Both teams to score: a bet on two independent failures
BTTS asks whether each side avoids a blank. It correlates with totals without being the same bet, comes from the same score distribution, and is where the independence assumption behind naive models fails most visibly.
Both teams to score pays out when each side scores at least one goal, regardless of the result. It is the probability that neither side is kept out — which is not the same as a high-scoring game. A 1–1 draw is a BTTS win and an under 2.5 win at the same time.
How does both teams to score relate to the total?
| Score | Total goals | Over 2.5? | BTTS? |
|---|---|---|---|
| 0–0 | 0 | No | No |
| 1–0 | 1 | No | No |
| 1–1 | 2 | No | Yes |
| 3–0 | 3 | Yes | No |
| 2–1 | 3 | Yes | Yes |
| 4–0 | 4 | Yes | No |
The two markets agree often enough that people treat them as interchangeable and disagree often enough that doing so is expensive. A team expected to win comfortably raises the total and lowers BTTS at the same time.
Pricing it from a score matrix
From a matrix of scoreline probabilities, BTTS is the sum of every cell where both coordinates are at least 1:
P(BTTS yes) = Σ P(home = i, away = j) for all i ≥ 1 and j ≥ 1Equivalently: 1 − P(home = 0) − P(away = 0) + P(0–0)The second form is worth remembering: subtract both clean-sheet probabilities, then add back the 0–0 case, which you have otherwise removed twice.
Where naive models break here
Independent Poisson treats each side's goals as unrelated. Real matches do not work that way: a side that concedes early attacks more, a side leading late defends. The dependence is small in most markets and concentrated in exactly the low-score corner that BTTS depends on.
This is why Dixon–Coles applies its correction to 0–0, 1–1, 1–0 and 0–1 specifically — those four cells carry a large share of the probability mass in football, and they are precisely the cells that decide BTTS. A model without that correction will systematically misprice this market. See Dixon–Coles explained.
BTTS is a useful diagnostic for any model you are considering following. Ask what its BTTS number is on a match where one side is heavily favoured. If it is close to what a naive independence assumption gives, the model has not accounted for goal dependence — and it will be wrong in the same direction every time.
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.