Branden Carlson — Two-Way Profile (2 Seasons)
Source: BBall Index (bball-index.com) — written with the Basketball Analytics Expert Sage active.
The Question
The complication on Carlson is sample. He has played 246 minutes (24-25) and 488 minutes (25-26) — a total NBA career of 734 minutes across two seasons, both as a third-string big behind Hartenstein, Chet Holmgren, and Jaylin Williams. The 488-minute year-two reading is the largest sample we have, and it's about a third of what a typical rotation player accumulates in a single season. Every reading here is provisional in a way the multi-year players in this audit aren't.
The framing question: a deep-bench big with very limited minutes whose component readings have moved positively year-over-year — what's actually there in the data, and how should OKC weight it for the role he occupies?
Part 1: The Big Picture
The composite picture is mildly positive but with a sample-size warning attached. LEBRON has gone -0.28 → 0.86 — first clearly net-positive year, but on roughly twice the minutes (still small in absolute terms). D-LEBRON at 0.65 (up from -0.04) is the cleaner positive of the two layers; O-LEBRON at 0.21 (up from -0.24) is barely net-positive. The component story is "both layers moved positive," which is the same year-two pattern Mitchell and Cason show — just on a much smaller sample.
LEBRON WAR at 1.02 is small in absolute terms but reflects the modest minutes: 488 min × ~2.1 LEBRON/100-poss ≈ ~1 win above replacement, which is about right for a third-string big. Offensive Talent at -1.16 is the one reading moving the wrong way (the talent layer says the player is slightly less productive on offense in year two, even as the headline impact has improved) — that's likely role-context noise on a small sample, but it's worth flagging.
Minutes Per Game at 11.62 is the cleanest indicator of the role: this is third-string-big territory, the player who plays when one of the top two bigs is in foul trouble or out for rest. Usage Rate at 18.9% is slightly higher than you'd expect for the role — when he is on the floor, he is being asked to take a real share of possessions.
| Metric | 2024-25 | 2025-26 | Read |
|---|
| LEBRON | -0.28 | 0.86 | First clearly net-positive year — small sample |
| O-LEBRON | -0.24 | 0.21 | Barely net-positive |
| D-LEBRON | -0.04 | 0.65 | Cleaner positive — the engine |
| LEBRON WAR | 0.34 | 1.02 | Right shape for the minutes |
| Offensive Talent | -0.78 | -1.16 | Talent layer moving the wrong way |
| Minutes Per Game | 7.7 | 11.6 | Third-string-big role |
| Usage Rate | 19.3% | 18.9% | Real share when on floor |
Part 2: The Offense
Stretch shooting on real volume — the load-bearing skill
The shape is what you want from a third-string big in a non-spacing offense. 3PT % at 36.05% is in the credible range for a stretch shooter, and Stable FG3% at 36.01% (vs. 35.43% rookie-ish) is essentially flat — the model treats this as a real ~36% shooter at this point. 3PT FGA Per 75 at 6.45 is real volume — he is taking nearly as many threes per 75 possessions as Hart (who takes very few) and Chet (~4.4) by design.
3PT Attempt Rate at 46.24% is below the 68% reading from his ultra-small 24-25 sample (where he was almost exclusively a corner shooter); the year-two role has him taking a more balanced shot diet, with Rim FG% at 77.22% on small volume showing he can finish at the rim when the defense rotates. C&S 3PT Shooting Talent at 0.33 is positive; the catch-and-shoot specifically reads as the load-bearing offensive skill.
| Metric | 2024-25 | 2025-26 | Read |
|---|
| 3PT % | 33.3% | 36.1% | Credible ~36% on real volume |
| Stable FG3% | 35.4% | 36.0% | Essentially flat — the prior |
| 3PT FGA Per 75 | 9.52 | 6.45 | Real volume; less extreme than rookie sample |
| 3PT Attempt Rate | 68.0% | 46.2% | More balanced shot diet year two |
| C&S 3PT Shooting Talent | 0.98 | 0.33 | Positive — load-bearing skill |
| Rim FG% | 77.8% | 77.2% | Solid finishing on small volume |
Creation and connector — the secondary layer
Box Creation at 2.58 (up from 2.23) is modestly positive; Playmaking Talent at -0.94 is meaningfully negative (worse than the rookie -1.04). Finishing Talent at -0.34 is in the slight-negative band. Screening Talent is in his low-positive band. On-Ball Gravity percentile is in the 60s; Off-Ball Gravity percentile in the 60s — defenders are paying him reasonable attention as a shooter.
