The Limits of Player Rankings

A ranking can calculate consistently without becoming an objective definition of greatness.

Every model chooses what matters

Selecting points, rebounds, assists, championships, or awards is already a judgment. Assigning weights adds another. Even a formula built entirely from official statistics reflects decisions about time period, minimum games, pace, role, and whether team results belong in an individual comparison.

Available statistics are incomplete

Traditional box scores understate positioning, rotations, screen quality, communication, gravity, matchup versatility, and off-ball decision-making. Modern tracking data fills some gaps but is unavailable across most of NBA history and can still require interpretation.

Context changes outcomes

Coaching, teammates, injuries, opponents, salary rules, expansion, travel, and league strategy influence both production and winning. Awards add knowledgeable human judgment, but voting standards and positional rules change.

Transparency beats false certainty

A useful model publishes its inputs, allows disagreement, corrects errors, and describes limitations. Facts Over Feelings lets visitors move weighting sliders because reasonable definitions of greatness differ. Its “Weighted Result” should be read as the output of those choices, not proof that discussion is over.

What rankings are good for

They organize evidence, expose inconsistent arguments, generate questions, and make comparisons repeatable. A fan who values longevity can explain that preference; another can emphasize playoff peaks. The goal is not to eliminate judgment but to turn hidden assumptions into visible ones.