# Basketball player archetypes

*Archetypes*

How our analysts describe each player as a mix of archetypes, to find who plays like the player you are about to lose, in 171 leagues.

## What it does

### A mix, not a label.

A player is rarely one thing. Every player-season is described as a share of each archetype: a guard can be half High Volume Scorer and a third 3&D Wing in the same season.

### Built per league.

Each league gets its own taxonomy, fitted on its own data, so an archetype means what it means in that competition. Values are not compared across different runs.

### The questions it answers.

Who plays like the player we are losing? Which candidates fit the profile we need before the window opens? Which players in our league share a role, and how close are they?

### The method.

The approach is described in our accepted paper at the MIT Sloan Sports Analytics Conference 2026: *Scouting Anyone: Probabilistic Player Archetypes for Any League.*

### Inside an engagement.

Archetypes is the tool our analysts use in client work. It is not sold as a separate product.

- **Compare players.** Up to four players side by side, each one as a mix of archetypes that adds up to 100%.
- **Similar players.** The nearest neighbours of any player in archetype space, ranked by similarity score.
- **Archetype comparator.** The average statistical profile of two archetypes, as percentiles within the league, metric by metric.

[Where to see this tool in public data](https://www.clutchdata.net/methodology#archetypes)

---

Clutch Data · basketball analytics consultancy · info@clutchdata.net · https://www.clutchdata.net/solutions/archetypes