How the Futures screen prices a whole field

What the Futures screen is

A futures market has one winner from a whole field: 32 teams for the Super Bowl, 239 for the college basketball title, 110 golfers for the Masters. The Futures screen puts every book's price on each runner side by side and prices the whole field at once: a fair probability for every runner, the best price at your books, the EV of that price and each book's margin on the market.

Futures are where books keep the most margin. On the 2026-10-13 capture the books held 5-26% on the Super Bowl market (most US books 19-26%) and 46-75% on the Masters, against 4-5% on a typical moneyline. Shopping the price matters more here than anywhere.

The problem: books list different fields

On a two-way market every book quotes both sides. On a futures market they don't. On the college basketball title one book listed 91 teams and another 204. A book that lists only the favourites can't be devigged to 100%: its runners don't carry all the probability. Treating them as if they did makes every runner look more likely than it is.

So we devig each book to the share of the field it actually lists, its covered mass. That share comes from the fair prices, and the fair prices come from the devigged books, so we solve for both together (a fixed point; it settles in a few rounds):

Worked example: a book that lists only two runners

A three-runner field. Book A lists everyone at 55%, 33% and 22% implied (a 10% margin on a true 50 / 30 / 20). Book B lists only the first two, at +100 and +200: 50% and 33.3%, 83.3% in total.

Devigging B to 100% would say X is a 60% shot. Counting only the share of the field B lists, X is about 49-50%, and B's real margin on its two runners is a few percent, not the "negative" 16.7% its raw total suggests.

Power, Shin or multiplicative

Books shade long shots hardest: the favourite-longshot bias. That matters far more on a 100-runner field than on a two-way line.

• Power (our default for futures) raises every implied probability to one power k so the field sums to its mass. Long shots give up the most margin.
• Shin models a share of insider money and also leans the margin onto long shots. On most fields it lands close to power.
• Multiplicative scales every price by the same factor. It overstates long shots on big fields; we show it for comparison.

The real 2026 World Series market on 2026-10-13 shows why the choice matters. DraftKings had Milwaukee +105, the Dodgers +350, Cleveland +400 and Tampa Bay +475: 108.4% in total, an 8.4% hold. Across every book, the fair chances were:

Pinnacle had Milwaukee at +112, the longest price on the board. Against the fair built without Pinnacle, that price was −1.3% EV with power, −2.1% with Shin and −4.3% with multiplicative. The method moved the EV by three points on a four-team field; on a 100-runner field it moves long shots far more. The Devig menu lets you check all three before you trust an edge.

Exchanges and prediction markets

• Exchange prices (Betfair Exchange on the golf majors today) show the quoted back price with a net line under it: what you get after the exchange's commission. The best price and EV use the net price.
• For a prediction market's fair price we use the middle of its bid and ask, not the ask: an exchange's spread is not a margin. A contract with no bid, a wide spread or little money at the ask is thin: shown, but it never builds the fair and is never the best price.
• Kalshi's futures contracts are not shown yet. The engine already prices them this way, and they will appear once we can publish them.

How to read the screen

• Market picker in the top bar: every open futures market, grouped by league, with the runner and book counts. The screen opens on the market with the most books. A market leaves the list once it settles or no book lists it.
• Best: the highest net price at your books (the Book selector). An offshore book is never the best price unless you picked it in My books. Thin, stale and off-market quotes never count.
• Fair %: the no-vig chance from every book, whatever you pick in the Book selector, with your book weights.
• EV: the best price against the fair rebuilt without that book. It needs two other books listing the runner. Runners under 0.5% fair show none: books disagree three to ten times over on prices that long.
• Book columns: each book's price, with its hold on the whole market under its logo. Green fill is the best price; green text beats fair; amber is a net price after fees, a thin quote or a flagged one.
• Age: when the best price was last checked. Book prices refresh every 6 hours; amber after 7 hours, red after 24.

Click a runner (or press Enter on it) to see every book's price, net price, EV, dollars at the ask and age. On a phone, each runner is a card with its best price, fair chance and a strip of every book.

Off-market and check flags

When at least three books list a runner, each quote is compared with the middle of the other books' quotes. A price more than three times away (and more than a point apart) is marked off-market and left out entirely: on the Masters, Betfair's exchange showed 2.0 on golfers the books priced at 1-4%. A price more than 1.75 times away is marked check: you still see its EV, but it never builds the fair and is never the best price. A stale line or a typo usually explains it.

Pitfalls

• Your money is locked up. A +5% EV future that settles in eight months earns far less per year than it sounds.
• Long-shot EV is noise. That's why runners under 0.5% fair show no EV.
• Limits are low. Books cap futures stakes, especially on long shots.
• The field assumption. We treat every runner any book lists as the whole field. Runners nobody lists are assumed to have almost no chance.
• Fewer books than game lines. Most futures markets have 4-10 books, so one book moves the fair more than it does on a moneyline.