CashCord / Public record preview

Show the whole record.

This is the proposed reporting method for CashCord’s public picks and creator profiles. The durable publishing service is not live yet.

What belongs in a pick record

Each entry needs a stable record ID, creator, sport, season, event, scheduled start, exact selection and market, line where applicable, sportsbook, decimal price, stake in units and a source reference.

Source posting time and CashCord’s own receipt time are separate fields. A historical screenshot does not establish that CashCord received or published the selection before the game.

Before-start and retrospective records

A future receipt service must stamp and preserve entries before the event begins. Retrospective imports will be labeled separately and excluded from ranked results. The current preview has neither kind of record.

Corrections preserve history

The proposed service appends settlement events. A correction must refer to the event it replaces and explain why it changed. It must not overwrite the original selection, odds, stake or settlement history.

Application-level append-only controls and a receipt timestamp are not an independent audit or a guarantee against database-owner tampering. Those stronger claims would require additional evidence.

How the metrics are calculated

Record and sample size: Wins, losses, pushes and voids are shown separately. Pending picks are excluded from settled performance. Win rate uses wins ÷ (wins + losses).
Net units: A win earns stake units × (decimal odds − 1). A loss loses the stake units. Pushes and voids contribute zero net units.
Unit-weighted ROI: Net units ÷ the sum of settled stake units, including refunded stakes, × 100. This is a unit-weighted measure; it is not a verified dollar return when a creator changes the cash value of a unit.
Flat-one-unit comparison: Every settled pick is assigned one hypothetical unit. A win earns decimal odds − 1, a loss loses one, and a push or void earns zero. Flat ROI divides that net by all settled picks, including refunded stakes.
Average price: The arithmetic mean of posted decimal odds across settled picks. It is not an average American line or a forecast.
Ranking eligibility: At least 50 wins or losses in the selected sport, bet type and period. This proposed floor limits tiny-sample rankings; it does not establish statistical significance or future profitability.
Time windows: The last 7 and 30 days select picks by CashCord receipt time, ending at the dataset’s stated snapshot time. Settlements known by that snapshot are included. Season filtering uses the stored sport-season label, not an assumed calendar-year boundary.

What a public record cannot prove by itself

Selection bias, missing bets, uncertain source timing, unavailable odds and correlated picks can limit interpretation. A profitable sample is not a guarantee of future results. The site will not label this preview independently verified.