NBA Value Explorer

Player Projections

Two gradient-boosted models. The first projects next-season WS/48 and BPM for current NBA players. The second projects NBA career value for the 2026 draft class (and every other D-I player), with the conference / strength-of-schedule adjustment built in.

Showing 50 of 1,793 filtered (2,970 total).
#
1Caleb WilsonCombineNorth Carolina · ACC · Fr · PF/C · 6-1057%9919.89.42.75.314.5891%100%17.115.56
2AJ DybantsaBYU · B12 · Fr · Wing F · 6-987%10025.56.83.76.594.0395%100%15.614.83
3Cameron BoozerCombineDuke · ACC · Fr · Stretch 4 · 6-984%10022.510.24.17.595.6497%100%14.413.90
4Nate AmentTennessee · SEC · Fr · Wing F · 6-1070%9916.76.32.33.794.2085%100%13.611.59
5Kingston FlemingsCombineHouston · B12 · Fr · Scoring PG · 6-479%9716.14.15.24.964.9296%98%10.910.41
6Zuby EjioforCombineSt. John's · BE · Sr · PF/C · 6-975%8916.37.33.54.964.5574%97%13.49.90
7Hannes SteinbachCombineWashington · B10 · Fr · PF/C · 6-1178%9918.511.81.65.584.6985%100%10.99.24
8Yaxel LendeborgCombineMichigan · B10 · Sr · Stretch 4 · 6-975%15.16.83.25.834.9675%96%12.09.01
9Christian AndersonCombineTexas Tech · B12 · So · Pure PG · 6-393%7718.53.67.46.114.4889%98%9.78.63
10Shelton HendersonMiami FL · ACC · Fr · Wing G · 6-678%9613.84.92.13.043.6682%97%10.48.56
11Brayden BurriesCombineArizona · B12 · Fr · Wing G · 6-474%9816.14.92.44.434.6888%100%9.38.13
12Bennett StirtzCombineIowa · B10 · Sr · Scoring PG · 6-494%19.82.64.46.174.4190%93%8.98.01
13Darius Acuff Jr.CombineArkansas · SEC · Fr · Scoring PG · 6-384%9823.53.16.46.663.4692%99%8.37.69
14Dailyn SwainCombineTexas · SEC · Jr · Wing F · 6-882%8217.37.53.64.873.9764%99%12.07.68
15Koa PeatCombineArizona · B12 · Fr · Wing F · 6-864%9914.15.62.63.554.0388%96%7.86.91
16Aday MaraCombineMichigan · B10 · Jr · C · 7-358%9712.16.82.43.354.0963%90%10.16.30
17Keaton WaglerCombineIllinois · B10 · Fr · Combo G · 6-684%5317.95.14.26.464.2368%99%9.16.16
18Pryce SandfortNebraska · B10 · Jr · Wing G · 6-783%8318.14.92.05.064.7945%84%13.66.12
19Chris Cenac JrHouston · B12 · Fr · PF/C · 6-1162%999.57.90.72.524.2278%97%7.65.96
20Labaron PhilonCombineAlabama · SEC · So · Scoring PG · 6-472%9322.03.55.05.993.3250%99%11.25.59
21Morez Johnson Jr.CombineMichigan · B10 · So · PF/C · 6-963%9413.17.31.24.344.1654%98%9.85.31
22Joshua JeffersonCombineIowa St. · B12 · Sr · Wing F · 6-973%16.47.44.84.405.0753%93%9.34.89
23Tyler TannerCombineVanderbilt · SEC · So · Scoring PG · 6-083%19.53.65.15.774.3234%100%14.54.87
24Ivan KharchenkovArizona · B12 · Fr · Wing G · 6-771%8910.44.32.32.614.2945%98%10.64.74
25Meleek ThomasCombineArkansas · SEC · Fr · Combo G · 6-575%9815.63.82.53.893.3979%98%4.93.88
26Flory BidungaCombineKansas · B12 · So · C · 6-978%9713.39.01.54.374.9535%96%11.03.81
27JT ToppinTexas Tech · B12 · Jr · PF/C · 6-964%7821.810.82.15.264.9743%74%8.83.78
28Tounde YessoufouCombineBaylor · B12 · Fr · Wing G · 6-582%9717.85.91.64.043.5869%99%5.43.77
29Alex CondonFlorida · SEC · Jr · PF/C · 6-1174%15.17.53.63.824.6134%96%10.83.67
30Baba MillerCombineCincinnati · B12 · Sr · PF/C · 6-1175%8913.010.33.73.295.1735%86%9.93.45
31Ja'Kobi GillespieCombineTennessee · SEC · Sr · Pure PG · 6-186%18.42.85.45.014.7836%78%9.23.27
32Jalen HaralsonNotre Dame · ACC · Fr · Wing G · 6-757%9616.24.02.63.232.3952%88%6.03.10
33Cameron CarrCombineBaylor · B12 · So · Wing G · 6-584%8918.95.82.64.783.6340%65%7.63.00
34Braylon MullinsConnecticut · BE · Fr · Wing G · 6-658%9712.03.51.42.704.0160%99%5.02.97
35Isaiah EvansCombineDuke · ACC · So · Wing G · 6-671%9715.03.21.33.954.1349%98%6.02.91
36Graham IkeGonzaga · WCC · Sr · PF/C · 6-969%19.98.02.44.934.8832%38%8.62.75
37Ebuka OkorieCombineStanford · ACC · Fr · Scoring PG · 6-282%5823.23.63.66.333.6930%95%9.12.71
38Joseph TuglerHouston · B12 · Jr · PF/C · 6-859%868.45.31.32.863.9023%95%11.12.58
39Henri VeesaarCombineNorth Carolina · ACC · Jr · Stretch 4 · 7-074%8517.08.72.14.754.2435%96%7.22.51
40Trey McKenneyMichigan · B10 · Fr · Wing G · 6-455%969.92.80.92.703.1224%99%10.02.39
41Boogie FlandFlorida · SEC · So · Scoring PG · 6-375%9611.62.53.52.824.2737%99%6.42.35
42Trevon BrazileCombineArkansas · SEC · Sr · Stretch 4 · 6-1076%6313.07.31.63.724.0826%97%8.82.31
43Thomas HaughFlorida · SEC · Jr · Wing F · 6-981%6217.16.12.14.834.8540%32%5.32.12
44Tarris Reed Jr.CombineConnecticut · BE · Sr · C · 6-1059%9314.79.02.34.274.6321%74%10.32.10
45Malachi MorenoCombineKentucky · SEC · Fr · C · 7-056%957.86.31.82.563.0543%13%4.82.03
46Cayden BoozerDuke · ACC · Fr · Scoring PG · 6-457%977.72.33.02.463.2917%99%11.71.92
47Juke HarrisWake Forest · ACC · So · Wing G · 6-787%8421.46.51.94.953.7327%98%7.11.89
48Allen GravesCombineSanta Clara · WCC · Fr · Stretch 4 · 6-957%11.86.51.84.123.1417%94%11.01.85
49Bruce ThorntonCombineOhio St. · B10 · Sr · Combo G · 6-290%9119.95.13.97.234.1921%99%8.21.68
50Milan MomcilovicCombineIowa St. · B12 · Jr · Wing F · 6-876%9216.93.11.05.114.1723%62%7.11.64
Projecting first-4-year NBA value for 25-26 college players · Draft-stock AUC 0.98 · WS Spearman 0.29 · How this works

