Betting analytics glossary

Backing every outcome of an event across different bookmakers at prices that guarantee a profit no matter what happens. An arb exists when the best available odds for each outcome imply a combined probability of less than 100%. In practice arbs are thin (often 1–5%) and disappear fast as prices correct, and books limit accounts that do it consistently — so profit comes from speed and volume, not any single big edge.

The discipline of deciding in advance how much of your dedicated betting money to risk, so emotion never sets your stake sizes. Good bankroll management is what stops a losing streak from turning into a blown bankroll: by staking a fixed percentage per bet, your stake shrinks automatically when you are cold and grows when you are hot. It matters more than handicapping skill for staying in the game long enough to have an edge.

Whether the price you bet was better than the price the market settled on at kickoff. Because the closing line is the sharpest number a market produces, consistently beating it is the strongest available predictor that you are a long-term winner — and it reads your skill on every bet, win or lose, long before your profit graph does. Positive CLV over a large sample means your process has a real edge.

Backing two or more selections in the same event, with stakes split so that whichever one wins you collect the same return. Unlike an arbitrage, dutching does not cover every outcome — if none of your chosen selections win, you lose the whole stake. It is most useful when you can confidently narrow a market to a few contenders but cannot pick a single winner, letting you bet a view efficiently.

The average result of a bet if you could place it infinitely many times — your edge expressed in money. A bet is +EV (positive expected value) when your true probability of winning is higher than the probability implied by the odds. EV is the foundation of every winning strategy: you are not trying to win each bet, you are trying to place bets whose average outcome is profitable, then let volume do the rest.

Placing a bet on the opposite side of a position you already hold, so you lock in a result instead of leaving it to chance. Hedging trades upside for certainty: you give up the chance of the full original win in exchange for a guaranteed, smaller profit — or a capped loss. It is most valuable when a bet has moved sharply in your favour and you would rather bank a sure profit than ride the variance.

The bookmaker’s built-in margin — the reason the implied probabilities of all outcomes add up to more than 100%. If both sides of a coin-flip market are priced at 1.91 rather than a fair 2.00, that gap is the vig. It is the fee you pay to bet, and it is why simply winning slightly more than half your bets is not enough to profit: you have to beat the price plus the juice baked into it.

A formula that sets your stake as a fraction of your bankroll based on your edge and the odds, maximising long-term growth while protecting you from ruin. Full Kelly assumes your probability estimates are perfect, so most bettors use a fraction (half or quarter Kelly) to cut volatility and forgive estimation error. It is the mathematically optimal answer to “how much should I bet?” when you genuinely have an edge.

How a betting line changes between when it opens and when it closes, driven by new information and incoming money. Watching line movement tells you where informed money is going: a line that moves toward the side you bet is a sign you got ahead of the market. Understanding movement is the practical skill behind capturing closing line value — you want to buy a price before the move, not chase it after.

Betting both sides of a total or spread at different lines, leaving a gap where both bets can win. For example, backing the Over at 45.5 and the Under at 48.5 means a result of 46–48 wins both. Middles usually cost a small amount when the window is missed and pay a large double win when it hits — a deliberately asymmetric bet you find when lines move or differ between books.

Your profit divided by the total amount you have staked, expressed as a percentage. ROI is the truest single measure of betting performance because it accounts for both how often you win and the odds you won at — unlike win rate, which ignores price entirely. Sustained ROI in the low single digits over a large sample is genuinely strong; anyone promising 20%+ long-term is selling something.

A “sharp” is a sophisticated, winning bettor whose money moves lines; a “square” is a casual, recreational bettor. Bookmakers watch which is which and often move lines toward sharp money and away from square money. The distinction matters because you want your betting to look — and perform — like a sharp’s: disciplined staking, good prices, and positive closing line value, rather than chasing popular teams and narratives.

A sudden, sharp line movement across many books at once, caused by a wave of informed money hitting the market. Steam signals that sharp bettors have acted decisively on something. The value is in getting ahead of it — taking a price at a slower book before it adjusts — not in chasing after the move has already happened, by which point the edge is usually gone.

Measuring every bet as a fixed slice of your bankroll — commonly 1% — instead of in raw dollars. A “unit” keeps stakes proportional to your bankroll and, just as importantly, keeps the dollar amount out of your emotions: you risk “a unit,” not “a hundred bucks.” Standard plays are 1 unit, with 2–3 units reserved for your strongest spots. It is the simplest, most effective discipline a new bettor can adopt.

The natural swing of results around your true expected value. Because betting is high-variance, even a genuinely winning bettor will hit long losing streaks, and a losing bettor can run hot for a while. Variance is why you cannot judge your skill from a small sample, and why metrics like CLV — which read your edge on every bet — are so valuable: they cut through the noise long before your bankroll confirms it.