--- title: DataDawg$ — auction conversion method contract as_of: 2026-08-28 source: Establish The Run half-PPR auction values (private subscriber snapshot, identified publicly only by SHA-256), converted to the JohnMaddenPepperoniNipplesXV room (Yahoo 773763; 14 teams, $200, no kicker, 2 flex, $0 bids legal) by the VOR-based conversion documented in /data/datadawg-dollars-method.md. Reproducible and red-teamed; not outcome-validated. staleness: A dated conversion of a dated ETR snapshot. Keeper inflation is not modelled and the Sep 8 keeper deadline will move real prices. canonical_url: https://datadawgs216.com/data/datadawg-dollars-method.md --- # datadawgs-ppn-auction-2026-v3 — method contract **As of:** 2026-08-28 · **Status:** reproducible conversion; not outcome-validated Translate ETR 12-team half-PPR auction values into the PPN league without replacing ETR player judgment. ## League { "teams": 14, "budget_per_team": 200, "minimum_bid": 0, "paid_slots_per_team": 15, "total_budget": 2800, "starters": { "QB": 1, "RB": 2, "WR": 2, "TE": 1, "FLEX_RB_WR_TE": 2, "DST": 1 }, "bench": 6, "IR": 2, "kicker": false } ## Scoring { "completion": 0.25, "incompletion": -0.5, "passing_yards_per_point": 20, "passing_td": 4, "interception": -2.5, "sack_taken": -1, "rushing_yards_per_point": 10, "rushing_td": 6, "reception": 0.5, "receiving_yards_per_point": 10, "receiving_td": 6, "return_yards_per_point": 25, "return_td": 6, "fumble_lost": -2.5, "pick_six_thrown": -1, "completion_40_plus": 0.5, "rush_40_plus": 1, "reception_40_plus": 1, "rushing_first_down": 0.25, "receiving_first_down": 0.25, "fractional_yardage": false } ## Authoritative prior { "provider": "Establish The Run", "input": "ETR Half PPR auction dollars", "source_snapshot_sha256": "1d772b1db0b5e79a06835b89d0589c5cc469930910798fef86d0975c3b511b8c", "source_rows": 456, "public_raw_source": false, "rule": "ETR controls player preference; outside projections may alter only league-format deltas." } ## Method { "source_floor": 1, "baseline_premium": "max(ETR_half - 1, 0)", "beta": 0.5325254870874315, "depth_pass_through": 0.5, "scoring_pass_through": 0.5, "central_weight": "max(0, baseline_premium + beta * [0.5*(VOR_14_standard - VOR_12_standard) + 0.5*(VOR_14_custom - VOR_14_standard)])", "exact_price": "2800 * central_weight / sum(central_weight for eligible, non-overridden players)", "rounding": "Hamilton/largest remainder to exact integer total of 2800", "replacement": "Fill mandatory starters, then allocate 28 flex positions to the highest projected RB/WR/TE; replacement is the best unselected player at each position.", "priced_pool": { "QB": 26, "RB": 68, "WR": 78, "TE": 24, "DST": 14 }, "sensitivity_scenarios": [ "budget_only 0/0", "cautious 0.25/0.25", "central 0.5/0.5", "full_depth 1/0.5", "full_conversion 1/1", "no_returns", "full_return_role", "ETR_INT_minus_1" ], "interval_meaning": "model_low/model_high are conversion-assumption sensitivity bounds, not player-outcome confidence intervals and not bid ceilings." } ## Auxiliary inputs { "season_projection": "FFToday 2026 raw component projections; used only for deltas", "event_rates": "2025 nflverse sacks, first downs, 40+ plays and fumbles; shrunk toward position means", "return_roles": "uncertain; no standalone return-TD premium" } ## Manual overrides [ { "player": "Jayden Higgins", "id": "00-0038130", "target": 0, "reason": "Torn ACL; out for 2026 season", "source": "https://www.nfl.com/news/texans-wr-jayden-higgins-torn-acl-out-2026-season" } ] **Amendment (2026-08-28, Claude):** Jayden Higgins does not exist as a row in the ETR source snapshot or the final board; the override is recorded for audit but zeroes nothing. Gate rewritten as 'Jayden Higgins absent or $0'. ## Rejected adjustments - positional games-played haircut (double-count risk) - concave price exponent labeled as certainty pricing - hand-set QB multipliers - uniform 14/12 price scaling - forced $1 league floor - arbitrary star cap ## Validation { "rows": 424, "non_kickers": 424, "positive_prices": 121, "target_sum": 2800, "duplicate_ids": [], "negative_prices": 0, "top_player": { "player": "Jahmyr Gibbs", "target": 90 } } ## Limitations - No keeper list or costs; values are pre-keeper inflation. - The 50% pass-through is declared shrinkage, not an empirically optimized coefficient. - One outside mean-projection system cannot reproduce ETR injury, ceiling or qualitative judgment. - No historical PPN auction-clearing data was supplied. - D/ST is left at $0 rather than inventing precision from unavailable rare-event forecasts. ## Falsification & update - Recompute when the ETR snapshot changes. - Recompute after keeper names/costs are known. - Track actual clearing prices and season outcomes; estimate pass-through and tail concentration next season. - A source hash, row count, total-budget or duplicate-ID failure blocks publication. ## v4 — the practical auction curve v3 allocated the whole $2800 by value over replacement, which concentrates it on the ~121 players who clear the replacement line and gives $0 to everyone else. That is a correct valuation and a poor bid sheet: 210 roster spots get bought on draft night, and a $0-minimum room still spends real money on mandatory starters, D/ST and late nominations. v4 takes a **soft behavioural reserve of $0.75 per auctioned slot** off the premium pool and hands it back across the 210 slots: soft_reserve_total = 210 x 0.75 = 157.5 remaining_premium_pool = 2800 - 157.5 = 2642.5 premium_scale = 2642.5 / 2800 = 0.94375 new_exact = 0.75 + 0.94375 * v3_exact (inside the 210-slot pool) new_exact = 0 (outside it) Integers come from Hamilton/largest-remainder to exactly $2800, tie-broken on the exact value and then the ETR rank. **The $0.75 is not a minimum bid.** $0 bids stay legal, no $1 floor is imposed, and every published integer falls out of the rounding. It models where a real room's money goes in aggregate, not what any single player must cost. **This is a dollar-allocation patch, not a ranking.** ETR still decides who is better than whom. The transform is monotone in the v3 exact value, so ETR order is preserved exactly; only integer-dollar ties move. **The 210-slot pool** is 26 QB, 68 RB, 78 WR, 24 TE, 14 DST. ETR leads where ETR speaks: every player it valued above zero is in the pool, in its order. Below roughly player 122 the ETR snapshot carries etr_half 0 for everyone and the rows sit in *alphabetical* order, so selecting on that rank would seat James Conner and CJ Stroud and bench Najee Harris and Geno Smith purely on spelling. The remaining slots are therefore filled by the site's own board rank, which is projection-driven and continuous. No outside vendor, ADP or consensus is consulted — board rank only orders players ETR itself declined to separate. **Jayden Higgins** is an explicit row at $0 under a season-ending ACL override, taking the payload from v3's 424 rows to 425. He is excluded from the pool: a player who cannot be rostered does not occupy one of the 210 slots the reserve is spread across. > DataDawg$ is an opening-state auction target built from ETR player values, translated to this league’s scoring, roster depth, $2,800 budget, and expected auction spending across 210 roster spots. It is a planning value, not a guaranteed clearing price or an automatic maximum bid. Reassess after major purchases based on remaining budget and roster needs.