Team: Houston Rockets · Status: standard contract · Positions: SG/SF · Age entering season: 23 · As of: 2026-07-18
2026-27 median projection
| GP | MPG | FG% | FT% | 3PM | PTS | REB | AST | STL | BLK | TO |
|---|---|---|---|---|---|---|---|---|---|---|
| 76 | 33.0 | .545 | .765 | 0.3 | 18.0 | 7.6 | 4.9 | 1.4 | 0.6 | 2.3 |
Nine-cat: top-25 to top-45, strongest in punt-threes builds. Reference points: 38.2 FP/G, grade A-. Dynasty: 9.5/10, foundational top-20 tier.
Role and evidence
Thompson started all 79 games in 2025-26 and averaged 18.3 points, 7.8 rebounds, 5.3 assists, 1.51 steals, and 0.61 blocks in 37.4 minutes while shooting .534 from the field, .216 from three, and .779 at the line (Rockets end-of-season game notes). The projection trims workload after VanVleet's return but preserves elite transition and defensive events.
Public price is unsettled. Dynatyze placed him 16th on its July 3 board, FantasyPros had him 22nd, a Reddit nine-cat board put him 31st on April 14, and SI ranked him 43rd on June 4. Those are mostly different valuations of his shooting categories rather than different views of his athletic impact.
The April 7 Dunker Spot video segment treated Thompson and Sheppard as Houston's swing players. VanVleet can lower Thompson's assists, but better spacing and transition hit-aheads can improve his efficiency.
Contract: the rookie-scale extension
This is the largest single dynasty input on the file and it is live and contested. Thompson became extension-eligible in July 2026 after Houston picked up his 2026-27 option. ESPN's Bobby Marks has him "at or close to a five-year, $251 million extension" — the 25% max — and specifically flags that Houston "will need to be careful of negotiating 30% language if Thompson is named Defensive Player of the Year next season," the Rose Rule escalator. Houston Chronicle beat writer Varun Shankar reports the other side: do not expect a full max, and Houston's precedent of team-friendly extensions "should keep him in the ballpark of 20-23% of the salary cap." The reference case is Şengün, who signed for five years and $185M when he was extension-eligible a year earlier.
He is simultaneously the only player Houston has told rival teams is near untouchable (Kelly Iko, Yahoo Sports, June 22 2026). The two facts are consistent: Houston intends to keep him and intends to pay under the max to do it.
Fantasy read: none of this moves his 2026-27 box score, and the 9.5/10 dynasty grade stands. It matters for dynasty in two indirect ways. A sub-max number preserves Houston's flexibility to keep talent around him, which supports his efficiency and assist environment; a full max with DPOY language does the opposite and raises the odds that the roster gets stripped around him. And an unsigned extension carried into the season is the one scenario that would put his name back into trade conversations he is currently insulated from.
Sources: SI · Shankar via The Dream Shake · Spotrac
Range, confidence, and reversals
- Ceiling: 20/9/6 with 1.8 steals, 55% FG, and a tolerable FT%; first-round value in a threes punt.
- Floor: VanVleet takes more creation, opponents ignore Thompson outside, and FT% regresses; useful but closer to round five in balanced nine-cat.
- Confidence: roster high; minutes high; counting rates high; shooting rates medium.
- Category plan: do not draft him and then accidentally chase threes at the expense of his strongest build. Monitor FT attempts, not percentage alone.
Career record
Pre-NBA: No college. Two seasons in Overtime Elite, the Atlanta-based pay-for-play league. In 2021-22 with Team OTE he averaged 14.0 points, 6.6 rebounds, 3.8 assists and 2.1 steals and lost in the finals; in 2022-23 with the City Reapers he averaged 16.4 points, 5.9 rebounds, 5.9 assists and 2.3 steals, made All-OTE First Team, and won the title in a 3-0 sweep of the YNG Dreamerz. Shooting splits from OTE are not reliably published and are not stated here. Note that some outlets list his 2022-23 line as 16.3/6.4/6.2, so the per-game figures are approximate. Draft: 2023, first round, fourth overall, Houston. Not traded.
| Season | Age | Team | GP | MPG | PTS | REB | AST | STL | BLK | 3PM | FG% | FT% | TO |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2023-24 | 21 | HOU | 62 | 22.4 | 9.5 | 6.6 | 2.6 | 1.3 | 0.6 | 0.1 | .536 | .684 | 1.5 |
| 2024-25 | 22 | HOU | 69 | 32.2 | 14.1 | 8.2 | 3.8 | 1.4 | 1.3 | 0.4 | .557 | .684 | 2.0 |
| 2025-26 | 23 | HOU | 79 | 37.4 | 18.3 | 7.8 | 5.3 | 1.5 | 0.6 | 0.3 | .534 | .779 | 2.4 |
| Career | HOU | 210 | 31.3 | 14.3 | 7.6 | 4.0 | 1.4 | 0.8 | 0.3 | .541 | .731 | 2.0 |
Honors: All-Rookie Second Team (2024), All-Defensive First Team (2025). In 2026 he missed both All-Defensive teams by the narrowest margin, tied with Stephon Castle as first player off the list.
Fantasy history: Season-end nine-category ranks are not verifiable from public sources and are not estimated here. The shape is unambiguous from the raw lines, though, and it has been the same all three years: field goal percentage, rebounds and steals are strong positives, three-pointers are close to a zero, and free throw percentage was a drag until 2025-26. The 2024-25 season was his best category profile in one specific respect — 1.3 blocks alongside 1.4 steals made him a two-way defensive-stat asset, which is rare from a guard. That collapsed to 0.6 blocks in 2025-26 even as everything else grew, so 2025-26 is the better points-league season and arguably not a clean upgrade in categories. Points leagues have rated him above his category rank in every season because the volume is real and the missing three-point production costs nothing there.
Durability: 62, 69, 79. The rookie-year absences were a grade 2 left ankle sprain early in 2023-24 plus a short G League assignment in early December 2023 for development, not injury. 2024-25 had no widely reported significant injury. 2025-26 was the cleanest year of his career: he missed one game to quad tendinitis and left a game against Washington in early March 2026 with a right ankle injury but returned and finished both the regular season and the playoffs. Availability is trending the right way and is currently a strength rather than a risk.
Trajectory: The three-year climb is real but a large share of it is minutes, and separating the two matters. Per 36 minutes his scoring went from roughly 15.8 in 2024-25 to 17.6 in 2025-26 — growth, but far less than the 14.1-to-18.3 headline suggests — while his rebounding actually fell per minute, from about 9.2 to 7.5, as Houston moved him further from the glass. The genuinely new skills are the passing, which rose per minute as well as in raw terms once he took on more on-ball creation with VanVleet out, and the free throw stroke, which jumped from .684 in each of his first two seasons to .779. That free throw move is the single most important development for his fantasy value, because it converts his high foul-drawing rate from a liability into a neutral or better. The unresolved issue is the block rate, which halved without an obvious explanation and should be treated as noise-prone rather than a settled decline. At 23 with 37.4 minutes already banked, he is at the point where further value has to come from the shot profile rather than workload, and the three-point attempt rate is the only lever left.
Career sources: ESPN career stats · StatMuse career per game · Wikipedia · SI, 2025-26 All-Defensive teams
Open questions
- Does Thompson's three-point attempt rate rise without damaging rim pressure?
- How much primary creation survives VanVleet's return?
- Does Udoka use him as the nominal shooting guard or as the point forward in closing groups?
- Does the extension land at Marks' near-max or Shankar's 20-23%, and does it carry Rose Rule DPOY language?