Betting glossary
75 terms, defined in plain English with the math where it matters — from odds formats and the vig to CLV, Kelly staking, and key numbers.
Odds & prices
American oddsThe odds format standard at US sportsbooks, written with + and −. A minus number is what you must risk to win $100 (−150 = risk $150). A plus number is the profit on a $100 stake (+150 = win $150). Minus marks the favorite, plus the underdog.ChalkSlang for the favorite, especially a heavy one; a “chalk bettor” habitually backs favorites. The term survives from pre-digital racetracks, where the most-bet horse's shortening odds were rewritten on chalkboards. Chalk-heavy betting produces high win rates and, frequently, slow losses to the vig.Decimal oddsAn odds format showing the total return per $1 staked, stake included: odds of 2.50 return $2.50 (a $1.50 profit). Standard in Canada, Europe, and Australia. Payout math is one multiplication (stake × odds), and implied probability is simply 1 ÷ odds.Even moneyA price that pays profit equal to your stake: +100 in American odds, 2.00 decimal, 1/1 fractional — an implied 50% probability. Some books display it as −100. A fair coin flip at even money has exactly zero expected value before any vig.FavoriteThe side the market prices as more likely to win, carrying a minus American price (−150) or decimal odds below 2.00. Favorites win more often but pay less than the stake risked — which is why a high win rate built on heavy favorites can still lose money.Fractional oddsAn odds format, common in the UK and horse racing, expressing profit relative to stake. 5/2 wins $5 for every $2 staked (+250 American, 3.50 decimal). Fractions below one, like 8/11, are favorites: you win $8 for every $11 risked.Implied probabilityThe win probability a betting price represents. For decimal odds it is 1 ÷ odds (2.50 → 40%); for −110 American it is 52.4%. Posted prices include the sportsbook's margin, so a market's implied probabilities sum to more than 100% until the vig is removed.LongshotA big underdog at long odds, such as +700 or more. Longshots cash rarely but pay heavily. Markets systematically overprice extreme longshots relative to their true chances — the longshot bias — making the longest odds some of the worst-value prices on the board.Pick'emA game the market rates as a true toss-up: the point spread is zero (or absent) and both sides are priced near even money, typically −110 each. Your side simply has to win the game; the −110 pricing on both sides is the book's margin at work.Underdog (dog)The side priced as less likely to win, carrying a plus American price (+150) or decimal odds above 2.00. Underdogs pay more than the stake because they win less often. Small underdogs on the moneyline historically win outright more often than casual bettors expect.
Markets & lines
Closing lineThe final odds available when an event starts. By close, the line has absorbed the most information and money, making it the market's sharpest estimate of the true probability. Beating the closing line consistently — positive CLV — is the strongest known predictor of long-term profit.FuturesLong-horizon bets settled at season's end: championship winner, division winner, season win totals, awards. Futures markets carry much larger margins than game lines — implied probabilities on a championship board often sum to 130%+ — and they lock up bankroll for months.LimitThe maximum stake a sportsbook will accept on a market. Limits start low at the opener and rise toward game time as the book gains confidence in its number. A book raising limits is inviting volume; a book cutting your personal limits is telling you you're winning.Line movementThe change in a betting line between open and close, driven by money (bets shifting the book's exposure) and information (injuries, weather, lineups). Direction and speed of movement hint at which force is at work — and whether the value has already been captured.Live betting (in-play)Betting on a game while it is being played, at odds that reprice continuously with the action. Live lines are generated algorithmically, carry higher margins than pre-game lines, and settle fast. The speed advantage belongs to the book, not the bettor.Market consensusThe aggregate price across many sportsbooks — commonly the median, de-vigged to fair probabilities. One book can post an off number; the consensus of a dozen books is hard to beat and serves as a practical stand-in for the true probability when hunting value.MoneylineA bet on which team wins the game outright, no point spread involved. Prices carry the probability: a −200 favorite must be risked at 2-to-1, a +170 underdog pays 1.7-to-1. The moneyline and spread are the same opinion split into different risk shapes.Off the boardA market a sportsbook has pulled from betting — no prices offered. Books take games off the board around major injury uncertainty (a starting quarterback's status), extreme news, or integrity concerns, and repost once the uncertainty resolves, usually at meaningfully different numbers.Opening lineThe first price a sportsbook posts for a market. Openers are set by models plus judgment, carry lower limits, and contain more pricing errors than later numbers — which is why sharp bettors attack openers, and why books let sharps shape the line early.Point spreadA handicap that levels a mismatch: the favorite must win by more than the spread (−6.5 means win by 7+), the underdog covers by losing by less or winning. Both sides are usually priced near −110, with the book's margin in the price rather than the number.Prop bet (proposition)A bet on something other than the game's outcome — most commonly a player statistic: passing yards, points, strikeouts, shots on goal. Props are priced with less attention and lower limits than main markets, which makes them softer, and prices vary more across books.Reverse line movementWhen a line moves opposite to the majority of bets — say 75% of tickets are on the favorite, yet the price moves toward the underdog. It suggests the smaller, sharper money is on the other side and the book respects it more than the public volume.SteamA sudden, synchronized line move across many sportsbooks at once, typically triggered by respected (sharp) money hitting the market. “Chasing steam” means betting the same direction at books that haven't moved yet — a race measured in seconds, and one books actively defend against.Total (over/under)A bet on the combined points scored by both teams: the book posts a number (say 44.5) and you bet Over or Under it. The teams' identities don't matter to the result — only the sum. Totals move on pace, weather, injuries, and scoring-environment news.
