Glossary

No jargon left unexplained. If a term shows up on the record page or the methodology page and you're not sure what it means, it's defined here.

Closing Line Value (CLV)

The difference between the price you got and the market's final price right before the game started. Beating the closing line is the earliest real evidence a bet was priced well — it converges roughly five times faster than win/loss record does, because a single result is mostly noise while the market's final number reflects everyone who bet after you. This is why Primo's record leads with CLV rather than profit.

De-vig (De-vigging)

Every posted price has the sportsbook's built-in margin baked into it — quote both sides of a two-way market and the implied probabilities add up to more than 100%. De-vigging strips that margin back out to estimate the market's true, "fair" probability. There's more than one way to do it, and the method matters — see how the model works for why one common method quietly favors favorites over longshots.

Edge

The gap between one book's price and the fair, de-vigged consensus built from every other book quoting the same line. A positive edge means a book is offering a better number than the market as a whole — a signal worth checking, not a guarantee of anything.

Favorite-Longshot Bias

A well-documented pattern where betting markets collectively overpay for longshots and underpay for favorites relative to true probability — part vig, part a human preference for lottery-ticket payouts. It shows up most at the extremes of the price range, which is part of why de-vig method choice matters more on a heavy favorite or a big underdog than on a near-even game.

Implied Probability

What a price says the odds of an outcome are, before removing the vig. -150 implies 60%; +150 implies 40%. Add both sides of a market together and, because of the vig, the total comes out above 100% — that gap is exactly what de-vigging corrects for.

Kelly Criterion (Quarter Kelly)

A formula for sizing a bet as a fraction of bankroll, scaled to how big the edge is — bigger edge, bigger stake. Full Kelly only maximizes long-run growth if the probability estimate behind it is exact, which a betting-market estimate never quite is. Sizing at a quarter of full Kelly punishes a bad estimate far less severely while still scaling stakes to conviction.

Line Movement

How a price changes between when it's first posted and when the game starts, as new information — injuries, lineups, weather — and betting volume push the market. Tracking movement over time, not just the final number, is how a closing-line archive like this one gets built in the first place.

Moneyline

A bet on which team simply wins the game, with no point margin involved. Priced as American odds: a negative number is the favorite (risk that much to win 100), a positive number is the underdog (risk 100 to win that much).

Overround (the Vig, or Juice)

The house's built-in edge, expressed as how far a market's implied probabilities run above 100%. A -110/-110 two-way line implies roughly 52.4% plus 52.4% — about 104.8% total — and that extra 4.8 points is the vig. It's exactly what de-vigging exists to strip back out.

Parlay

A single ticket combining multiple bets, all of which have to win for the ticket to pay out. Payouts multiply with every leg added — but so does the compounded vig, which is why more legs tilts the math further against the bettor, and why correlation between legs matters more here than on a single straight bet. See the Parlays page for how conservative and higher-variance builds are kept separate rather than blended together.

Push

A bet that ties — the final result lands exactly on the number — and the stake is simply returned. A push counts as neither a win nor a loss in any honest record.

Regression to the Mean

The statistical reality that a small sample of outlier performance — a hot streak, a bench player's lucky week — is mostly noise and tends to move back toward the true underlying rate as more data comes in. Trusting a small sample at face value overstates real ability. See how the model works for exactly how much of each stat gets trusted at different sample sizes.

Run Line (Point Spread)

A bet on the margin of victory rather than simply who wins. In baseball it's almost always set at ±1.5 runs; in football and basketball, the spread is set to try to make both sides equally attractive to bet.

Sharp Book

A sportsbook — Pinnacle and Circa are the standard examples — known for taking large, informed action and setting prices that other books often follow, rather than reacting to public betting patterns. A sharp book's price tends to carry more information than one that mostly mirrors the rest of the market.

Total (Over/Under)

A bet on the combined final score of both teams relative to a posted number, independent of which team actually wins.

Unit

A fixed, self-defined stake size used to describe bet sizing without revealing — or implying — actual dollar amounts. "Risking half a unit" scales to whatever bankroll someone is actually using, which is how staking gets discussed in the abstract, including in Primo's own record.