Basketball Analytics

How a 6'4" Guard Out-Rebounds Centers

The viral take: "analytics can't explain Josh Hart — watch the games." Actually, analytics explains him beautifully. Pick a player below and watch how positioning, effort and box-outs beat raw height on the 3D court.

Drag to rotate · The shaded zone shows where missed shots actually land; the orange figure is your selected player

6'4"
Height
8.3
Rebounds/game
13.8%
Rebound rate (TRB%)
Elite for any position
Verdict

The stat that settles it: rebound rate

Raw rebounds per game reward minutes and pace. Total rebound percentage (TRB%) fixes that: the share of all available rebounds a player grabs while on the floor.

Player (2023-24)HeightRPGTRB%
Josh Hart6'4"8.313.8%
Myles Turner6'11"6.912.1%
Jaren Jackson Jr6'10"5.59.6%
League avg center~6'11"~15%

So the tweet's core observation is right — Hart really does out-rebound some starting centers — but it's the calculator that proves it, not just the eye test. Analytics and watching the games agree here.

Why height isn't destiny

  • Miss geometry: most misses rebound 6–12 feet from the rim, often to the weak side — long threes carom even farther. Being 7 feet tall directly under the rim can be the wrong spot.
  • Box-outs create real estate: a well-timed box-out matters more than a 7-inch height gap once the ball is in the air.
  • Anticipation: elite rebounders start moving on the shot's release, reading arc and angle. That's a skill, not a measurement.
  • Motor: Hart crashes from the perimeter on nearly every miss; many bigs leak out or ball-watch. Effort compounds over 70 misses a game.
  • Role design: Turner and JJJ are rim-protectors who contest shots first — contesting often takes you out of rebounding position. Low TRB% is partly the job, not just the player.

Analytics vs the eye test — a false fight

The best framing: the eye test generates hypotheses, analytics tests them. "Hart rebounds like a big" is an eye-test claim that TRB% confirms. "JJJ is a bad rebounder" is an eye-test claim the numbers complicate — his blocks and contest rate explain a chunk of the gap.

When they disagree, dig deeper: check per-possession stats, role context, and tracking data (contested vs uncontested boards). When they agree — as with Hart — you've probably found the truth.

Enjoy this tool? Build your own with Super