A useful explanation shows what is known, when it was known, and what the model leaves unanswered.
Design preview · MLB, NFL and NCAAF model views distinguish current source availability from saved records. Saved records are dated snapshots; loading them does not renew their source or quote times. SIM output is illustrative. Missing inputs, provenance and settlement remain unavailable.
01 / DIFFERENT SPORTS, DIFFERENT OUTPUTS
Let the evidence set the terms.
Connected feeds provide game data and market observations. A separate model artifact provides a decision, and a separate settlement record provides its grade. One cannot stand in for another.
NBA, men’s Division I NCAAB, NHL, and golf pages are design examples. Their probabilities, factor contributions, fictional players and settlements are SIM fixtures, not an approved or fitted basketball, hockey or golf model. No basketball, hockey or golf decision producer is connected. Scores, schedules and odds observations cannot create a model pick or settlement.
Golf form charts use authored strokes-gained examples per round. Course-fit explanations are illustrative. Real golf records carry provider timing and visible data conflicts; they never inherit sample probabilities. Golf odds comparisons require matching tournament, player, market and settlement rules.
MLB
A probability, with context.
MLB records can carry a model probability, a market reference, coverage tier, and supporting signals. A missing probability or market reference is shown as unavailable.
The market reference is absent when the two-way price basis is missing, stale, supplied by fewer than five sources, or ambiguous for a doubleheader. A difference is not calculated when either side is missing.
The NFL artifact carries a predicted cover margin at the T−24 decision line and a 90% residual interval. It does not supply a win probability or market edge.
Prediction-error experiment. The model has no demonstrated advantage over the market line: on the holdout its error was not distinguishable from the market baseline (the 95% interval crosses zero). Nothing here is a recommendation to act.
NCAAF records carry a home-win probability and, when supplied, a naive comparison probability. Prior-season team support and the home or neutral-site setting provide input context. They are not a measured contribution from each factor.
An abstained record keeps its reason without an invented prediction. A final outcome is shown only when its settlement matches the exact decision. An observed price alone is not a model selection.
The NFL decision line is 24 hours before kickoff. A pending state means the receipt is awaited; it does not imply a lean. A pregame recorded output becomes frozen history at kickoff, or when a usable kickoff clock is missing.
The input capture time and decision as-of time stay beside the record. A later quote does not rewrite the original input snapshot. An NFL board older than three hours is stale, and its picks are not presented as current.
NCAAF records preserve their source observation and publication times; the NFL timing rule does not establish their freshness. Saved records and the preview’s SIM boards are explicitly labeled. Freshly retrieving an old artifact does not make its contents current.
03 / READING THE STATES
A decision to wait is visible, too.
Current pick / recorded
A valid decision exists before the game starts. Its output and provenance are available.
Pending · T−24
Awaiting the 24-hour receipt. No decision numbers are shown.
Football abstained
The producer recorded an abstain reason. No numbers or hypothetical selection are shown.
MLB · not ranked for coverage
Low coverage is distinct from a model refusal. Available model numbers may still be shown, with the coverage reason.
Withheld
Publication is withheld; the reason remains visible and decision numbers stay hidden.
Kicked off · frozen
The pregame output is historical. It is no longer current.
Final / graded outcome
A grade is shown only when the settlement record supports it. A final score alone does not grade a pick.
Conflict
Inconsistent records require reconciliation. No current selection is displayed.
04 / FOOTBALL ABSTAIN REASONS
Say why.
Producer reason text is preserved. The currently emitted NFL abstain reasons are:
“The schedule carried no usable kickoff clock.”
“The scoring pass ran after kickoff, so no decision was recorded.”
“A retry after kickoff found no matching earlier record, so nothing was recorded.”
For a schedule conflict: “The schedule changed after the decision froze; the frozen record is withheld.” An unknown code is identified as unknown, without inventing a reason.
05 / KEEP THE RECORD HONEST
Only count what is graded.
The separate MLB aggregate archive counts only the published eligible cohort: graded records equal eligible records and wins plus losses. Historical quarantine is separate. Its reason counts overlap, so they are never summed. Voided or postponed exclusions do not identify individual game dispositions in this aggregate feed. Coverage abstentions on the forecast board remain separate from graded performance.
A win rate needs eligible graded observations. Flat-one-stake source ROI needs eligible priced observations and uses the source’s profit divided by that priced count. Zero denominators mean unavailable, never 0% performance. Individual grades, voids, dated performance and settlement histories are not reconstructed from team names or dates.
In that MLB archive, published output is recorded with freshness unverified. The model generation time, feed generation time and retrieval time are separate. The game file becomes historical after its recorded start; elapsed time alone never establishes a final result. Model version, serving receipt and captured spread, total, moneyline, book and line timestamp are unavailable when absent. An artifact digest is not a model receipt. Stable game addresses retrieve only the selected feed, not an archive.
The separate MLB archive price snapshot has no canonical game identity, quote observation time or timezone on event times. It stays unverified and separate from the source odds comparison. No current best-price rank, pick join or model edge is inferred from these individual prices.
The MLB and NFL model-record feeds do not currently supply individual settlements. NCAAF records can carry outcomes bound to the exact decision; the saved September 10, 2026 snapshot contains four such outcomes. They stay separate from unsettled predictions and abstains. A score alone does not grade a model record.
The current model feed supplies no aggregate evaluation, so its record pages do not infer model-quality or economic performance summaries. Any descriptive evaluation in a saved snapshot applies only to that snapshot and its stated sample, not to the current board.
We do not turn a missing price into zero, an incomplete feed into a prediction, or an illustrative example into a track record.
Understand the game. Keep the uncertainty.
BetMetrics is not betting advice. Outcomes remain uncertain. Nothing on these pages is a promise or a recommendation to act.