← All terms
Impact models
DARKO
Predictive points per 100 possessions versus an average player.
What it teaches
- How much a player is expected to help or hurt a team’s scoring margin.
- A blended offense + defense view, not just points scored.
- Who projects as high-impact going forward, not only who stacked counting stats yesterday.
What it doesn't
- Who “tried hardest” or who had the flashiest highlights.
- Chemistry, coaching schemes, or playoff-specific matchups by themselves.
- A guarantee. It’s a projection, not a box-score fact.
Upsides
- Puts stars and role players on one comparable scale.
- Updates as new games come in.
- Splits offense and defense so you can see where the value lives.
Downsides
- Opaque internals: you can’t recompute it from a box score alone.
- Can lag sudden role changes or brand-new injuries.
- Easy to overtrust one number without minutes, age, or context.
How to apply it
- Rank trade targets and free agents on the same board.
- Pair with minutes and age: a +2.0 at 36 minutes is different from +2.0 at 12.
- Use offense/defense splits when building lineups, not just the total.