NBA Blocks and Steals Props: Why These Markets Have the Highest Win Rates

Blocks at 69.9%, Steals at 61.9%: the Win Rates Bookmakers Struggle to Price

If I told you there was a prop market where the under side hit at nearly 70% over a full NBA season, you would assume the bookmaker had priced it accordingly and there was no edge left. You would be wrong. Blocks and steals props remain the most consistently mispriced markets in NBA player prop betting, and the data from the 2025-26 season confirms what sharps have known for years: the under on defensive stats hits at rates that no bookmaker model has fully accounted for.

Blocks under props hit at 69.9% in backtested data from the 2025-26 season. Steals under props hit at 61.9%. Compare those figures with points props at 55.7% or PRA combos at 54.7%, and the gap is striking. These are not marginal differences — blocks unders are nearly 15 percentage points more likely to hit than points props on the equivalent side. The question is why, and the answer lies in the statistical properties of low-volume events that the bookmaker’s models handle poorly.

Why Low-Volume Stats Are Harder for Bookmakers to Model

Points are a high-volume stat. A scorer who averages 24 points per game takes 18-20 shots, goes to the free throw line 5-6 times, and accumulates his total through dozens of individual events. The law of large numbers keeps the outcome relatively close to the mean on most nights, and the bookmaker’s model, which is built around that mean, is therefore accurate.

Blocks are a low-volume stat. A centre who averages 2.0 blocks per game might block 0, 1, 2, 3, or 4 shots in any given game. Each block is a discrete, binary event — either the shot is blocked or it is not — and the total for the game is the sum of a small number of these events. When the number of events is small, the variance relative to the mean is enormous. A player can go from 0 blocks to 4 blocks based on a handful of moments, and the distribution around the mean is not the smooth bell curve that the bookmaker’s model assumes.

The specific problem is that blocks and steals follow something closer to a Poisson distribution than a normal distribution. In a Poisson distribution, the median is below the mean when the average is a small number. A player averaging 2.0 blocks per game has a median closer to 1.7 or 1.8, because the distribution is right-skewed — most games he gets 0-2 blocks, with occasional spikes to 4 or 5. When the bookmaker sets the line at or near the mean (say, 1.5 or 2.5), the under hits more often than the 50% the line implies, because the median is below the line.

This is the same mechanism that drives the mispricing patterns across all prop categories, but it is most pronounced on blocks and steals because the volume is lowest and the skewness is greatest. The bookmaker knows about this effect in theory but has not fully corrected for it in practice, partly because the blocks and steals market is not liquid enough to justify the computational investment in a more sophisticated model.

Blocks Props: Rim Protection Matchups and Shot-Blocking Patterns

Understanding blocks props requires thinking about what generates a block in the first place. A block happens when a defensive player — almost always a centre or power forward positioned near the rim — times his jump correctly to deflect an opponent’s shot attempt. Three conditions need to align: the offensive player must attempt a shot at the rim, the defender must be in position to contest it, and the timing must be precise enough to make contact with the ball rather than the shooter’s arm.

The matchup factor that matters most is how often the opposing team attacks the rim. Teams with dominant drivers who slash to the basket on every possession generate more block opportunities for the opposing centre than teams that rely heavily on three-point shooting. If the opposing team’s primary scoring comes from perimeter shots, the centre is standing under the basket with no one to block. His prop is set based on his season average, but tonight’s game might produce half the rim attempts that the average was built on.

Player-specific shot-blocking patterns add another layer. Some centres are high-volume shot-blockers who contest every attempt near the rim, which produces a wider distribution of outcomes — more games with 3-4 blocks but also more games with 0. Other centres are selective, blocking only the shots they are confident they can reach cleanly. The selective blocker has a narrower distribution and a more predictable output, making his prop easier to model but harder to find mispricing on.

Foul trouble is the wild card in blocks props. A centre who picks up two early fouls will play cautiously in the second quarter to avoid his third, reducing his shot-blocking aggression and his minutes. That conservative approach can easily cost him 1-2 blocks relative to his average, tipping an over/under bet decisively. Foul trouble is impossible to predict pre-game, which adds irreducible variance to the blocks market.

Steals Props: Turnover-Forcing Metrics and Opponent Ball Security

Steals are even more volatile than blocks because they depend on defensive positioning and offensive mistakes in equal measure. A steal is not something a defender creates entirely on his own — it requires the offensive player to make an error, whether that is a careless dribble, a telegraphed pass, or a moment of inattention in traffic.

The most useful metric for evaluating steals props is opponent turnover rate. Teams that rank high in turnovers per game create more steal opportunities for the opposing defence. If your prop candidate is facing a team that averages 16 turnovers per game, the steal opportunities are significantly higher than against a team averaging 12. Not all turnovers are steals — some are offensive fouls, shot-clock violations, or unforced errors — but the overall turnover rate is the best available proxy for steal opportunity volume.

Player archetype matters too. Guards who play aggressive on-ball defence and jump passing lanes generate steals in a fundamentally different way from wings who use their length and anticipation to deflect entry passes. The on-ball defender’s steal rate is more dependent on the opposing point guard’s ball security, while the off-ball defender’s rate depends on the opposing offence’s passing precision. Both are valid paths to steals, but they respond to different matchup variables.

The practical takeaway for steals props is similar to blocks: the under hits at 61.9% across the season, the distribution is skewed, and the bookmaker’s model tends to overstate the probability of the over. The edge is largest when the matchup analysis confirms the direction — when the opposing team protects the ball well and the steal opportunities are genuinely below average for the game.

When to Bet Blocks and Steals — and When to Leave Them Alone

The persistently high under win rates on blocks and steals do not mean you should bet every under blindly. The bookmaker adjusts the price on each specific line, and some lines are priced efficiently despite the broad market inefficiency. A blocks under at 1.55 implies a 64.5% probability, which is close to the actual win rate and leaves little room for edge after the vig.

The edge is largest when the matchup data supports the under and the price has not fully adjusted. If a centre’s blocks are set at 2.5 with the under at 1.70, and the opponent is a perimeter-oriented team that rarely attacks the rim, the true probability of the under might be 72-75% — well above the implied 58.8%. That is a clear positive expected value bet. But if the same centre faces a team that ranks first in paint scoring, the under probability drops to 60-62%, and the price at 1.70 is now roughly fair. Context always matters more than the headline win rate.