Over/under totals: the market that prices the game, not the winner
Totals ask how much scoring there will be, not who wins. The line types, what a push costs you, why alternative lines are priced worse than they look, and how a single score distribution prices every total at once.
An over/under bet is on the combined score of both sides against a line set by the bookmaker. Over 2.5 goals wins on 3 or more; under 2.5 wins on 0, 1 or 2. Half lines cannot end level, so there is no push. Whole lines can: over 3.0 with exactly 3 goals returns your stake.
Line types and what they do
| Line | Over wins on | Push on | Under wins on |
|---|---|---|---|
| 2.5 | 3+ | never | 0–2 |
| 3.0 | 4+ | exactly 3 | 0–2 |
| 2.75 | 4+ (full), 3 (half) | — | 0–2 (full), 3 (half loss) |
| 2.25 | 3+ | — | 0–1 (full), 2 (half win) |
Quarter lines behave exactly as they do on Asian handicaps: the stake splits across the two neighbouring lines and each half settles independently. The half-win and half-loss states must be weighted at 0.5 in any expected-value calculation — see Asian handicap for why this matters more than it sounds.
What is a push worth?
The chance of landing exactly on a whole line is not small in low-scoring sports. In football, exactly 2 goals occurs in roughly 22–25% of matches and exactly 3 in about 19–21%, depending on the league.
| Total | Approximate frequency in football |
|---|---|
| 0 goals | ~8% |
| 1 goal | ~19% |
| 2 goals | ~23% |
| 3 goals | ~20% |
| 4 goals | ~14% |
| 5+ goals | ~16% |
A push returns the stake, which is better than a loss and worse than a win. Comparing a 2.5 line at one price against a 3.0 line at another without accounting for the push probability compares two different bets.
Why do alternative lines usually cost more?
Books quote the main line to everyone and the alternatives to fewer people. Fewer participants means less correction, which means wider margins. An over 4.5 in a match priced around 2.5 may be quoted by three books rather than twenty — see thin markets.
One score distribution prices every total at once. That is the whole argument for building a matrix rather than estimating markets one at a time: the probability of over 2.5 and the probability of over 3.5 come from the same object and cannot disagree. Models that estimate each line separately routinely produce sets of prices implying two different games — see the Poisson model.
What does a view on totals actually require?
Totals reward information about tempo: whether both sides attack, how the referee treats time-wasting, whether a team needs a goal or a point. They punish information about quality, because a stronger side does not necessarily produce more goals — it may produce a comfortable 1–0.
If your reasoning about a match is "the home side are much better", that is an argument about the result, not the total. Placing it on a totals line is expressing the wrong opinion in the wrong market.
Check any of this against our record
Every signal CONSENSUS publishes carries the bookmaker odds fixed before the event starts and the settled result afterwards — including the drawdowns and the losing runs. The running total is on the front page and every entry is in the log.