College → NBA Archetype Translation

This model is the college → NBA analogue of the D2 → D1 translation work: the college archetype is the input, NBA performance is the outcome.

Matching

• Each NBA player is matched to their final college season by name, after stripping accents, punctuation, and suffixes — so V.J. Edgecombe, Egor Dëmin, and Kasparas Jakučionis resolve.
• A chronology guard requires the NBA debut to fall after the final college year, dropping veteran name-collisions.
• Captures drafted and undrafted players; draft slot is carried in the college data itself.

Outcome metrics (six categories)

• Rotation rate — share reaching 20+ MPG in a season. Stick rate — share with 2,000+ career minutes.
• Quality rate — share reaching league-average BPM (≥0) in a 500+ minute season.
• Best BPM, WS/48 — peak value, minutes-qualified (sub-500-min seasons excluded, clipped ±12). Career minutes — opportunity earned.

Draft-slot control

Better archetypes tend to be higher picks, so raw rankings can just re-measure the draft. Each outcome is residualized on log draft pick (undrafted = slot 75); "vs. draft expectation" shows performance relative to where players were taken.

Projections: Outlook vs. Projection

The two columns on the Projections tab answer different questions, and this is the distinction that matters for a player like Cameron Boozer:

• Outlook is an archetype base rate — the historical rotation rate of that archetype, nothing about the individual. Athletic Bigs land ≈49% of players in NBA rotations, which is why the archetype reads "Mixed." It is deliberately blind to talent, so an elite prospect and a fringe one in the same bucket get the same Outlook. Tiers: Translates well ≥50%, Mixed 40–50%, Below average 25–40%, Translates poorly <25%.
• Projection is the player-level score: 0.50 × stick score + 0.20 × team quality + 0.20 × youth + 0.10 × archetype base rate — weights set by analytics and basketball logic, not tuned to any draft. The stick score also includes a size-fit term: each player's height is compared to the NBA-makers of their archetype (Pure Shooters who reached the league average ~6'7"), so an undersized player for their role is discounted and an NBA-sized one is credited.. The stick score weights each archetype on its signature skills — Pure Shooters and Stretch Bigs on 3P%/FT%/efficiency, Playmakers and P&R Creators on assists and ball security (low turnovers), Shot-Making Guards on shot creation and usage, Athletic Bigs on rim protection and rebounding — plus a universal two-way-impact block (RAPM, WS/40) that separates stickers in every archetype. Each current player is then benchmarked against the historical NBA-makers of their archetype: a stick score of 60 means "beats 60% of the players from this archetype who actually reached the NBA," and the median college rotation player scores about 30. Youth is class year (below). Team quality is the program's KenPom AdjEM percentile, and the archetype term is a light prior.
• Youth is the single strongest signal. In the matched players, class year correlates −0.40 with NBA career minutes — stronger than any box stat. Freshmen who reach the league stick at 61% and average ~2,460 minutes; seniors stick at 35% and average ~800. So a freshman posting a given line projects far better than a senior posting the same line, which is why a 5th-year stat-stuffer no longer outranks a freshman blue-chipper. Mapping: Freshman 100, Sophomore 82, Junior 48, Senior 28, Graduate 20.
• Floor vs. Ceiling — two different questions. Translate (floor) asks will they stick and is driven by role-player production, youth, and level. Ceiling asks how good could they get and is a separate score: it weights overall value (WARP/40, WS/40), size and rim presence (blocks, rebounds), efficiency (TS%), impact (RAPM), and youth — the traits that actually predicted NBA quality among past players (WARP/40 r≈0.36, blocks r≈0.29). Critically, 3-point volume is left out of Ceiling, because it was slightly negative for NBA quality — the clean basketball split is that pure shooters have a high floor but a modest ceiling, while young two-way bigs and creators carry the ceiling.
• Hits & Misses — the retro test. For players already in the NBA, the same college-based model gives an Expected value; we compare it to how good they Actually became (peak BPM + minutes + total VORP, as a 0–100 percentile among matched players). The gap is the verdict: Exceeded (e.g., Jalen Williams, a Santa Clara guard the profile rated middling who became an All-NBA-level player), Met, or Fell short (hyped college bigs whose production didn't translate). It's both a sanity check on the model and a map of which archetypes and profiles the box score systematically over- or under-rates.
• Tiers are calibrated to draft reality. Only ~45 college players are drafted each year, so the cutoffs are set by rank, not by generous score bands: Elite ≈ top 20 (first-round look), Elite + Strong ≈ top 60 (draft-caliber), Solid ≈ next ~150 (pro radar / two-way range), then Fringe and Long shot. Out of 2,380 current rotation players, only 61 grade Elite or Strong.

Coverage & caveats

• Window: college 2020-21 → 2025-26; NBA box + advanced 2022-23 → 2025-26.
• 2021 draft class excluded from the archetype study — the calculator starts at 2021-22, so 2020-21 finishers (Franz Wagner, Evan Mobley, Scottie Barnes) have no archetype.
• Recent classes truncated: 2025 draftees have only partial 2025-26 NBA data.
• Small samples: Stretch Big (n=7) is directional only.
• College archetype → NBA outcome. A true archetype-to-archetype crosswalk needs NBA shot-location data.