Pure Hicks Analytics
Pure Hicks Analytics

Basketball stats

Loading league…

Sleeper ↗

Loading stats…

Connecting to Sleeper draft…Live draft & picks ↓

Player stats

Sample and minutes options
Optional: apply one minutes assumption to compare projected FP/game.

Availability estimates are waiting for the Sleeper board.

Availability estimate assumptions

The two estimated-availability columns show the chance a player lasts until your next pick and your following pick, from the current board. The later estimate assumes you pass on him at the earlier turn and take the best remaining ADP alternative. Back-to-back turns therefore have the same availability. Actual drafted players are removed automatically; a pick undo restores them.

Each update runs 1,000 remaining-board simulations. Players are ordered by Sleeper aggregate ADP plus normal variation with standard deviation equal to the larger of 2 picks or the selected percentage of ADP. The spread is an assumption, not measured mock-draft variance. These are model estimates, not calibrated league probabilities. They omit opponent roster needs, keeper reservations, traded picks, players without ADP, and new news not reflected in ADP. Your historical FP/min model still determines player value. Estimates below 1% or above 99% avoid implying certainty from the simulations; an available player on your current turn is shown at 100%.

Stat definitions

FP uses the league’s enabled scoring categories. FP/min divides fantasy points by total minutes played. Per-36 scales that rate to 36 minutes. Actuals show games played and respect the minimum games filter.

Model rank uses the selected model’s FP/game across the full player pool and stays fixed when you filter or change the table rate. ADP − rank is positive when Sleeper ADP is later than the custom points rank. It does not account for positional scarcity or injury absences. ADP is Sleeper’s standard NBA market reference.

Historical models use 2025–26 league fantasy points divided by total minutes, multiplied by historical MPG or Sleeper estimated MPG. Historical MPG divides total minutes by games actually played, not 82. Every enabled category scales linearly, including bonus and foul-event rates. This assumes stable production and does not model nonlinear bonus thresholds. At least 10 NBA games and 100 minutes are required; smaller/no-history samples use a labeled Sleeper fallback.

Sleeper points remain available as a comparison. The selected projection model also drives draft recommendations, simulations and roster values; viewing actuals does not change that model. Minutes scenarios are always FP/game. Historical scenarios scale all categories; Sleeper/fallback scenarios scale counting stats with bonuses and fouls held fixed. None is a season availability forecast.

Live draft

Your league loads automatically. Picks sync every 5 seconds during the draft, or 15 seconds before it starts, while this tab is visible. Keep Sleeper open to make your selections.

Connecting to Sleeper…

Pick log (0)
    Read a different Sleeper draft
    League scoring

    Double-double and triple-double bonuses stack. Missing projected bonus/foul fields use prior-season rates.

    Draft simulations

    Each trial simulates all 14 rounds against 11 opponents. Strategy scores sum the selected model’s FP/game in the best legal nine-player lineup. Bands show the 10th–90th percentile across drafts. These are model comparisons, not win probabilities. Opponent variation is assumed, not measured from individual mock results. Injury absences and weekly Game Pick/Lock-In decisions are not simulated.

    Simulation assumptions and limits

    Opponents follow a shared noisy order centered on Sleeper ADP. Every strategy sees the same opponent draw. Custom-points strategies consider players within roughly 18 picks of the current market and are heuristics, not a globally optimal draft policy. The roster-fit strategy values the marginal gain to an exact position-eligible starting lineup plus 12% of a player’s FP for bench depth. It is evaluated using the same projections that drive its picks, so its apparent advantage needs out-of-sample validation.

    Included players with valid standard ADP enter the pool, including flagged injured players. The default value uses historical FP/min × Sleeper MPG with labeled insufficient-history fallbacks; changing Model / source to another projection basis requires new simulations. Review those flags; the model does not assume they are healthy. Keeper selections, pick trades, actual opponent preferences and schedule-specific availability need to be incorporated before treating this as a final draft plan. The league’s draft order is authoritative only after Sleeper assigns it.

    My roster

    Player detail