Tall OrderDigital

NBA · 2026-27 · Week 2

Role Call

One week in, a player's shooting is noise. His role isn't. Here's whose role has really changed this season, and whose only looks like it.

By Shawn Nafziger · Data through Oct 27, 2026

0.85How well a player's after 4 games predicts the rest of his season (, 2025-26)
0.83How well his minutes per game after 4 games predict the rest of his season
0.21How well his after 4 games predicts the rest of his season

How to read these: each number is a correlation, a score from 0 to 1.0 for how well a player's first 4 games predict the rest of his season. 0 means they tell you nothing; 1.0 means they predict it perfectly. Around 0.7 and up, early numbers are a good guide.

Coaches decide roles. Luck decides makes.

How many minutes a player gets, how often the offense runs through him and where he shoots from are decisions, made by coaches and players before the ball goes up. Whether the shot goes in is partly luck. Among the 367 players who played at least 40 games last season, the decisions showed up within 4 games. The makes didn't.

How well a player's first 4 games predict the rest of his season

Correlation across players, 2025-26 regular season (1.0 = perfect).
Orange: role decisions, which settle fast.
Gray: shooting, which is still noise.

Players with 40+ games; rate stats need 40+ minutes in the first 4 games. Source: NBA Stats API.

Whose role changed

Each dot is one player. Across: how his minutes per game changed since last season. Up: how his changed, meaning how much of the offense runs through him.

The shaded box shows what a normal first week looks like: about ±10 minutes and ±6 usage points of random . Being inside the box isn't good or bad. It means we can't yet tell his change from noise.

Orange dots are outside the box: a change bigger than a noisy week explains, so it reflects a real decision about that player's role. 47 of 249 players are there.

Light orange dots are at the edge: at least 80% of the way out of the box, so probably real but not yet certain.

Whether that's good news depends on the player. More minutes for a second-year player is usually good; a smaller share for a star on a new team is a question.

Change since 2025-26: minutes per game and usage

2026-27 through Oct 27 (56 games). Hover or tap a dot for details.

Changed more than a normal weekAt the edge of the boxInside normal week-1 wobbleHollow dot: new team this seasonShaded box: a normal first week
Show players outside the box as a table (* = new team)
PlayerTeamMinutes (change)Usage (change)Points

Three kinds of change

  • New team, smaller share. Jaylen Brown (BOS to PHI) is using 24.9% of plays, down from 35.1%, while playing more minutes (34.4 to 36.8). Ja Morant (MEM to POR): 30.5% to 20.7%. Jonathan Kuminga (GSW/ATL to MIN) is playing 29.8 minutes a game, up from 23.1, but his usage fell from 22.6% to 14.2%.
  • Bigger share of the offense. Naz Reid (MIN to CHA): 21.3% to 28.1%. Brandon Miller: 27.2% to 34.4%. Duncan Robinson (15.2% to 21.3%) is the player behind Detroit's new shot diet. Anthony Edwards, already a high-usage star, is up to 36.7%, right at the edge of the box.
  • More minutes. Khaman Maluach went from 8.9 to 23.4 minutes a game. Dylan Harper (22.6 to 31.1) is at the edge of the box: probably real, but not yet beyond what a noisy week could do.

Same team, smaller role happens too: Alex Sarr's usage is 13.8%, down from 25.2%.

Roles can still change with an injury, a trade or a lineup tweak. But a role change this size in week 1 is a decision, not a fluke.

Method