Self-exclusion in numbers.
This is not a press release. It is the figures that exist, with sources and a downloadable file, and the figures that do not exist yet. The page is updated when the regulators answer our public-records requests.
B4GAMBLE earns commission from licensed casinos. Nothing in these figures links to one; the site menu does.
How many are self-excluded
| Sweden, 31 December 2023 | 104,301 | Spelinspektionen, annual statistics |
|---|---|---|
| Sweden, 31 December 2024 | 120,607 | Spelinspektionen, annual statistics |
| Sweden, 31 December 2025 | 134,484 | Spelinspektionen, annual report 2025: 101,803 men, 32,681 women |
| Sweden, 31 March 2026 | ≈ 138,000 | Spelinspektionen, quarterly statistics, rounded |
| Denmark, 31 May 2025 | 60,000 | Spillemyndigheden: 58 percent of registered men are under 40 |
The tables may be used freely with attribution (CC BY 4.0). The figures are the regulators'; we only collected them. Denmark: /data/rofus-registrations.csv.
The periods
Spelpaus: 10 days (since 29 June 2026, the first change since 2019), 1, 3 or 6 months, or until further notice (12 months minimum); cannot be cancelled; accounts reactivate automatically. ROFUS: a 24-hour pause, 1, 3 or 6 months, or permanent exclusion; a temporary exclusion cannot be lifted early. Neither covers unlicensed sites.
What nobody publishes
The regulators publish how many people are excluded. They do not publish:
- how many choose the shortest period, and whether they come back and register again
- how many exclusions expire each month
- how many pauses end in the week the salary arrives
- what happens the day after
Our public-records requests
We have requested monthly registrations by period, sex and age, the number of people registered more than once, and the number of exclusions that expired each month, from both regulators. The answers are published here as they come, with case numbers, even when they are refusals.
Status: requests sent in October 2026.
What it sounds like afterwards: 3,299 coded posts
We read 3,299 open forum posts about gambling and debt and coded every post on its own against a fixed scheme. The clearest thing in the material: where a self-exclusion register is named, a way round it or a way back comes up in more than a third of the posts. No quotes are reproduced.
| Coded posts | 3,299 | three open forums; the two Swedish ones coded in full, Reddit as a fixed selection |
|---|---|---|
| Write about their own gambling | 925 | of 3,299 posts (28.0 %); the rest are answers and relatives |
| Name a self-exclusion register | 174 | of 3,299 posts (5.3 %) |
| Of those: a way round it or a way back | 65 | of 174 posts naming a register (37.4 %) |
| Stand at the end of a pause | 8 | of 3,299 posts — the day almost nobody writes about |
Every figure carries its denominator, and what the material cannot show – no payday week, no trend over the years, no Danish figure – is printed on the study's own page together with all the aggregates as a CSV.
The day after
The question is not how many pause. The question is what happens the day the accounts open again. For that day there is a free plan without an account, and the plan itself contains no casinos.
The plan for the day after