Both teams to score, explained: when it wins and how often it happens

Both teams to score — BTTS, «Goal» on an Italian coupon — wins if each side scores at least once. «No» wins if at least one finishes on nought: 0-0, 1-0 and 4-0 are No, 1-1 and 3-2 are Yes. Two outcomes, no third case.

How often it happens: across our whole archive of matches with a result, both sides scored in 51.82% of them — 131,881 matches out of 254,498. It is a market close to even: unlike a 1X2, whose three outcomes start from very different frequencies, here the two sides are worth roughly the same before you look at the teams.

What counts as a goal, and until when

Goals in normal time count, stoppage time included. Extra time and a shoot-out don't: a BTTS market closed at the 90th stays closed. It is the Under/Over rule, and the one our tracker uses to decide whether a logged bet won or lost.

Own goals count like any other, credited as the scoresheet credits them: a match can turn Yes without the second team having a shot on target. Disallowed goals don't count.

How often it happens, league by league

That 51.82% averages hundreds of very different competitions. League by league the spread is wide and follows the goals-per-game average: where more goals are scored both sides scoring is easier, and the two columns below put the seven in almost the same order.

openfootball archive, CC0 licence, from the snapshot stamped «aggiornato: 2026-09-19T08:16:51.350Z»: the file is rewritten several times a day, and the date on its own fixes nothing. Two rows are the same match if they share the season, the teams once names are normalised, the score, and dates no more than a day apart. The count lives in scripts/verifica-goal-nogoal.mjs, which re-checks every figure on this page.

Each league is held together by country plus name, with every label it uses: in the archive the same league changes name from season to season. Serie A 2025-26 sits under «Serie A», the other seasons under «Italian Serie A»; it happens in five rows of seven, and by label alone each lost its last completed season — 380 matches in Italy, England and Spain, 306 in Germany and France, and for Spain three seasons, not one. The country is needed because the name isn't: «Premier League» unqualified, in 2025-26, also covers 853 Kenyan, Rwandan, Tanzanian and Sudanese matches.

The seven in the table are the top divisions of seven European countries: an editorial choice, not a criterion — the archive holds many other continuous sources, from the Championship to the 2. Bundesliga. The filter on the data, a source with no missing seasons, only excludes: it keeps out Serie B, which the archive carries from 2013-14 but with four seasons missing.

⚠ Duplicates spelled too differently escape the key, so 254,498 is a ceiling. The Bundesliga proves it: a second label carries 636 rows whose season, date and score are already among the canonical ones — complete at 306 a season, where others wouldn't fit. We took them out of the table by hand; 192 survive in the total. Three shortfalls then remain: Ligue 1 2019-20 at 279 matches and the Eredivisie 2019-20 at 232, seasons cut short, and one match of 306 missing from the Eredivisie 2025-26.

An average over many seasons hides how much the number moves

Over a long series the percentage looks constant. It isn't: in Serie A, across the last three complete seasons, BTTS came in at 52.37%, then 51.58%, then 45.26%. Six points in a year, with nearly the same teams.

With the labels normalised, all seven have the same three complete seasons: 2023-24, 2024-25, 2025-26. The gap between highest and lowest runs from the Bundesliga's 4.9 points to the Eredivisie's 7.4 — Premier League 5.5, Primeira Liga 5.6, Ligue 1 6.2, Serie A and La Liga 7.1. An archive percentage says where the centre is, not how wide the target is.

BTTS and Under/Over are not the same market

You often hear that Yes and Over go together: more goals, more chance both sides score. As a tendency that is true. But a 1-1 is Yes and Under 2.5, a 3-0 is No and Over 2.5. The question is how much the crossed ones weigh.

The same 254,498 matches, from the same snapshot: the four combinations are exhaustive and add to 100.

The bottom two rows are worth 22.31% together: more than one match in five where the two markets settle opposite ways. And it barely moves between leagues — across the seven above, from Serie A's 22.23% to the Eredivisie's 25.67%.

⚠ Our Under 2.5 piece measures the same leagues and gives lower totals — La Liga 4,981 against 5,379, Ligue 1 3,958 against 4,273 — because it predates the label discovery: those gaps are the seasons that were missing. It needs redoing, and we flag it here because this is the page sending you there.

The other market of the pair, with its own archive frequencies: What Under 2.5 means: how to read the Under/Over market

What this market costs

A market's price is judged by the margin, what the operator keeps before paying. Before the numbers, one thing holds up the rest: the two markets are not priced on the same matches. Across the 1,128 matches that hadn't started, Betsson prices the match result on 1,128 and BTTS on 1,045; Betflag 700 against 686; Netwin 702 against 672. BTTS is always a subset, and the two columns below sit on the same matches: the ones where that operator prices both.

Median margin, taken on 19 September 2026 at 18:53 Italian time, in a single read of the list: 1,242 matches, 114 in play and excluded. «Major competitions» means the five big leagues plus the Eredivisie, Primeira Liga, Champions and Conference League: 118 matches, enough to keep the three operators above twenty, the threshold below which we don't publish a median.

Across the whole list the two-outcome market costs less than the three-outcome one at two operators: Betsson by almost four points, Netwin by a little under one. At Betflag it doesn't — 10.61% against 10.47%, with BTTS on the expensive side. Fourteen hundredths of a point is not a result in either direction: in a read taken the same evening, forty minutes earlier, the comparison came out with the opposite sign.

⚠ Put like that, that row looks as though it contradicts our Under 2.5 piece, which makes the same comparison and concludes that «two outcomes cost less than three» is a reasonable hypothesis the data do not confirm. The difference isn't in the numbers, it's in the base. That piece measures competition by competition; the «whole list» row measures a set dominated by minor competitions, where the 1X2 widens far more than BTTS: outside the majors Betsson's median 1X2 goes from 6.46% to 11.60%, its BTTS only from 5.98% to 7.65%.

It is that widening that produces the apparent advantage of the two-outcome market, not the number of outcomes. On the other piece's base — the majors alone — the count comes out identical: one operator in three, and by half a point, with Betflag and Netwin going the other way by 1.15 and 2.20 points. Which is why the panel below fills from the feed: a ranking photographed once is not a rule.

How the margin is worked out, and why the ranking between operators changes from market to market: What the bookmaker's margin is, and how to measure it on a 1X2

Three things these numbers do not say

• They don't say how a match will end: a frequency over thousands of matches does not apply to one, which depends on who is playing.
• They don't say one league is better than another: they say that in the Bundesliga and the Eredivisie goals spread across both sides more often than in Portugal.
• ⚠ They don't say that a gap between the archive frequency and today's implied probability is a market error: the archive looks at an average across many seasons, the price at those two sides today.

The comparison between operators, match by match, is on the odds page; a match's own page puts the books in a column with each side's goal averages.