Operator Vig and Fee Index

Every row is a verbatim sentence read from the named source's own page through our fetch service on the date shown, hash-pinned to the stored read (the sha of the page's visible text; the raw page is mirrored to R2).

2 of the 25 named sources that could be read published a usable figure; 23 published none and are recorded as not published, never filled in; 9 could not be read (blocked, dead or unreachable) and count nowhere.

The headline figures (median, minimum, maximum, quartiles) are over ONE figure per source in % (a source's median where its page yielded several accepted sentences), so N counts sources, never sentences; 99 candidate sentence(s) were refused by a reviewer with the reason recorded.

The fuller derivation this dataset is chartered to grow into (not yet what the figures above are; they are each source's own published figure as read): one row per operator per market type: the explicit fee where the operator publishes one (Polymarket's taker and maker table, Kalshi's fee formula, an exchange's commission percentage, PredictIt's profit and withdrawal fees) and - for every sportsbook, none of which publishes a margin - the OVERROUND computed from that operator's own posted prices on identical markets, which is the only way the number exists at all.

The derivation nobody publishes: the vig by operator, sport and market type side by side with the fee, restated as one comparable cost of taking a position, and the fair no-vig probability our calculator returns beside the operator's implied one; plus the measured share of operators that publish any cost figure at all, which this harvest already answers - of the operators read, the CFTC-regulated venues and the exchanges publish a fee schedule and not one traditional sportsbook publishes a margin

99 candidate sentences on these pages were refused by a reviewer; each refusal and its reason is recorded in the review ledger.