The share of an opponent's possessions that end in a steal while a player or team is on the floor. A measure of active ball-hawking.
Raw steal totals reward big minutes and fast pace. Steal % takes both out and asks: of every 100 possessions the other team has, how many end with you taking the ball? It is a cleaner read on activity in the passing lanes, and it captures what defensive rating alone does not, the ability to create turnovers rather than just contest shots.
STL% = Steals ÷ Opponent Possessions × 100
STL%: Steal Percentage / STL: Steals / Poss: Possessions
Teams that gamble for steals get punished when the gamble fails, a defender out of position gives up a layup. Some high-steal teams have a surprisingly ordinary defensive rating for exactly that reason. Steals are high reward and high risk.
Guards and wings top steal lists because they live in the passing lanes. Centers, who guard the rim, will show lower rates whatever their quality. Never compare steal rates across positions without allowing for where each player stands.
Barcelona forced a steal on 9.8% of opponent possessions in 2025–26, the highest rate in the league, nearly one in ten. But steals and a great defense are not the same thing. Barcelona finished mid-table at 22-18, because chasing the ball is a gamble: when it works you get a run-out, and when it misses a defender is out of position and the rim is open. Steal % measures the reward. It says nothing about the risk.
Steal percentage estimates how often a player steals the ball per opponent possession while on the floor. It measures ball-hawking adjusted for playing time.
Divide a player's steals by the number of opponent possessions during his minutes, as a rate. That makes a starter and a bench player directly comparable.
Not on its own. Steals show activity and anticipation, but gambling for steals can also break a defense. Read steal percentage together with the team's defense when he is on the floor.