CourtVision / Measurement lab
Compare measurements, inspect rows, and export your selection. Descriptive exhibits, each with a source JSON link. Analytics hub · Methodology
Compare pitch usage with the center and spread of measured velocity.
Median velocity19 published rows; 16 numeric
Published definition
19 rows
Summaries describe the filtered rows with equal row weight. Row counts are not independent game or event counts.
Showing 16 of 16 matching numeric rows. Open a row to inspect all its fields. Rankings apply only to the published subset.
693,037 pitches from the published 2025 Statcast pull; exact observation dates are not published.
Velocity and pitch-mix denominators differ. Missing velocity rows can reflect omitted groups below the 20-observation floor or missing pitch type or speed; they do not prove no pitches were measured. Pooled pitch-type velocity variation is not pitcher consistency or quality, and pitch-type rates do not establish prediction accuracy.
Full method and source artifactPublished experimental metrics
These are the project's published metric formulations. Prior-art labels include incremental contributions and novel packaging; inclusion does not establish universal originality or predictive value.
soccer_intl completes half its pre-game line motion by 2.7h before tip; nba not until 4.8h. tennis, wnba move most of their line >6h before tip.
bucket_move=n*mean_abs_move per horizon bucket; G(h)=cum far->near / total; half_life = h where G(h)=0.50 (interpolated).Data integrity
This timing measurement is revision 2, rebuilt on the segment-clean corpus (2026-09-17). The stored corpus contains 178 MLB and 27 international soccer games. Individual measurements can use fewer games after eligibility checks; read their own denominators. Revision 1 values are withdrawn and kept in the timing regeneration receipt.
MLB: LCF 0.7368 at 3 runs (inning clock; 174 usable score paths). INTERNATIONAL SOCCER: LCF 0.5994 at 1 goals (minute clock; 26 usable score paths). Thresholds and clock units differ; these values do not rank sports.
LCF = decided_clockfrac_median at the non-masked threshold whose decided_frac_of_games is closest to 0.50 (higher = contested later).Most one-star-fragile: DEN (Nikola Jokić, delta_winprob 0.5822). Two estimators name the same #1 player for 1/30 teams (directional cross-check, different seasons).
LBI_a = max_player (p_win_with - p_win_without) [Elo on/off]; LBI_b = max_player (win_rate_active - win_rate_missed) [raw]; agreement = same #1 player.Data integrity
This timing measurement is revision 2, rebuilt on the segment-clean corpus (2026-09-17). The stored corpus contains 178 MLB and 27 international soccer games. Individual measurements can use fewer games after eligibility checks; read their own denominators. Revision 1 values are withdrawn and kept in the timing regeneration receipt.
MLB / 1: MFP 0.051 (3589 checkpoint observations); MLB / 10: MFP 0.411 (46 checkpoint observations). INTERNATIONAL SOCCER / 0: MFP 0.6779 (150 checkpoint observations); INTERNATIONAL SOCCER / 90: MFP 1.4861 (33 checkpoint observations). These endpoint snapshots have different observation counts and do not establish an in-game trend.
MFP(t) = ((naive-market)/naive - (naive-model)/naive) / max(entropy_market_bits, 0.15)Data integrity
Composed on 2026-07-25 from market_overreaction and market_disagreement_profile. Both inputs were rebuilt at revision 2 on the segment-clean corpus, so every figure here is still the revision 1 composition and is under review until it is recomposed. Stale input: market_overreaction, market_disagreement_profile.
soccer_intl: overshoot 0.286 but model matches or improves on the reference only 22% of the time at max disagreement -> OHG 0.081 (a failure to harvest, not an advantage); mlb: overshoot 0.072 but model matches or improves on the reference only 38% of the time at max disagreement -> OHG 0.009 (a failure to harvest, not an advantage)
OHG = (n-weighted mean |moved_to - outcome| over {3-6pt,6-10pt,10pt+}) * (0.5 - model_closer_rate at >=.10 disagreement).Across 511,807 adjacent pitch pairs in the 2025 Statcast season a pitcher throws the same pitch type twice in a row 0.3533 of the time against the 0.3166 his own count-conditioned mix implies, an excess of 0.0367 (95 percent interval 0.0348 to 0.0385). The excess is largest when he is behind (0.0476, n=180,871) and smallest at an even count (0.0165, n=127,054), and at the same count the repeated pitch draws MORE swings and misses than the switched pitch (+0.0098, interval 0.0079 to 0.0120).
excess = mean over adjacent pairs of 1{type == previous type} minus (n(pitcher, count, previous type) - 1{type == previous type}) / (n(pitcher, count) - 1), by count class, raw count, previous pitch family, pitcher.Across 4793 NBA games the home win frequency rises from 0.4944 when the visitor is the better-rested side by two or more days (n=180) to 0.6409 when the home side is (n=259), a 0.1465 spread, while the two symmetric-congestion groups show no visible difference: at an equal differential both on a back-to-back win 0.5510 (n=245) against 0.5518 (n=2394) when neither is.
P(home win | rest_days_home - rest_days_away in cell), and on the priced subset the same frequency minus the mean devigged pregame market forecast for that cell.Data integrity
The Schedule Fatigue Tax ORtg estimate is withheld. Its cited -1.73 result measures scoring margin (no-rest minus rested), not offensive rating per 100 possessions. The historical artifact misstates both the unit and subtraction order; its ORtg-derived values cannot be interpreted as validated measurements. Schedule counts and frequencies remain descriptive. The review date is not an observation date.
Under review: the claimed ORtg fatigue tax used a scoring-margin contrast as its anchor. Published team-season back-to-back counts, frequencies, and the separate raw descriptive composite remain available; the ORtg effect is withheld.
ORtg-derived SFT is under review: the cited -1.73 is a scoring-margin contrast (no-rest minus rested), not an offensive-rating effect. The separate descriptive composite is b2b_freq * (-0.315 composite/36).Across 52078 MLB team-starts from 2010 to 2021 the starting pitcher's rotation day barely moves the win frequency: 0.5027 on four days rest (n=24333) against 0.5023 on five (n=18222) and 0.4999 on six or more (n=9014), and measured against the recorded closing reference forecast those gaps are -0.0019, -0.0003 and 0.0082, each inside its own 95 percent interval of zero.
P(team win | starter days of rest in bucket) over team-starts, and the same frequency minus the mean devigged closing reference forecast for that team, reported overall and within reference-forecast bands.