Six through
the cut.
Eight slots,
one cap.
Survival is the objective, not the ceiling. Salary is read as a ceiling and never as a target, and nothing under sixteen tour rounds is rosterable.
DraftKings pool, ESPN rate stats, scored on DK's own table. No market layer, because none is needed to price a projection.
Lineups
Survival board
What this will not do. It will not roster a golfer with fewer than sixteen tour rounds behind him, because that one filter is the measured edge and the salary cap will never force it. It will not select on low variance: controlling for skill, variance is worth two points of cut probability, and a pure floor-first build finished below random in 249 out-of-sample events.
Lineups
Player pool
What this will not do. It will not claim an ownership read. No NBA ownership feed is reachable, so the GPP weighting here is a salary-implied chalk proxy and is labelled as one on every response. An uncalibrated proxy of exactly this kind priced a thirty-eight percent running back at six percent, which is why it never masquerades as a measured field.