SPORTS BETTING GLOSSARY
86 terms, defined the way a sharp friend would explain them. No fluff.
Odds and pricing
American odds
The default odds format at US books. Negative numbers (-110) tell you how much you'd risk to win $100, and positive numbers (+150) tell you how much you'd win on a $100 bet. So -110 means risk $110 to win $100, and +150 means a $100 bet returns $150 in profit.
Decimal odds
The format most of the world outside the US uses. The number is your total return per dollar staked, stake included, so 2.50 means a $100 bet returns $250 total ($150 profit). Handy because a bigger number always means a bigger payout, no sign-flipping like American odds.
Fractional odds
The old-school UK format, written like 5/2 or 10/1. The fraction is profit over stake, so 5/2 means you win $5 for every $2 you risk. You'll mostly see these on horse racing and futures boards.
Implied probability
The win rate baked into a price. For negative American odds it's odds/(odds+100), so -110 implies 110/210, which is 52.38%. For positive odds it's 100/(odds+100), so +150 implies 100/250, or 40%. If your honest estimate of the true probability beats the implied number, you've got a bet.
Vig (juice)
The book's cut, built right into the price. A fair coin flip should be +100 both ways, but books post -110/-110, and that extra bit is the vig. It's why you need to win 52.38% of your -110 bets just to break even, not 50%.
Overround
What you get when you add up the implied probabilities of every outcome in a market and it comes out above 100%. A standard -110/-110 spread market sums to about 104.8%, and that extra 4.8% is the book's margin. The bigger the overround, the worse the market is for you.
No-vig line (fair line)
The price after you strip the book's margin out. Take both sides' implied probabilities and scale them so they sum to 100%, and what's left is the market's actual opinion. Sharps compare their numbers to no-vig lines, not posted ones, because posted prices are rigged to sum over 100%.
Line
The number you're betting against, whether that's a spread (-6.5), a total (48.5), or a moneyline price. Lines move as money comes in and news drops. Where a line sits relative to where it opened tells you a lot about which side the market respects.
Opening line
The first number a book posts on a game, usually set by a market-making book and copied everywhere else. Openers are the book's best guess before the betting market weighs in, which makes them softer and lower-limit than anything you'll see later in the week.
Closing line
The final number before the game starts, after all the money and information has hit the market. It's widely treated as the most accurate estimate available, which is why beating it (see closing line value) is the cleanest skill signal in betting.
Key numbers
The margins games most often land on, which matters most in football, where 3 and 7 dominate final margins. The difference between -2.5 and -3.5 is way bigger than one point suggests, because so many games land exactly on 3. Sharps pay real attention to crossing key numbers; squares mostly don't.
Buying points
Paying extra juice to move a spread or total in your favor, like turning -3.5 into -3 for a worse price. It's almost always a losing proposition at standard pricing, with the classic exception of buying on or off key numbers in football when the price is right. If you're routinely buying half points off 6.5, you're just donating.
Off the board
When a book pulls a game or market and won't take bets on it. Usually it's an injury question (star quarterback suddenly doubtful), breaking news, or weird money the book doesn't want exposure to until things settle. The game usually comes back later with a fresh line.
Circled game
A game the book is nervous about, so limits get slashed and point-buying options disappear. Injuries, weather, back-to-backs, anything that adds uncertainty can get a game circled. If a game's circled, the book is telling you it doesn't trust its own number much.
Bet types
Moneyline
The simplest bet there is: pick who wins the game, no spread involved. Favorites carry negative prices (-200 means risk $200 to win $100) and underdogs carry positive ones (+170 means $100 wins $170). All your team has to do is win.
Point spread
A handicap that turns a lopsided game into a coin flip. Lay -6.5 and your team needs to win by 7 or more; take +6.5 and you cash as long as your team loses by 6 or fewer (or wins outright). The spread is a bet on margin, not on who wins.
Total (over/under)
A bet on the combined points both teams score, over or under a posted number. If the total's 48.5 and the game ends 27-24 (51 points), overs cash. You're not picking a winner, just how the game gets played.
