Predicting a player’s future shooting performance is a difficult task. Analysts spend hours poring over box score stats like Effective Field Goal%, trying to find evidence to build their predictions around. In the last decade, Shot Making stats have come onto the scene. They work differently from shooting percentages by looking at how a player performs relative to their shot quality, instead of raw makes and misses. Recently, Krishna Narsu, our head of R&D at Basketball Index, looked into which of these two stats (Effective Field Goal% and Shot Making) is better at predicting future results. 

 

Here are quick explainers of how they work:

Effective Field Goal% – This correctly weighs three pointers as 1.5 more valuable than two pointers. eFG% = (FGM + 0.5 * 3PM) / FGA

Shot Making – It looks at how well a player performs relative to their shot quality. It creates an expected Effective Field Goal% based on whether it is a pull-up or C&S attempt, where the nearest defender is, and how much time is left on the shot clock. Then it takes your actual eFG% and subtracts the expected eFG%

 

Effective FG% has long been used as a proxy for shooting skill. However, the issue with eFG% is that you have to make some kind of adjustment for a player’s shot quality. A play finishing big is going to have a higher eFG% than a post-up player self-creating their own looks. This is where Shot Making stats have an edge. Because it looks at how a player performs relative to their shot quality, it has built-in context for a player’s performance. The year-to-year correlation for Effective Field Goal% is 0.28, while Overall Shot Making is 0.66. This makes Shot Making a far better metric for predicting future shooting performance, due to its strong year-to-year consistency.

When a player’s role or team situation changes, their shot quality can vary vastly. A player going from the worst team in the league to playing alongside Jokic or Luka is likely to see their shot quality sharply rise. Higher shot quality means higher expected eFG%, and in turn, it’s likely a player’s raw eFG% will rise. If they again change teams to another bottom-feeder with poor playmakers, they would likely see their shot quality drop, along with their eFG%. A player’s Shot Making changes less dramatically in these situations because it looks at performance relative to circumstance. 

Imagine two role players with identical physical traits and age. They both averaged 8 points a game last season and will both be stepping into larger offensive roles next year. Player A has high shot making, but an average eFG%, and player B has a high eFG%, but average shot making. Player A’s shooting will likely scale better because they are less reliant on shot quality as the key to their efficiency. 

The point of this article isn’t to tell you shooting percentages are bad or useless (they’re not), but to explain why analytics has an edge in predicting future results. Understanding shot quality and the relationship it has with eFG% is the key to understanding the two stats.

Thank you to Krishna Narsu for doing all the math. Follow me on Twitter for more @taylormetrics