The book's edge
Break-even win rateThe win percentage a price requires just to neither win nor lose money — equal to the price's implied probability. Even money needs 50%; −110 needs 52.4%; −200 needs 66.7%. Any honest evaluation of a betting record starts by comparing the actual win rate to the break-even rate of the prices bet.De-viggingRemoving the sportsbook's margin from posted prices to recover the book's true probability estimate. In the simplest method, divide each side's implied probability by the market's total: a −110/−110 market de-vigs to 50/50. De-vigged consensus prices are the standard baseline for finding +EV bets.HoldThe sportsbook's margin on a full market, measured by summing both sides' implied probabilities: a fair market sums to 100%; a real one sums higher, and the excess is the hold. A −110/−110 spread holds about 4.5%; parlays and futures hold far more.No-vig (fair) oddsThe odds a market would offer with the bookmaker's margin stripped out — the price matching the de-vigged probability. If −110/−110 de-vigs to 50/50, the no-vig price is +100 each side. Any available price better than the fair number is, by definition, positive expected value.OverroundThe amount by which a market's implied probabilities exceed 100% — the same quantity as the hold, expressed from the odds side. A two-way market at −110 each shows 52.4% + 52.4% = 104.8%, an overround of 4.8 points. Removing it yields the fair, no-vig probabilities.Reduced juiceMarkets priced with a smaller-than-standard margin — for example −105/−105 instead of −110/−110. The break-even win rate drops from 52.4% to about 51.2%, which compounds into serious money across a season. Reduced-juice books are among the most valuable accounts a bettor can hold.Vig (juice)The sportsbook's built-in commission on every bet. At the standard −110/−110 market you risk $110 to win $100 on either side — roughly a 4.5% margin, requiring a 52.4% win rate just to break even. Also called juice; the market-level version is the hold or overround.
Bet types & tactics
Arbitrage (arbing)Betting both sides of a market at different books whose prices disagree enough that you profit whichever way it lands — possible when the combined implied probability falls under 100%. Real but thin (typically 1–2%), fast to vanish, and books limit accounts that do it aggressively.Buying pointsPaying extra juice to move a spread or total in your favor — for example from −3 to −2.5 at −120 instead of −110. Books price the half-point below its true worth almost everywhere except around football's key numbers 3 and 7, which is precisely where they charge the most.HedgingBetting the opposite outcome of a position you already hold to lock in a guaranteed result — most often cashing a futures ticket that is one game from winning. Hedging always sacrifices expected value for certainty; it is risk management for outsized positions, not a profit strategy.MiddlingBetting both sides of the same market at different numbers so a range of results wins both: take +3.5 at one book and −2.5 at another, and a 3-point margin wins twice. You risk only the vig for a shot at a double win — when books disagree enough to allow it.ParlayA single ticket combining multiple bets, all of which must win. The legs' decimal odds multiply — three −110 legs pay about +596 — but so do the probabilities and the book's margins. A parlay of standard-priced legs carries several times the house edge of a straight bet.Round robinA bet that automatically creates every smaller parlay combination from a set of picks — from four teams, all six two-leg parlays, for instance. It spreads risk versus one big parlay, but each mini-parlay still carries compounded vig; it is a convenience wrapper, not an edge.Same-game parlay (SGP)A parlay whose legs all come from one game — a quarterback's yards, his receiver's catches, the team to win. Because such legs are correlated, the book reprices the payout downward with a dedicated engine; the convenience is real, but SGP pricing is among the highest-margin on the board.TeaserA parlay in which every leg's line is moved in your favor — commonly six points in football — in exchange for a much lower payout. A standard two-team, six-point teaser at −120 needs each leg to win about 73.9% of the time to break even; most teased legs don't come close.The hookThe half-point on the end of a spread — the difference between −3 and −3.5. On football's key numbers it converts pushes into losses (or wins): losing “by the hook” means the half-point beat you. Half a point around 3 or 7 is worth far more than it looks.Wong teaserThe narrow teaser pattern with a historical mathematical edge, named for author Stanford Wong: tease only NFL favorites of −7.5 to −8.5 and underdogs of +1.5 to +2.5, so all six points cross the key numbers 3 and 7. Modern teaser pricing has shaved much of the edge — payout matters.