Team total
A total for just one team's points instead of the combined score. It's useful when you have an opinion on one offense or one defense but not the whole game script. Make sure it's graded against that team's score only, because sloppy graders have been known to settle these against the full game total.
Parlay
Multiple bets bundled into one ticket where every leg has to win. The payout multiplies, but so does the book's edge, since you're eating the vig on every leg at once. Fun for a small sweat, terrible as a main strategy.
Same game parlay
A parlay built from correlated legs inside one game, like a quarterback's passing yards stacked with his team's win. Books reprice that correlation heavily in their favor, which makes SGPs some of the highest-margin products they sell. If it's advertised during every commercial break, that should tell you who wins on it.
Pleaser
The teaser's evil twin: you move the line against yourself for a bigger payout, so -3 becomes -9 and every leg still has to win. The math is ugly and books love these. There's a reason it sounds like a bet the house invented to please itself.
Prop bet
A bet on something inside the game other than the final result, like a player's rushing yards or the first team to score. Props get less market attention than sides and totals, and books post them with lower limits because they know their prop numbers are softer. That combination is why serious bettors spend so much time there.
Futures
A bet on something decided down the road, like a championship winner or a season win total. Your money's tied up for months, and the built-in margin on futures boards runs far higher than on game lines. Fine for entertainment, rough as an investment.
Live betting
Betting on a game while it's being played, with lines repricing on every possession. The speed cuts both ways: pricing algorithms can lag a big momentum swing or an injury, but books also charge more vig in-play. If you can't beat the closing line pregame, live betting won't save you.
Middling
Betting both sides of the same game at different numbers, hoping the result lands between them. Take over 45.5 early, watch the total climb, then take under 49.5, and any game landing 46 through 49 cashes both tickets. It costs a little vig when you miss and pays double when you hit, and it only exists because lines move.
Arbitrage
Betting every outcome of the same game across different books at prices that guarantee a profit no matter what happens. It shows up when books disagree enough that the combined implied probability drops under 100%. It's real but grindy: margins are thin, and books limit or close accounts that do it.
Run line
Baseball's version of the spread, almost always fixed at 1.5 runs. The favorite at -1.5 needs to win by two or more; the dog at +1.5 can lose by exactly one and still cash. Since the number rarely moves, the real shopping happens on the juice attached to it.
Puck line
Hockey's spread, fixed at 1.5 goals just like baseball's run line. A -1.5 favorite needs a two-goal win, which is why empty-net goals in the final minute cash or kill so many puck line tickets. Budget your emotions accordingly.
First five innings (F5)
A baseball bet settled on the first five innings only, built for betting starting pitchers without the bullpen roulette. If your read is on the starters, F5 isolates exactly that. It also ducks ninth-inning heartbreak entirely.
Cash out
The button that lets you settle a live ticket early for a guaranteed amount. Convenient, but the offer is priced with an extra margin on top of the vig you already paid, so you're taking a second haircut. It's occasionally right as risk management and usually just the book buying your ticket cheap.
Bankroll and staking
Bankroll
The pool of money you've set aside strictly for betting, separate from rent and groceries. Every staking decision flows from this number, and it should be money you could lose entirely without it changing your life. If you don't know your bankroll, you don't have a staking plan, you have vibes.
Unit
Your standard bet size, commonly 1% to 2% of bankroll. Units make records comparable across bettors with different wallets: +45 units means the same thing whether a unit is $10 or $1,000. Any capper quoting dollar profits instead of units is hiding the sizing story.
Flat betting
Staking the same one unit on every play regardless of confidence. It's the most honest test of picking skill because sizing can't rescue or ruin the record. Boring, and that's the point.
Percentage staking
Betting a fixed percentage of your current bankroll instead of a fixed dollar amount, so stakes shrink during downswings and grow during upswings. It compounds when you're running well and slows the bleed when you're not. The tradeoff is that a cold stretch right after a hot one hurts more in raw dollars.
Kelly criterion
A formula for bet sizing that maximizes long-run bankroll growth by staking a fraction based on your edge and the odds. The catch is that it assumes you actually know your edge, and betting full Kelly on an overestimated edge is how bankrolls die. Serious bettors run fractional Kelly (a quarter to a half) to survive being wrong about themselves.
