← All guides / Glossary

Glossary

Every term used across SharpSharks, in plain language. If you remember nothing else, remember EV%, CLV, and devig.

Core math

TermDefinition
EV%Expected value as a percent of stake. EV% = (fair_prob × payout_multiplier) - 1. A +5% EV bet returns $1.05 per $1 staked on average.
CLVClosing line value. The difference between the price you got and the price at game close. Positive CLV is the gold standard of sharp betting. Full guide.
DevigDe-vigging. Stripping the bookmaker's vig out of a price to recover the no-vig fair probability. Full guide.
Fair ValueThe vig-free price. The price you would expect at a fair market with no bookmaker margin.
Vig / JuiceThe bookmaker's margin baked into prices. -110 / -110 is ~4.5% vig.
No-vig PriceThe fair price after stripping vig. If a book quotes -110/-110 (52.4% / 52.4%), the no-vig price is 50/50.
Hold %Total bookmaker margin on a market. 1/odds_a + 1/odds_b - 1.
Implied ProbabilityThe probability of an outcome implied by its price. -110 = 52.4%, +150 = 40%.
Blended ProbabilityOur weighted average of multiple sharp books' fair probabilities. The v4 anchor.

Books

TermDefinition
PinnacleOffshore sportsbook with the lowest vig and highest accepted limits in the industry. Their lines are the closest thing to truth.
CircaLas Vegas book. Sharp on NFL/NBA spreads. Posts opening lines early and aggressively.
LowVigSharp offshore book with very low hold. Good cross-check for Pinnacle.
BetOnlineAGOffshore book that takes sharper action than most US books.
Sharp BookA book whose prices are well-calibrated and that accepts large bets without limiting (Pinnacle, Circa, sometimes Bookmaker, BetCRIS).
Soft BookA book whose prices lag the market and that limits winning players quickly (most US retail books to varying degrees).
Consensus PricingThe fair value derived from multiple sharp books, weighted by calibration. Our v4 method.

Edge categories

TermDefinition
+EV BetA bet with positive expected value. Long-run profitable assuming the fair value estimate is correct.
Tier (A/B/C)Our internal confidence rating. A: multi-book sharp confirmation, model agreement, calibrated book. B: solid but missing one element. C: speculative.
ArbArbitrage. Two opposing bets across books that lock in profit regardless of outcome. Full guide.
MiddleTwo-sided bet where both can win if the result lands in a specific window. Full guide.
ScalpA near-arb. Both sides at small loss in worst case, large win in best case (e.g., line moved). Same idea as middle, narrower window.
HedgePlacing the opposite side of a bet you already have, to lock in part of the profit (or cut losses).
SteamA rapid price move, usually triggered by sharp action. Following steam can be an edge, but the line moves before you bet.

Markets

TermDefinition
h2h / Moneyline / MLPick the winner. No spread.
SpreadPick a team to cover a points handicap.
Total / Over-UnderBet on combined points scored over or under a number.
Team TotalBet on a single team's total points.
Player PropBet on an individual player stat (points, rebounds, assists, etc.).
Game PropBet on game-level outcomes other than the main markets (first scorer, exact score, etc.).
Live / In-playMarkets offered after the event has started.

v4 method codes

CodeMeaning
v4_fullAll three primary sharp books contributed (Pinnacle 45%, LowVig+Circa 30%, BetOnlineAG 25%). Highest confidence.
v4_pinPinnacle was the dominant or only sharp anchor.
v4_consensusTwo or more sharp books contributed but not all three.
v4_singleOnly one sharp book had a quote. Use with extra caution.
v4_fallbackNo sharp books available, fell back to broader consensus or market median. Lowest confidence.

See Reading the v4 enrichment for the full breakdown.

Bankroll

TermDefinition
Unit1% of bankroll, by convention. A 1u bet on a $10,000 bankroll is $100.
Kelly CriterionMathematically optimal bet sizing given edge and odds. f = (bp - q)/b. Most pros use 1/4 to 1/2 Kelly to reduce variance.
ROIReturn on investment. Profit divided by total wagered.
VarianceHow much W/L bounces around expected value over a sample. High variance = need bigger samples to see the edge.
DrawdownThe largest peak-to-trough decline in bankroll. Even +EV bettors face long drawdowns. Plan for them.

Why we don't push parlays

A parlay is a sequence of bets that all must hit. Books love them because the vig compounds: a 2-leg parlay with -110 legs hides about 9% hold, vs 4.5% on a straight bet. Even when individual legs are +EV, the combined parlay is usually negative EV after the multiplier.

Our position: parlays are entertainment, not edge. We do not surface them. If you want to bet correlated outcomes, use same-game-parlay (SGP) only when the correlation is priced incorrectly, and even then, treat it as the exception.

Props vs Game Lines

Game lines (ML / spread / total) are highly liquid, sharp, and tightly priced. Edges are smaller and rarer, but more reliable.

Player props are less liquid, slower to update, and more error-prone. Edges are bigger but noisier. Books also limit prop bettors faster than game-line bettors. Different game, different style.

Related