Math & measurement
Closing Line Value (CLV)The difference between the odds you took and the closing odds, expressed as a percentage: bet +150 on a line that closes +120 and you hold about +13.6% CLV. Because the close is the market's sharpest price, consistently positive CLV is the strongest known predictor of long-term winning.CorrelationWhen outcomes move together: a quarterback's passing yards and his top receiver's receiving yards, a blowout and the under on the losing team's props. Correlated parlays would beat fair multiplication of the odds — which is exactly why books reprice or block them.DrawdownThe decline from a bankroll's peak to its subsequent low — a $2,000 bankroll that touches $1,400 has taken a 30% drawdown. Every betting strategy has them; their expected depth scales with stake size and variance, and surviving them is what staking plans are for.EdgeThe percentage by which your expected return exceeds break-even on a bet — a 55% true probability at even money is a 10% edge. Real, durable edges in modern markets are small and usually come from price (line shopping, beating the close) rather than from out-predicting the market.Expected value (EV)The average profit or loss a bet would produce if repeated many times: (win probability × profit) − (loss probability × stake). A bet is +EV when the odds pay more than the true probability justifies — decimal odds × true probability greater than 1 — and long-term profit is simply accumulated +EV.Regression to the meanThe tendency of extreme short-run results to drift back toward the long-run average: a team winning every close game, a bettor on a heater, a 70% month. Betting as if the extreme will persist is one of the most common and expensive mistakes in handicapping.Risk of ruinThe probability that a betting bankroll goes to zero before the long run arrives, given a stake size, edge, and variance. Overstaking is its main driver: even a bettor with a real edge who bets 10% of bankroll per play carries a serious chance of busting on an ordinary cold streak.ROI (return on investment)Profit divided by total amount staked, the honest headline number of a betting record: $500 profit on $10,000 of stakes is a 5% ROI. Unlike win rate, ROI accounts for the prices paid. Sustained ROI above roughly 5% at scale is exceptional in sharp markets.Sample sizeThe number of bets behind a statistic. Small samples are dominated by luck: a 60% win rate over 50 bets is weak evidence, over 1,000 bets it is compelling. Higher-variance sports (hockey, baseball) require materially larger samples than basketball or football before any conclusion holds.Units wonProfit measured in units — a bettor's standard stake — rather than dollars: +12.4 units means twelve-and-a-half standard bets of profit. Units normalize records across bankroll sizes and stake changes, making them the standard way bettors compare results honestly.VarianceThe natural swinginess of betting results around their expected value. Even a genuinely profitable bettor endures long losing stretches — and a losing one enjoys hot streaks — purely by chance. Variance is why short-term results prove almost nothing and why staking discipline exists.
Bankroll & staking
BankrollThe pool of money set aside exclusively for betting — an amount you could lose entirely without touching real finances. Every staking decision is expressed as a fraction of it. Walling the bankroll off from rent and groceries is the first, non-negotiable rule of betting solvency.Bankroll managementThe discipline of sizing bets so that no realistic losing streak can bust you: a dedicated bankroll, a fixed staking rule (flat or proportional), units of 1–2%, and no size changes based on emotion or recent results. It doesn't create an edge — it keeps variance from destroying one.Flat bettingStaking the same amount on every bet regardless of confidence or recent results — one unit, every time. It forfeits some theoretical growth versus edge-proportional staking, but it is simple, immune to overconfidence, and eliminates the classic ruin pattern of sizing up after losses.Fractional KellyStaking a fixed fraction — half, quarter, eighth — of the full Kelly recommendation. Because Kelly is only optimal if your probability estimates are perfect, fractioning down sharply cuts drawdowns and estimate-error damage while retaining most of the growth. Quarter-to-half Kelly is the practical standard.Kelly criterionThe staking formula that maximizes long-run bankroll growth given an edge: stake fraction = (b × p − q) ÷ b, where b is decimal odds minus one, p your win probability, q = 1 − p. At +100 with a 55% estimate, full Kelly stakes 10% of bankroll — aggressive, which is why fractions are standard.Proportional stakingBetting a fixed percentage of the current bankroll rather than a fixed dollar amount, so stakes shrink through losing streaks and grow through winning ones. It cushions drawdowns automatically and compounds gains — the family of strategies to which the Kelly criterion belongs.Stop-lossA pre-committed limit — daily, weekly, or per-session — at which you stop betting after losses: say, three units in a day. Its power is that it is set while calm and executed mechanically, interrupting tilt and loss-chasing precisely when judgment is most compromised.UnitA bettor's standard stake, typically 1–2% of bankroll: on $1,000, a 1% unit is $10. Thinking and recording in units keeps sizing consistent and emotionally neutral, and lets records be compared across bankrolls — +10 units means the same thing whoever says it.
