An NFL MVP market needs more checks than a single-game YES/NO contract. Before comparing candidates, identify the season, the award definition, the available outcomes and what each contract pays under its own rules. This guide uses hypothetical prices; it does not quote the current MVP race or demonstrate a profitable strategy.
Start with the actual contract rules
Open the current market on Kalshi and read its settlement source and exceptional-outcome rules. Do not infer the ticker, tie payout, eligibility after a trade or injury, or treatment of unlisted candidates from another season's market. If you compare the market with a sportsbook or another exchange, their definitions must match.
A basket calculation requires a complete set of mutually exclusive outcomes that covers every possible settlement. A list of popular candidates may not satisfy that requirement. Check whether an “other” outcome exists and what it covers.
What a 38c quote means
Buying a contract for 38c with a $1 winning payout has a 38% gross break-even win rate, before costs. Its equivalent gross decimal payout is 1 / 0.38 = 2.632, or approximately +163 American odds.
That is a payout conversion, not evidence that the candidate's true probability is 38%. A last trade, bid, ask and midpoint may differ. Use the ask and available size when studying a purchase, and the bid and size when studying an exit. A 3c quote likewise does not establish that 97% of such contracts actually lose; that requires an outcome study of a defined cohort.
Basket prices and overround
Suppose a hypothetical complete set has exactly one $1 winner. Buying one contract on every outcome costs $1.20 in total, all orders fill, and there are no other costs. The basket pays $1, so its loss is $0.20, or 16.67% of the $1.20 stake. A price sum of 120% is a 20-percentage-point excess over 100%, not a 20% stake loss.
Equal cash bets on candidates are a different portfolio: they buy different contract quantities, so the payout depends on which candidate wins. Neither calculation establishes the economics of a real MVP market.
A sum below $1 is only a candidate for investigation. It needs compatible and exhaustive settlement rules, simultaneous executable asks, sufficient size on every leg, fees and a plan for partial fills. Summing last trades or mixing bids and asks does not produce an executable basket.
Depth and volume answer different questions
Trade volume reports activity under the venue's unit definition. It does not show how much you can buy at the current ask. Inspect the ladder for your intended quantity and calculate the volume-weighted acquisition price. A top quote can disappear before you trade.
For historical analysis, retain quote and print timestamps separately. Comparing a print with a later book cannot reliably identify whether it lifted the earlier ask. Our Kalshi tape walkthrough explains the timestamp, capture-gap and size-unit checks; it is not an MVP-specific liquidity study.
Testing a probability model
In a hypothetical 38c contract, a model estimate of 45% gives 0.45 - 0.38 = 7c gross expected margin per share. If the actual probability is 31%, the margin is −7c. Both statements assume a $1 payout and exclude costs.
That sensitivity is why a model disagreement is a hypothesis. Declare the prediction time, collect multiple seasons without future information, inspect calibration and account for the shared constraint that only one candidate can win. One award outcome per season limits the independent evaluation sample. Our Kelly formula guide covers sizing assumptions; a smaller position does not validate the probability.
Earlier claims we withdrew
The previous version described a seasonal window and a 5–7% threshold as if they identified tradable edge. It also made unsourced claims about typical liquidity and retail losses. Those claims are removed. The following corrections were already published on September 25 and remain part of the editorial record; their internal-study figures are dated disclosures, not new October measurements.
Correction (2026-09-25): "most extractable" describes market structure, not a demonstrated result — our own live bought-side calibration gap is negative in every sport we trade, roughly −7pp to −31pp (internal fill study, September 2026). Read this as a description of when the market is most responsive to new information, not a claim that we or anyone else have shown edge here.
Correction (2026-09-25): the 5-7% figure is an illustrative risk-control threshold for the sizing math below, not a demonstrated edge — our own live bought-side calibration gap is negative in every sport we trade, roughly −7pp to −31pp (internal fill study, September 2026).
Use the odds converter for payout arithmetic and free schemas to inspect the data fields before designing a historical study. Neither resource proves an MVP trading opportunity.