’s suppose that every single red circle…

# Story Underneath Usage: Incompleteness

In the era of possession-based statistics, we often look at items such as per-possession or per-100 possessions. This type of parsing makes sense as a possession is deemed to be the period of time a team “controls” the basketball. The technical definition of a possession defines the end of the “control” period as the point of…

# Manifold Nonparametrics: Which Way Do Passers Pass?

As a player traverses across the court, they break down and process every event that they see: the assertion of defensive players, the alignment of their teammates, and current state of the game. The ability to perceive all three components simultaneously is what I call court vision. It is one of the primary instigators that…

# Exercising Error: Quantifying Statistical Tests Under RAPM (Part IV)

In the Regularized Adjusted Plus-Minus (RAPM) model, one of the perceived challenges is understanding the error associated with the resulting posterior RAPM value a player receives. In a previous post, we noted that RAPM is a Bayesian model in which we assume that “player contribution” can be estimated through weighted offensive ratings conditioned on the…

# Considering Consistency of an Analytic

Warning: This is a lecture from one of previous statistics courses taught over the years. It will be theoretically heavy, but offers insight on some of the research process required for developing analytics. (End Disclaimer)   Thought Exercise: Perimeter Defense Whenever we develop an analytic to help describe the game, we typically have to ask three things.…

# Transitioning Turnovers: Case Study of Golden State and Toronto

In Dean Oliver’s Four Factors, we are interested in effective field goal percentage, offensive rebounding percentage, free-throw rate, and turnover percentage. If a team cannot dominate a couple of these categories, then it will be unlikely for that team to win. For instance, let’s consider effective field goal percentage. The Golden State Warriors have posted a .558 eFG% while limiting their…

# The Randomness of Ratings

Fresh off a Boston Celtics sweep of the Indiana Pacers, across Twitter phrases of “Remember that the Boston starters were -3.5 Net Rating for the series” piped up. The implication was that the Boston starters should have lost the series, and the most common rumblings came about the depth of Indiana’s bench costing them the…