Loading DraftADynasty…
Experimental · Frozen source
11,978 data points · 1,562 fields
Signal or noise · historical research
These verdicts summarize the frozen 2024–2025 evaluation. They do not promote experimental forecasts into current recommendations. A single correct pick is not proof of a repeatable pattern.
One game
80% margin range: ±18.0 points; observed coverage 82.3%
A single upset or correct pick cannot tell us whether a forecast has a lasting edge. Judge frozen predictions over many games.
The range is calibrated from historical errors; it does not capture every roster or injury change.
Across games
Model − Elo Brier -0.0078; 95% block interval -0.0140 to -0.0018 (n=569)
Lower Brier is better. The paired interval stays below zero across the pooled held-out games.
This is a historical model comparison, not proof of an advantage on the next matchup.
Across seasons
1 of 2 seasons have a model-favoring 95% interval versus Elo
2024: -0.0139 to 0.0042 · 2025: -0.0186 to -0.0024
Only the evaluated seasons are represented. More unseen seasons and model-change controls are needed before calling this durable.
Player to next season
Candidate MAE 3.103 vs age plus production 3.098 (3009 player-seasons)
The candidate does not beat the strongest existing baseline on this point-error measure.
Inspected historical seasons and excluded rookie or thin-history cohorts limit any claim about next-season skill.
Historical market check
Model − market Brier 0.0107; 95% block interval 0.0040 to 0.0185
Positive values favor the historical market on games with matched odds.
Historical odds lack archived quote times. This is not evidence of a real-time or tradable edge.
Frozen artifact d5b0aa30726c · Snapshot Sep 5, 2026, 1:00 AM UTC
This-page CSV uses field path, type and value columns, including source context. Full dataset downloads are above.
11,978 matches · 11701–11750 shown. Every field remains available in the full download. Filters are saved in this page’s address.
| Field path | Value & type |
|---|---|
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2024"]["interval_count"] | 434number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2024"]["mae"] | 3.2453158767972754number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2024"]["mean_interval_width"] | 10.6283168145923number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2024"]["spearman"] | 0.6675199290589215number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2025"]["count"] | 419number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2025"]["coverage"] | 0.8400954653937948number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2025"]["interval_count"] | 419number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2025"]["mae"] | 2.9640951368979676number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2025"]["mean_interval_width"] | 10.56746041321445number |
| $["player"]["next_season_ppr_ppg"]["summaries"]["prior_production"]["perSeason"]["2025"]["spearman"] | 0.7051225058357592number |
| $["playerLimitations"][0] | Retrospective research: 2019-2025 outcomes were already inspected before this experiment; no pristine confirmation set.string |
| $["playerLimitations"][1] | Source snapshots are later-retrieved nflverse history, with assumed season-complete feature timestamps, not archived publication-as-of receipts.string |
| $["playerLimitations"][2] | Original football is an unchanged historical comparator containing recent_team scoring attribution; the nested candidate excludes that index and constant age.string |
| $["playerLimitations"][3] | Full-PPR season totals train directly on explicit dataset totals; missing labels are excluded, never filled with zero. Existing dataset zero-target semantics are inherited.string |
| $["playerLimitations"][4] | Prior-season eligible-player cohort excludes rookies and other players without qualifying prior production; results do not cover every draftable player.string |
| $["playerLimitations"][5] | No historical Sleeper, ESPN, Yahoo consensus or prediction-market probabilities are synthesized or used in these historical folds.string |
| $["playerLimitations"][6] | Earlier outer residuals give out-of-sample temporal conformal-style intervals; dependence and distribution shift mean nominal coverage is not guaranteed.string |
