Welcome back, friends.
Remember when I started this project? The Reader’s Digest version is, using multiple methodologies honed in other disciplines (including crime analysis, geographic information systems, animal movement, transportation, and social networks), I argue that a player’s shot selection is patterned and predictable.
And although it’s been a while, the initial results were promising: when examining a specific player in a specific game, indeed shots are predictable at reasonably (useful) accurate and precise levels. And they more we predict, the better we get.
Unfortunately, ShotCaller started with Carmelo Anthony, who unsurprisingly has decided to shut it down for the remainder of the season. Despite this, the project keeps it movin’…enter Damian Lillard.
In the downtime, there’s been a bit of data enrichment, to include measures for schedule, opponent, fatigue, and shot preferences. While this is not the final draft, this iteration hopefully further increases the predictive accuracy and precision.
The result? Here’s the prediction for 1st quarter shots for Dame for Wednesday night’s Blazers vs. Spurs game:
Why only the 1st quarter? It’s the purest of all the quarters (huh?), meaning that each subsequent quarter is directly related to the ebb and flow of the game; this makes predicting quarters 2-4 before the game begins highly volatile (and potentially stupid). Expect attempts into those quarters before the season is over.
For now, while you enjoy the potential playoff matchup Wednesday night, expect Damian Lillard to hoist four shots from those red areas to start the game.
Data and photo support provided courtesy of NBA.com, Basketball-Reference.com, and data extraordinaire Darryl Blackport.