Bettor culture & practice
Bad beatLosing a bet that was all but won through an improbable late twist — a meaningless touchdown flipping a spread, an empty-net goal blowing an under. Bad beats are variance wearing its cruelest mask; their emotional weight, not their frequency, is what damages bankrolls afterward.Cash outA book's offer to settle a live ticket early for a guaranteed amount — effectively the book buying your position back, at a price that embeds a fresh margin on top of the original vig. Convenient for risk management; systematically accepting cash-outs surrenders expected value twice.Chasing lossesIncreasing stakes to win back what was just lost — the single most reliable way bettors go broke. Doubling after losses stacks leverage onto variance exactly when judgment is worst. If betting stops being fun or controllable, help exists: ConnexOntario 1-866-531-2600, ncpgambling.org.Grading (settlement)Marking a finished bet as won, lost, pushed, or void and crediting the result — done by the book for your wager and by your tracker for your records. Accurate grading against official final statistics is what makes a betting record trustworthy enough to learn from.HandleThe total dollar volume wagered — on a game, at a book, in a state's monthly report. Handle measures activity, not profitability: a book's revenue is handle × hold. Public handle reports are where the true scale of legal betting markets becomes visible.Key numbersThe final margins games land on most often. In the NFL, roughly one game in six ends with a 3-point margin and 7 is next — because field goals and touchdowns quantize scoring. Half-points that cross 3 or 7 carry outsized value; the same logic is far weaker in basketball.Limited (getting limited)Having your maximum stakes cut by a sportsbook — often to trivial amounts — because your action profiles as winning: beating the close, arbing, or hitting soft lines. Recreational-focused books limit aggressively; it is the industry's backhanded certificate of skill.Line shoppingComparing odds across multiple sportsbooks and always taking the best available price for the bet you already intend to make. It requires no prediction skill, yet compounds into one of betting's most reliable edges: −105 instead of −110 on every bet cuts the vig you pay by nearly half.Public moneyThe aggregate action of casual bettors, measured in ticket count more than dollars. It gravitates to favorites, home teams, overs, and brand-name franchises. Books lean prices a touch toward public tendencies on big events, which is the seed of fade-the-public strategies — and their overuse.PushA tie against the line: the favorite wins by exactly the spread (−3 in a 3-point game) or the total lands exactly on the number. Stakes are refunded, and in a parlay the pushed leg simply drops out. Half-point lines exist precisely to eliminate pushes.SharpA professional or consistently winning bettor whose action sportsbooks respect — and often respond to by moving lines. Sharps grind small edges through price: early numbers, line shopping, and closing-line value. The opposite archetype is the recreational “square” or public bettor.Sharp moneyWagers from bettors the book considers winners — identified by account history, timing, and size. Sharp money moves lines out of proportion to its ticket count: a few respected bets can shift a number that thousands of public tickets could not. Reverse line movement is its visible footprint.SquareSlang for a recreational bettor who bets favorites, overs, and famous teams at whatever price is posted. Square money arrives late, concentrates on popular sides, and is the volume books shade their lines toward — creating, occasionally, value on the other side.TiltEmotionally compromised betting after losses — bigger stakes, worse prices, forced action — named from poker. Tilt converts a normal, survivable losing stretch into a catastrophic one. Its antidotes are mechanical: pre-set unit sizes, stop-losses, and logging every bet where you'll have to see it.Void betA bet cancelled by the book with the stake returned — a postponed game, a player prop where the player never plays, an obvious pricing error (“palp”). Void rules differ meaningfully between books and are worth knowing before, not after, an unusual settlement.
Put the vocabulary to work
Free calculators for the concepts that matter most — odds conversion, Kelly staking, CLV, no-vig fair prices — and a tracker that measures all of it on your own bets.