| $["playerLimitations"][7] | Player-block bootstrap preserves repeated players but not all shared season/team shocks; intervals describe conditional retrospective uncertainty.string |
| $["playerLimitations"][8] | Public ranking influence and promotion are false; this experiment cannot establish market leadership.string |
| $["playerReportSha256"] | 9fb75caa6a26e48973092102a4b3760fada59eb211151b9153ccf62da2edf340string |
| $["providerStatus"] | Existing Sleeper, ESPN and Yahoo connections preserved; this frozen game experiment does not yet use their live fields.string |
| $["publicRankingInfluence"] | falseboolean |
| $["schema"] | draftadynasty.forecast-lab/v1string |
| $["simulationMethod"] | Seeded independent game draws with fixed current team strength; scheduled rest only varies. Ties not simulated; historical ties count half a win.string |
| $["sourceRetrievedAt"] | 2026-09-05T01:24:43.335236+00:00string |
| $["sourceSha256"] | 08447bfca7733c18e969b2675e0f2220f3d76ee1cb29d8597b4eb004a6a10735string |
| $["sourceUrl"] | https://raw.githubusercontent.com/nflverse/nfldata/c31c50f12973293c5a770c36a53c323419d41383/data/games.csvstring |
| $["specification"]["calibration"] | 2023 out-of-sample residualsstring |
| $["specification"]["development"] | 2018-2022 rolling annual foldsstring |
| $["specification"]["featureNames"][0] | eloDifferencestring |
| $["specification"]["featureNames"][1] | homeFieldstring |
| $["specification"]["featureNames"][2] | scoringDifferencestring |
| $["specification"]["featureNames"][3] | allowedDifferencestring |
| $["specification"]["featureNames"][4] | scoringSumstring |
| $["specification"]["featureNames"][5] | allowedSumstring |
| $["specification"]["featureNames"][6] | marginDifferencestring |
| $["specification"]["featureNames"][7] | winRateDifferencestring |
| $["specification"]["featureNames"][8] | homeGamesstring |
| $["specification"]["featureNames"][9] | awayGamesstring |
| $["specification"]["featureNames"][10] | homeReststring |
| $["specification"]["featureNames"][11] | awayReststring |
| $["specification"]["featureNames"][12] | restDifferencestring |
| $["specification"]["features"] | Earlier completed season-week blocks only; no current-game odds or results.string |
| $["specification"]["initialTraining"] | 2000-2017string |
| $["specification"]["intervalMethod"] | Absolute 2023 out-of-sample residual finite-sample 80% quantile, held fixed for test and forecast.string |
| $["specification"]["lockedEvaluation"] | 2024-2025 annual refit using earlier seasons only; 2024 results enter 2025 training without reselectionstring |
| $["specification"]["mode"] | offline_researchstring |
| $["specification"]["outcomeData"] | Finalized nflverse game outcomes; reconstructed historical research, not strict as-known evidence.string |
| $["specification"]["probabilityMeaning"] | Home victory conditional on a decisive result; ties excluded from probability training and scores.string |
| $["specification"]["publicModelInfluence"] | falseboolean |
Source date: Sep 5, 2026, 1:00 AM UTC
272 scheduled games, 32 team outlooks, probability calibration, score errors and player-production experiments. Frozen preseason research; does not change public rankings.
Original capture date; opening or downloading does not refresh the source. Per-record timestamps take precedence. Definitions may describe unavailable inputs.
JSON bracket paths; [] means every array element. CSV uses concrete zero-based indices. Types are observed, not a promise that future fields cannot change. null is unavailable; empty array/object means an empty collection. Zero is preserved as a measured or supplied numeric value. Source-native units and scoring. Consult metric-definitions for formulas; ambiguous units remain unspecified, never inferred.
sha256 of UTF-8 JSON.stringify(data), excluding this envelope. Source receipt hashes remain inside data. CSV prefixes formula-looking text with an apostrophe; numeric signs are preserved. JSON keeps original strings.
ecbde32e2751aef21c0bf12e3dfcd6ef10021f583afb03d6db2f3bce002f9316
Published observations, approved source-labeled aggregates, research and definitions are provided within each dataset's stated scope. Read each record's source, period, scoring and availability. Research visibility does not promote a model into public rankings. Missing means unavailable, never zero. No account records or provider write access.