Player Efficiency Rating (PER): What It Measures and Where It Falls Short

What is PER?

Player Efficiency Rating (PER) was developed by basketball statistician John Hollinger and popularised in the early 2000s. It attempts to summarise a player's complete statistical contribution in a single number, normalised to a league average of 15. A PER of 20 is very good; a PER of 25 is elite; a PER of 30 or above is all-time historical territory.

What PER measures

PER combines positive contributions (points, rebounds, assists, steals, blocks) with negative ones (turnovers, missed shots) using a formula that weights each event by an estimated value. It also adjusts for pace — ensuring that a player on a fast-paced team is not unfairly rewarded compared to one on a slow team — and for playing time.

PER's genuine value

For high-level screening of player productivity, PER is useful. It clearly distinguishes elite players (30 PER) from average starters (15 PER) from fringe roster players (8-10 PER). For quickly assessing whether a player is performing at a high level in raw statistical terms, it is a reasonable starting point.

Where PER falls short

PER has significant and well-documented weaknesses that the analytics community has largely moved past:

• Defence is poorly captured — blocks and steals are the only defensive inputs. The entire quality of a player's defensive positioning, communication, and help rotations is invisible in PER.
• No context for shot quality — PER does not distinguish between a player who creates open threes for teammates and one who forces contested mid-range shots at the same frequency.
• Team context ignored — a player on a great team benefits from better spacing, better screening, and more open shots. PER does not control for team quality.
• Heavily skewed toward high-volume players — since PER is based on counting stats, players who take more shots, make more passes, and appear more in the box score naturally accumulate higher PER even at the same efficiency level.

What to use instead

Modern basketball analytics has largely replaced PER with impact-based metrics — Real Plus-Minus (RPM), BPM (Box Plus/Minus), RAPTOR, and LEBRON — that attempt to measure a player's total impact on team performance rather than just their individual statistical contributions. These are not perfect either, but they capture defensive impact, lineup context, and opponent quality in ways PER cannot.