Odds and probability

Football Prediction Glossary: Odds, Markets and Model Terms A to Z

A plain English glossary of football prediction and betting terms, from 1X2 and accumulator to closing line, expected goals, overround, value and yield.

By the HereBets editorial team ·

Wireframe illustration of a card index box with tabbed dividers next to a tactics board

A price of 1.25 on a home win looks like easy money until it is read as 80 percent, which leaves one match in five going another way. Most of the vocabulary around football predictions exists to make that kind of translation possible. The entries below are short on purpose; each points to the guide that covers the idea properly.

80%implied by a price of 1.25
1 in 5matches left for another result
5.06total odds of four legs at 1.50

Numbers and A–C

1X2
The three-way result market: 1 is a home win, X a draw, 2 an away win, settled on normal time plus stoppage time. Covered in the guide to 1X2, double chance and draw no bet.
Accumulator (acca)
One bet combining several selections, where every leg must win and the odds are multiplied together. Four legs at 1.50 give total odds of about 5.06. The accumulator guide shows why the margin compounds with every leg.
American odds
A format shown as plus or minus numbers. +150 means a 100 stake returns 150 profit; −200 means 200 must be staked to win 100.
Asian handicap
A market that gives one team a goal head start or deficit, often in half or quarter goals, which removes the draw. A quarter line such as −0.25 splits the stake across two neighbouring lines.
Banker
The selection someone considers most reliable, used across several slips. It is a feeling, not a measure.
Brier score
A measure of how good probability forecasts are: the squared difference between the forecast and what happened, averaged across many matches. Lower is better.
BTTS
Both teams to score, yes or no. Explained alongside goal lines in the guide to over/under and BTTS.
Calibration
Whether forecasts match reality over time. A well-calibrated forecaster's 60 percent calls come true about 60 percent of the time.
Closing line
The odds available just before kick-off, when the most information is priced in. Beating the closing line consistently is widely treated as a sign of genuine skill.
Correct score
A market on the exact final score. Prices are long because even the most common scorelines, such as 1–1 or 1–0, happen in only a small share of matches.

D–I

Decimal odds
The total return per 1 staked, including the stake. A 10 stake at 3.20 returns 32. This is the standard format across most of Africa and Europe.
Double chance
A bet covering two of the three results: 1X (home or draw), X2 (away or draw) or 12 (either team wins).
Draw no bet (DNB)
A bet on one team where the stake is returned if the match is drawn.
Elo rating
A rating system, borrowed from chess, that moves points between teams after each match based on the result and the expected result. Club Elo publishes ratings of this kind for European club football.
Expected goals (xG)
The probability that a given shot becomes a goal, based on thousands of similar past shots, added up across a match. Data firms such as Opta publish xG figures. The method and its limits sit in the prediction models guide.
Favourite-longshot bias
The long-observed pattern that long shots tend to be priced more generously than their true chances deserve, so they lose more per unit staked than favourites.
Fractional odds
The UK format showing profit relative to stake. 5/2 means 5 profit for every 2 staked, equal to decimal 3.50.
Hit rate (strike rate)
The share of predictions that won. Without the odds attached, it says almost nothing about whether a method makes money.
Implied probability
The chance a price suggests, calculated as 1 divided by the decimal odds. 2.00 implies 50 percent; 4.00 implies 25 percent. The guide to reading odds walks through the conversions.
In-play (live) betting
Betting while the match is being played, with prices moving after every attack, card and goal.
Price 1.251.2580%
Price 2.002.0050%
Price 4.004.0025%

L–R

Log loss
Another scoring rule for probability forecasts that punishes confident wrong calls very heavily. A forecaster who says 95 percent and is wrong pays dearly.
Margin (overround, vig)
The bookmaker's built-in edge. Add the implied probabilities of every outcome in a market; the amount above 100 percent is the overround. A 1X2 market adding to 106 percent carries a 6 percent overround.
Over/under
A market on total goals above or below a line such as 2.5. Half-goal lines cannot finish level, so there is no refund.
Poisson model
A statistical model that estimates how many goals each team will score from their attacking and defensive strength. Maher's 1982 paper is a well-known starting point, and Dixon and Coles refined it in 1997.
Push
A bet that ends level and returns the stake, common on whole-number handicap and goal lines.
Return on investment (ROI)
Profit divided by total amount staked over a period. Often used interchangeably with yield.

S–Z

Stake
The amount placed on a bet.
Sure win
A phrase for a prediction claimed to be certain. No football outcome is certain, and a few minutes with a 1.10 favourite's implied probability shows why.
Treble
An accumulator with exactly three legs.
True (fair) probability
An estimate of the real chance after the margin is removed from a price, or from an independent model.
Value
A price whose implied probability is lower than the real chance of the outcome. If a side truly wins 50 percent of the time, odds of 2.20 offer value and 1.90 do not. Spotting value requires a reliable estimate, which is the hard part.
Variance
The natural swing of results around the long-run average. A sound method can lose for weeks; a poor one can win for a month.
Yield
Profit as a percentage of total stakes. A yield of 3 percent over a thousand bets is considered strong; a yield claimed over twenty bets means very little.

One term missing from most lists is sample size. It belongs in every conversation about a forecaster’s record, and it is usually the first thing a sales pitch leaves out.

FAQ

Your questions, answered.

What is the most useful betting term for a beginner to understand?

Implied probability. Once a price is read as a percentage, every other idea follows: the margin is the amount by which those percentages exceed 100, value is a price that implies less than the real chance, and a long accumulator is visibly a string of small percentages multiplied together. Divide 1 by the decimal odds and the price starts to make sense.

How is a banker different from a double chance leg?

A banker is a label a person puts on a pick, while double chance is a market. The banker can be any selection, usually a short-priced home win, that the bettor trusts enough to place in every line of a system bet. A double chance leg covers two of the three results, such as home win or draw, at a lower price. Many slips use a double chance leg as the banker because it loses less often, but it also pays much less.

Is a banker the same as a sure win?

No. A banker is simply the pick a person feels most confident about, often placed in every line of a system bet or at the heart of an accumulator. Confidence is not certainty. Heavy favourites priced around 1.20 still fail roughly one time in six on their own odds, and a banker that loses sinks every slip it appears in.

What is the difference between odds and probability?

Probability is the real chance of something happening. Odds are a price set by a bookmaker, which reflects their estimate of that chance plus a margin. Converting odds to implied probability gives the bookmaker's view including that margin. Removing the margin gives a fair estimate, which can then be compared with your own.

What does BTTS stand for?

BTTS means both teams to score. A yes bet wins if each side scores at least one goal in normal time, whatever the final result, so a 1–1 draw, a 3–2 home win and a 1–2 away win all count. A no bet wins if at least one team fails to score, including any 0–0.