Drawdown
The drop from your bankroll's peak to the lowest point that follows, usually quoted as a percentage. Even genuinely winning bettors sit through long, ugly drawdowns; that's variance doing its thing. A record's maximum drawdown tells you more about the ride than the final number does.
ROI
Return on investment: profit divided by total amount staked. A bettor who wagers 500 units and finishes +25 has a 5% ROI, and against -110 pricing, a sustained single-digit ROI is genuinely strong. A big ROI on a tiny sample means nothing, so check the volume before you're impressed.
Expected value (EV)
What a bet is worth on average across all outcomes: win probability times the amount won, minus loss probability times the amount lost. A +150 bet you win 45% of the time is positive EV even though it loses more often than it wins. EV is the whole game; the result of any single bet is just noise around it.
Variance
The gap between what should happen on average and what actually happens over any stretch you'll personally live through. Good bets lose, bad bets win, and short samples routinely make winners look like losers and vice versa. Respecting variance means judging process over results and sizing so a normal cold streak can't end you.
Standard deviation
The stat that measures how spread out results are around their average. In betting terms it tells you how big your normal swings should be, so you can tell an expected downswing from a genuinely broken process. High-odds bets like dogs and parlays carry far more of it than -110 sides do.
Sample size
The number of bets behind a record, and the first thing to check before believing one. A 15-2 heater is a coin doing coin things; hundreds or thousands of graded picks is where skill starts to separate from luck. When a capper's landing page shows a hot month and hides the lifetime graph, sample size is exactly what they don't want you asking about.
Break-even percentage
The win rate you need at a given price to neither make nor lose money. At -110 it's 52.38% (110/210); at +150 it's 40% (100/250). Every handicapping claim should be measured against this number, not against 50%.
Chasing
Raising your stakes after losses to try to win it all back fast, the single most reliable way to torch a bankroll. The math never supports it, and the psychology behind it is exactly what books are built to monetize. If you feel the urge, that's the signal to log off, not to double up.
Tilt
The emotional state after a bad beat or a losing run where you stop making decisions and start reacting. Tilted bettors chase, bet leagues they don't follow, and fire at halftime lines out of spite. Everyone tilts sometimes; the skill is recognizing it before the next bet instead of after.
Records and verification
Handicapper (capper)
Anyone who analyzes games and shares or sells picks, professional or otherwise. The title is self-appointed and says nothing about skill, which is why the only thing that matters is a complete, independently verified record. Judge the ledger, not the confidence.
Tout
A pick-seller, and in practice the word carries a smell. The classic tout playbook is loud marketing, selective memory about losses, and guaranteed locks for a monthly fee. Some pick-sellers are honest about their records; touts, by definition, aren't.
Verified record
A record where every pick was logged before the game started and graded by a neutral third party against official results. It's the opposite of a screenshot collection. That's the entire premise of the CAPTRACKER leaderboard: picks get captured at post time and settled automatically, so nobody's grading their own homework.
Self-reported record
A record the capper compiles and publishes themselves, with no outside check on what got counted. Even honest people misremember losses, and dishonest ones simply delete them. Treat any self-reported win rate as marketing copy until someone else has verified it.
Auto-settlement
Grading picks automatically against official score data instead of trusting a human to fill in results. Machines don't forget losses, don't fudge pushes, and don't quietly skip a bad weekend. CAPTRACKER grades every tracked pick against ESPN settlement data, with the whole process documented on the methodology page.
Immutable ledger
A record where entries can't be edited or deleted after the fact, so a pick logged on Tuesday still says exactly the same thing in March. It kills the classic tout moves: retroactive edits, vanished losers, restarted records. CAPTRACKER writes picks to an append-only ledger for exactly this reason.
Timestamped pick
A pick logged with the exact time it was made, provably before the game started. Without a timestamp, a winning pick is just a claim; with one, you can check it against the line that was actually available at that moment. It's the difference between evidence and a story.
Graded pick
A pick that's been settled against the final result as a win, loss, or push, at the odds it was logged with. Ungraded picks are where bad records hide, since pending forever is a convenient place for a loser to live. A trustworthy tracker grades everything or tells you exactly why it couldn't.
