Can basketball be reduced to a numerical formula? Even KenPom has his doubts

[jbcarol]

Del Harris, a longtime NBA coach, said that analytics are a poor tool for evaluating high school or college players

There are too few games played to get a good sense of a player through numbers, he said. Plus, the players can play against wildly different levels of competition.

Harris also said basketball is not the best sport to try to reduce to a mathematical formula.

"Stat numbers in football and baseball are better indicators because it's a stop-start game in both those sports," Harris said. "Basketball is a fluid dynamic-type game where players are never in the same position at any time during the game. That is you'd never have 10 players in the same geometric position twice in the same game. It's like a snowflake..."

Pomeroy: "That's one thing about people who dive into basketball stats. You sort of have to have that understanding, or you're occasionally going to sound like a fool."

[ShadowHawg]

Exactly right. It isuch a dynamic sport that it generates way too much information to accurately analyze.

Player performance is affected by many factors that aren't accounted for. Like would Stockton have been nearly as productive without Malone and vis versa? Where is the analytic for a relationship on the court where teammates are so familiar with each other they know what the other will do in a given situation?

[ThundrHawg]

Can basketball be reduced to a numerical formula?

Absolutely. The mathematical model might not be there yet, but if we can predict the movements of subatomic particles by using math, basketball can absolutely be predicted with the correct model. You have to realize, gravity was a mystical force up until 350 years ago. With today's models, no. But I'm sure at some point we will be able to much more accurately predict outcomes of basketball games and even individual performances.