How THE LINEUP verifies public +EV results.
The public results workflow is designed to separate market-based +EV decisions from model projection context, preserve source labels, and make the settled record inspectable after each game.
Fair-price check
EV+ compares a sportsbook's offered price against fair odds from sharp or historically validated market sources. Projection models do not create +EV picks.
Point-in-time history
The promotion-grade record uses an immutable first-seen sportsbook price captured before the exact event start. Legacy rows without that immutable price remain visible only as audit context; they cannot support advertised ROI. Rows first written after kickoff, scanner backfills, synthetic prices, and later duplicate quotes are excluded.
Stronger receipt subset
Newer picks also bind selection identity, sportsbook price, policy, quote time, publication time, and event start into an immutable receipt. Receipt coverage is shown as a verification tier, not substituted for the larger point-in-time record.
Settlement review
One flat unit is graded at the first archived pregame price, or at the receipt-bound price in the receipt view. The page exposes result, units, ROI, sample size, calendar gaps, drawdown, and CLV coverage.
Strategy approval
A historically interesting bucket is not an approved strategy. A rule must be frozen before use, survive chronological validation and untouched holdout, then collect a separate forward-shadow sample before it can be activated on the EV Board.
Portfolio versus sport proof
A positive cross-sport portfolio does not prove that every sport or market is profitable. Portfolio evidence is tested for sport, sportsbook, fair-source, and market concentration; individual sport and market claims require their own frozen rule and later validation.
Model scorecards
Each sport, market, model head, and artifact version keeps its own accuracy or calibration cohort. Model betting ROI remains unavailable until that output has a separate frozen wagering rule tied to an executable price.
What the public record can and cannot prove
A public betting record can show what was logged, how it was settled, which source path produced it, and whether the market moved favorably by close. It cannot guarantee future outcomes or replace sportsbook access, bankroll rules, or user discipline.
- EV+ source
- Offered price versus fair odds
- Model source
- The NBA Projection Model is a separate LightGBM v7_beta system with 157 configured features for player stat context; other sports expose model, shadow, or statistical projection context only where supported.
- Public surface
- Results ledger with filters and source labels
- User check
- Date range, scope, source, CLV, and settlement status
After the methodology
Review any offer after the proof rules.
You reviewed how THE LINEUP separates EV+ fair-price checks, projections, source labels, and settlement review. If an eligible partner account fits your next step, use code LINEUP from the disclosed cards below.
FAQ
Are THE LINEUP +EV picks created by the projection model?
EV+ compares a sportsbook's offered price against fair odds from sharp or historically validated market sources. Projection models do not create +EV picks.
What does the projection model do on results pages?
The projection model provides separate player-stat context. It can help users understand a player range or matchup, but it is not the source of the EV+ fair price.
Why do source labels matter?
Source and receipt labels distinguish a point-in-time pregame observation from a cryptographically receipt-verified publication or a later research reconstruction. Those evidence classes must not be blended.
What should users inspect before trusting any result?
Inspect the date range, number of active days, sample size, policy cohort, exact sportsbook quote, publication and event timestamps, settlement status, CLV coverage, and any validation exclusions. A high ROI from a small sample is factual but not yet durable.
Does a profitable historical market automatically appear as a proven strategy?
No. Discovery results can be useful research, but they stay WATCH until a frozen rule passes later chronological windows and forward observation. The page reports zero approved strategies when none have passed.
Does a positive portfolio ROI mean every sport was profitable?
No. The portfolio result is an aggregate. The page separately reports sport breadth, concentration, clustered confidence intervals, and remove-one-group checks. A sport or market is not called proven until a separately frozen strategy passes chronological and forward validation.