The composite offensive picture: a stretch-and-finish big whose value is concentrated in the perimeter shot and the rim finish, with limited connector value beyond. That's the right shape for the role; the Playmaking Talent -0.94 is the cleanest signal that this isn't a passing-hub big in the Hartenstein mold.
| Metric | 2024-25 | 2025-26 | Read |
|---|
| Box Creation | 2.23 | 2.58 | Modestly positive |
| Playmaking Talent | -1.04 | -0.94 | Meaningfully negative — not a passing hub |
| Finishing Talent | -0.7 | -0.34 | Slight-negative band |
Part 3: The Defense
Help defense on small sample — the surprise
The defensive component readings are the part most worth looking at, with the sample-size caveat applied heavily. Help Defense Talent at 2.24 (up from 1.66) is in the elite band — for context, Jaylin Williams is at 2.14 and Hartenstein is at 1.99 on much larger samples. Help Defensive Activity at 2.27 (up from 1.92) is the volume layer saying the same thing. The honest read is that if the sample holds up, this is a genuinely valuable help defender; the sample (488 min) is too small to call it the player's true level with confidence.
Rim Protection at 0.23 is mildly positive; Post Defense at 0.21 has recovered to neutral from -0.73 last year. Blocks Per 75 at 1.96 is decent for a backup big.
The matchup distribution confirms the role shape. Matchup Difficulty at -1.89 is the most extreme negative on the team — he is being matched up against the easiest offensive players on the floor by a wide margin. That's the right scheme for a third-string big with very limited minutes; he's not being asked to take playoff-level assignments and the impact reading should be weighted accordingly.
| Metric | 2024-25 | 2025-26 | Read |
|---|
| D-LEBRON | -0.04 | 0.65 | Net-positive year two — small sample |
| Help Defense Talent | 1.66 | 2.24 | Elite reading — sample caveat |
| Help Defensive Activity | 1.92 | 2.27 | Elite volume — same caveat |
| Rim Protection | 0.10 | 0.23 | Mildly positive |
| Post Defense | -0.73 | 0.21 | Recovered to neutral |
| Blocks Per 75 | 3.02 | 1.96 | Decent for backup big |
| Matchup Difficulty | -1.86 | -1.89 | Easiest matchups on the team |
Ball screen navigation — the absorbed weakness
Ball Screen Navigation at -1.44 is meaningfully negative and worse than the rookie -1.17. Like every non-elite big in the rotation, he is hunt-able in pick-and-roll. Defensive Positional Versatility at 62.23 is solid for a big — he can credibly defend the 4 and 5, which matters for the small-ball lineups the staff uses him in.
| Metric | 2024-25 | 2025-26 | Read |
|---|
| Ball Screen Navigation | -1.17 | -1.44 | Hunt-able in PnR |
| Defensive Positional Versatility | 61.1 | 62.2 | Solid — defends 4 and 5 |
| Defensive Role Versatility | 46.4 | 58.2 | Wider role year two |
Part 4: Team Fit — Strengths and Absorptions
What OKC gets
A third-string big who fits the bigs rotation in the right shape: a real ~36% three-point shot on real volume (3PT FGA Per 75 6.45) and a help-defense reading that — if the small sample holds up — is in the elite tier. The stretch element is what matters most for the role; in a non-spacing OKC offense, having a third big who can credibly hit the corner three is the difference between the bench unit having usable spacing and the bench unit collapsing the floor. The component layers (C&S 3PT Shooting Talent 0.33, Help Defense Talent 2.24, Defensive Positional Versatility 62) all point the right direction.
The role-fit case: he plays in spot minutes (foul trouble, blowouts, when one of Hart/Chet/Jaylin needs rest) and the staff hides his defensive matchups (Matchup Difficulty -1.89). That's the correct deployment for what the data says he is.
What OKC absorbs
The sample-size limitation is the load-bearing absorbed cost. 734 NBA minutes across two seasons is not enough to know whether the year-two component readings are the player's true level. Help Defense Talent at 2.24 on 488 minutes could be elite-true-talent or sample-noise variance; the multi-year prior to weight against it doesn't yet exist. Forward-looking, the model is reading him optimistically; a forecast probably regresses meaningfully toward the rookie-year readings.
The pick-and-roll vulnerability that Ball Screen Navigation at -1.44 reflects is the second cost — when he's the only big on the floor, the perimeter coverage has to do more work, and the role narrows accordingly. The third cost is the Playmaking Talent -0.94 and Offensive Talent -1.16 readings, which both suggest the offensive value is concentrated in the perimeter shot and doesn't scale beyond it. He is not a connector or a creator; he is a stretch-and-finish big whose role doesn't ask for more.
The two-way verdict
The honest two-year read is a deep-rotation stretch big whose year-two component readings support a useful third-string role — credible perimeter shot, surprising help-defense reading, defensive matchup deliberately easy — but whose sample is too small to be confident the year-two impact is the player's true level. For the role he occupies (12 MPG, third-string), that's the right answer: useful in the regular-season rotation, with the playoff-minutes question resolved by which of Hart/Chet/Jaylin is on the floor at any given time. Forward-looking, the data is encouraging without being load-bearing on the team's playoff math.