Sportsbooks vs. prediction markets, for futures specifically

A sportsbook is a counterparty

When you bet a futures market at a sportsbook, the book is on the other side of it. They set one line, they hold the liability if it hits, and they manage that liability by adjusting the price, limiting action, or — when a market stops being worth the trouble — pulling it outright. A team that's mathematically drawing dead has nobody left betting the over and nobody left worth taking the under from at a meaningful price, so the book has no reason to keep quoting it. Pulling the market isn't a bug. It's a risk manager deciding the position isn't worth carrying.

An exchange isn't carrying anything

Kalshi and Polymarket don't take the other side of your bet — they match you with whoever else is trading that contract. There's no book managing directional risk, so there's no moment where the risk stops being worth carrying. As long as the event hasn't resolved, someone can quote a price, even if that price is 1¢ because everyone agrees the outcome is close to settled. The market doesn't get pulled because nobody's job is to decide it's not profitable anymore.

What that means for a number you're tracking

Mark-to-market only works if a mark exists. When a sportsbook pulls a futures market mid-season, there's no live price to value your ticket against — you're stuck with the number from the day you bet it, or a guess. When the same market is also trading on an exchange, there's still a real, tradable number all the way to resolution, and it's one you could actually act on if you wanted to hedge or cash out early, not just look at.

It works the other direction too. A team catching fire late in the season can see its exchange price run up well before a sportsbook gets around to reposting a market it pulled months earlier — the exchange never stopped pricing it, so there's no re-opening lag to wait out.

It isn't a strictly better number

Exchange prices come with their own tradeoffs. Futures contracts on Kalshi and Polymarket are often thinner than a sportsbook's headline markets, so a price can move on relatively little volume and a large order can move it further than the "true" probability would suggest. There's also no vig baked in the way a sportsbook's two-sided line has one — cents are closer to a raw probability, which is useful, but it means the price reacts to thin order flow in a way a book's managed line doesn't.

Why we track both

This is the actual reason Kalshi and Polymarket sit next to DraftKings and FanDuel on every board: not because an exchange price is better, but because it's still there when a sportsbook's isn't. If your ticket's book pulls the market, your mark-to-market value doesn't have to go blank — it falls back to whatever's still quoting.