Push
A tie against the number: the game lands exactly on the spread or total, and your stake comes back. Lay -3 in a game decided by exactly 3 and nobody wins. Pushes are why half points exist and why they cost money.
Units won
Profit measured in standard bet sizes rather than dollars or win rate. It's the cleanest single number for comparing records because it accounts for odds: hitting 55% on heavy favorites can be worth fewer units than 45% on underdogs. That's why it's the headline stat on the CAPTRACKER leaderboard rather than raw win percentage.
Closing line value (CLV)
How your bet price compares to the closing line: bet a team at +7 and it closes +6, you beat the close. Because closers are the market's most accurate numbers, consistently beating them is the best early evidence of real skill, and it shows up long before win-loss records stabilize. It's also only measurable when picks carry honest timestamps, which is a core reason CAPTRACKER logs pick times.
Survivorship bias
The illusion created when you only see the winners because the losers disappeared. Have ten thousand people flip coins and the few who hit ten straight heads will start selling coin-flipping courses. Every hot capper you've heard of stands in front of a crowd of busted ones you never will, and complete, un-deletable records are the only antidote.
Cherry picking
Showcasing your wins while quietly omitting the rest. A capper going 9-1 in their last 10 NBA dog plays means nothing when they chose the ten, the sport, and the category after seeing the results. The only defense is a full record where every pick counts, no exceptions.
Record laundering
Restarting a public record after a cold streak so the losses vanish from view. New account, new system, new era, same guy. It's the reason lifetime records on an unerasable ledger matter more than anything a capper says about their recent run.
Market and sharp concepts
Sharp
A bettor who wins over the long run, whose action books respect and often limit. Sharps care about closing line value, shop every book, and bet numbers rather than teams. The term gets self-applied constantly; actual sharps are identified by books moving lines off their bets, not by their bios.
Square
A recreational bettor: favorites, overs, and whoever's on national TV. There's nothing wrong with it as entertainment; the term only stings when squares mistake themselves for sharps. Books build their whole retail product, from parlays to promos, around square behavior.
Sharp money
Bets from players the book knows are winners, which move lines even at modest dollar amounts. When a line jumps on light volume, that's often a respected account firing. Spotting real sharp action from the outside is hard, so take any content promising to show you sharp money with a shaker of salt.
Public money
The mass of recreational bets, heavy on favorites, home teams, and overs. On big TV games it can push a line off fair, which is where the fade-the-public idea gets its appeal. It's a real signal in narrow spots and a lazy strategy everywhere else.
Steam
A sudden, sharp line move across the whole market at once, usually because respected money hit one book and everyone else instantly followed. Chasing steam (betting the same side before slower books adjust) worked in a slower era, but books now move in seconds and flag accounts that try. The useful read today is informational: somebody big liked one side, a lot.
Reverse line movement
When the line moves against the side getting most of the bets, like the spread dropping while most tickets sit on the favorite. The usual read is that fewer but bigger, sharper bets are on the other side. It's one of the few public-data signals with real logic behind it, though the ticket-percentage data feeding it is often junk.
Line shopping
Checking multiple books for the best price before betting, the closest thing to free money in this whole business. Getting -105 instead of -110, or +3.5 instead of +3, compounds enormously over a season. If you only have one sportsbook account, opening a second one is the cheapest upgrade available to you.
Limits
The maximum a book will let you bet on a market. Limits start low on openers and props, rise toward game time, and get slashed personally for players who win. Being limited down to pocket change is annoying, but it's also the most sincere compliment a sportsbook can pay you.
Syndicate
An organized group betting serious money off shared models and coordinated accounts. Syndicates exist because winning players get limited fast, so the operation spreads its action across many outlets. When a line steams globally in seconds, it's often syndicate money, not one guy with a feeling.
Market maker book
A book that posts lines early, takes sharp action at high limits, and lets winners keep betting because their bets sharpen the number. Retail books mostly copy these origin lines and ban their winners instead. The market maker's closing prices are the reference point the rest of the industry, and all of CLV math, is built on.