Three stacked gradient-boosting models, scored together for every D-I player above the minute filter:

  1. Draft-stock classifier — P(this player gets drafted). Trained on 33,705 player-seasons of Torvik data (2014–2024), with the matching drafts.csv lookup giving a 1/0 label per season. Holdout (2023+2024 seasons) AUC 0.984, recall@top-5% 86% (87/6289 positives).
  2. Value regressor — if drafted, E[cumulative NBA Win Shares across first 4 seasons]. Standardized horizon means a 2014 pick and a 2022 pick are scored on the same outcome window (no career-length confound). Leave-one-draft-class-out CV across 2014–2022: MAE 5.19 (vs mean baseline 5.45), Spearman 0.29.
  3. Made-it classifier — P(played ≥500 NBA min in first 4 seasons | drafted). LODCO AUC 0.71, base rate 78%. Surfaced for interpretation alongside the value regressor.

The leaderboard Scorecolumn is P(drafted) × max(0, E[WS yr1-4]) — this combines "is this even a real prospect?" with "if drafted, how good?" The draft-stock term is what keeps the model from rewarding mid-major upperclassmen who happen to put up big numbers against weak competition.

Feature set (71 total): Torvik opponent-adjusted box + composite ratings, shot-diet (rim/mid/dunk rates), bio (height, age, class, HS recruit rank), multi-year trajectory deltas, role one-hots, and NBA Combine measurables (wingspan, standing reach, max vertical, lane agility, three-quarter sprint). Players who attended the combine are flagged in-row.

Data sources: Bart Torvik (barttorvik.com) advanced college stats; Basketball Reference (NBA advanced + per-game); NBA.com stats (draft combine anthro + drills). Model v2. Generated 2026-06-01T21:35:10Z. 2,970 prospects projected, 417 drafted players in the value-model training set.