Consensus
Where the market as a whole sits on a game, whether that's the average line across books or the majority side among bettors and cappers. Consensus line data is genuinely useful for spotting an outlier price worth grabbing. Consensus pick data (most bets are on the favorite, again) is mostly entertainment.
Convergence
When multiple independent sources land on the same side of the same game without copying each other. One capper liking a play is an opinion; a bunch of unrelated ones plus a supporting line move is a signal worth a look. It's a core ranking input on CAPTRACKER's daily feed, which surfaces spots where independently tracked cappers pile onto the same side.
Fade
To bet against someone, deliberately. People fade cold cappers, public darlings, or that one friend who's always wrong. Fading a bad picker sounds clever, but it only works if they're genuinely bad rather than briefly unlucky, and telling those apart takes the same sample-size discipline as finding a good one.
Tail
To copy someone else's pick and ride along on their read. Tailing a verified winner beats tailing a loud stranger, but you'll usually get a worse number than they did, and that slippage quietly eats the edge. If you tail, tail fast and check the line they actually got.
Handle
The total dollar amount wagered on a game, a sport, or at a book, before any winnings are paid out. Handle measures volume, not profit; a book can take a huge handle on a game and still lose on it. When states brag about their betting markets, handle is usually the big flashy number in the press release.
Hold
The percentage of handle the book keeps after paying winners. Theoretical hold comes from the overround; actual hold swings around it with results. Parlays and same game parlays hold several times more than straight bets, which is exactly why the apps push them so hard.
Slang
Chalk
The favorite, and by extension the boring, obvious side. A chalky card is one loaded with heavy favorites. Betting chalk isn't wrong, but laying -300 all day means one upset erases three wins.
Dog
Short for underdog, the side getting points or a plus price. Dog bettors win less often but get paid more when they do, and whole profitable styles are built on catching numbers the market has given up on. A live dog is an underdog with a real chance, not one that's merely cheap.
Bad beat
A bet that was basically won and then died in some cruel, low-probability way. Up eight with a minute left, then a garbage-time three, a meaningless free throw, a backdoor touchdown, and there goes your ticket. Every bettor has a bad beat story; the good ones don't let it change the next bet.
Backdoor cover
When a team that's been beaten all game scores late, meaningless points that flip the spread result. The winning team's fans don't care, but the folks who laid -7.5 and just watched a garbage-time touchdown make it a 6-point game care a great deal. Backdoor covers are where bad beats and dog money meet.
Hook
The half point on a spread or total, as in seven and a hook for 7.5. Hooks can't push, so they decide every game that lands right on the number. Losing by the hook (your team loses by exactly 7 after you laid -7.5) is its own specific flavor of pain.
Juice (slang)
Street name for the vig, the book's built-in cut on every bet. Asking what's the juice just means asking what the price is. You'll also hear a line described as juiced up when it carries more than the standard -110.
Lock
A pick sold as a guaranteed winner, which is your cue that you're the product, because guaranteed winners don't exist in a market with vig. Anyone selling locks is waving a red flag: if they truly had a can't-lose bet, they'd bet it themselves, not retail it to strangers. Treat the word as a filter; hear it, walk away.
Degen
Short for degenerate, worn half as a badge and half as a confession. It means betting for the rush: 3 a.m. table tennis, twelve-leg parlays, action for action's sake. Self-aware degens having fun with money they can afford to lose are fine; the word stops being funny when the money's needed elsewhere.
Sweat
The experience of watching a game you have money on, as in a great sweat or sweating the under. A good sweat is half the product bettors are actually buying. If a bet's outcome doesn't make you feel anything, congratulations, you've sized it correctly.
Ticket
A single bet slip, from the paper stubs at a Vegas window to the digital version in your app. Having a ticket on something just means you bet it. Old-timers still tell stories about winning tickets that never got cashed, and the books keep every dollar of that money.
Cooler
A game or a stretch that kills everything you touch, borrowed from old casino slang for a run-ending presence at the table. Saying the book sent a cooler is how bettors joke about a night where every leg dies late. It's pure superstition, and every bettor indulges it a little; variance doesn't need a personality, but we give it one anyway.
Now that you speak the language, see which handicappers actually win. Every record auto-settled by ESPN data.
VIEW LEADERBOARD →