Skip to content
Market Trends & Consumer Behaviour

Gambling participation by age: Britain’s growth is men over 55

Key Takeaways

  • Gambling participation by age moved in only one direction in Great Britain between 2023 and 2025: men aged 55 and over, whose past-year non-lottery gambling rose from 27% to 32% in the Gambling Commission’s Trends Report of 16 July 2026.
  • Men aged 18 to 34 went from 54% to 51% over the same period, at a p value of 0.211, so Britain’s much-repeated story about young men is weaker in the official statistics than in the commentary.
  • The United States is the opposite. The Wheatley Institute’s Gambling With Our Future, published on 13 October 2026 from a survey of 2,548 adults, found 67% of men aged 18 to 29 had placed a mobile sports bet in the past year, against 41% of young women.
  • One customer risk model across both markets is therefore calibrated on an American age profile, and that shows up first in gambling harm detection thresholds.

Why is older men gambling the only age trend that actually moved?

Because it is the only one that passed a significance test. The Commission’s Gambling Survey for Great Britain: Trends Report 2023 to 2025, published on 16 July 2026, puts past-year non-lottery participation among men aged 55 and over at 27% in 2023 and 32% in 2025, with past-four-week participation up from 18% to 22%, at p values of 0.003 and 0.001.

Now set the group everybody talks about alongside that. Men aged 18 to 34 ran 54%, then 55%, then 51% on the same measure, at a p value of 0.211, so the fall cannot be separated from sampling noise, and men aged 35 to 54 barely moved, 49% to 48% at 0.859. The Commission’s wording is that the older male pattern “was not observed for other age groups and was not evident among female participants”.

I have heard the opposite version told from stages for two years, and I do not think anyone is being careless on purpose. The American research is alarming and it travels easily, so it gets repeated by people who read the headline finding rather than the trend tables.

What do the American numbers actually say?

They say young men, clearly and recently. The Wheatley Institute at Brigham Young University published Gambling With Our Future on 13 October 2026, from a survey of 2,548 US adults fielded by Qualtrics with an oversample of 18 to 29 year olds across 16 activities. Among men aged 18 to 29, 67% had placed a mobile sports bet in the past year against 41% of women the same age, and 59% gamble at least a few times a week against 34% of young women. Past 30 those weekly figures fall to 33% and 20%.

Americans aged 18 to 29 rank making money (53%) above entertainment (50%) as the reason they gamble, and everyone over 30 reverses it, 51% against 37%. Men are around three times as likely as women to report losing more than $1,000 in a single day. That motive split is what I would take into a product meeting ahead of the participation rate.

None of that is wrong and none of it is British. Mobile sportsbooks have been legal state by state in the United States only since the Supreme Court struck down the Professional and Amateur Sports Protection Act in May 2018, and they arrived with the promotional spending of a race for share, whereas remote betting has been licensed in Great Britain since 1 September 2007. Somebody who met mobile betting at 22 in 2019 is not the customer who met it at 22 in 2008.

What breaks if gambling participation by age is shifting older?

The harm detection model breaks first. Most of the logic in use was tuned on a young male spend pattern, fast session escalation, in-play volume and deposit velocity, because that is where the early online losses were. A customer in their sixties on a fixed retirement income, losing a steady amount every week, can reach a larger annual loss than a 24 year old without ever crossing a velocity threshold, and nobody around that customer is raising it either: help sought because of somebody else’s gambling among the over-55s fell from 2.3% in 2023 to 1.1% in 2025, and among women in that age group from 2.9% to 1.2%.

The advertising argument breaks next, and that one costs more. The House of Lords Liaison Committee’s follow-up report of 17 September 2026 made the case for a full ban largely on exposure among young people. An operator defending itself on youth exposure is answering a question Great Britain’s participation figures do not pose, the restriction arrives anyway, and the credibility is spent for nothing.

The Commission has also begun publishing the cuts a plain age and sex breakdown hides. Its release of 27 October 2026, Gambling behaviour and harm among people from different ethnic groups, puts the share of the population scoring 8 or more on the Problem Gambling Severity Index between 1.9% and 7.7% by group, against 2.4% nationally, and the gap in reported adverse consequences was widest at the low PGSI scores. A programme that escalates only on a high score is slowest where the difference is largest.

Why move on this before the next survey wave is published?

Because the regulator has said what it wants to know next and does not have the answer. Helen Bryce, its Head of Statistics, wrote on 16 July 2026 that the trend data shows an increase among males aged 55 and over, and the report asks for further analysis of which activities those men are taking up. An operator who can answer that about its own customers first is in a different conversation when the next customer interaction findings arrive.

The work is narrow enough to start this quarter. Re-cut the customer base by age band and report three numbers for each band rather than one: participation share, net loss per active customer over 12 months, and the median time from the first recorded risk marker on an account to the first restriction applied. Then take the gambling harm detection thresholds and establish in writing which cohort they were calibrated on, and whether anybody has validated them against customers over 55.

TGB’s Customer Segmentation Framework sets out the dimensions, profiles, activation and governance for the segmentation itself, so the age-band-against-risk validation above is the step to run on top of it. If the model turns out to have been built on under-35 behaviour and never retested, that is the finding, and it is cheaper to have it now than to have a regulator arrive at it first.

Boardroom Questions

  1. Which cohort was our harm detection model calibrated on, and when was it last validated against customers aged 55 and over?
  2. What is our net loss per active customer by age band over the last 12 months, and which band is largest?
  3. How long, at the median, from the first recorded risk marker on an account to the first restriction applied, and does that differ by age band?
  4. Do we run one customer risk profile across our US and Great Britain licences, and whose data built it?
  5. If the Commission publishes an activity breakdown for men aged 55 and over, could we produce our own equivalent the same week?
  6. Where we escalate on a high PGSI score, what happens to a customer reporting real consequences at a low one?

Sources

  1. Gambling Commission, Gambling Survey for Great Britain (GSGB) Trends Report 2023 to 2025, official statistics, published 16 July 2026, including trend figures for participation in the past 12 months and past 4 weeks by age and sex, and help sought for own and others’ gambling.
  2. Gambling Commission, Gambling Survey for Great Britain: Annual Report 2025, official statistics, published 16 July 2026, based on 20,775 adults in Great Britain across four quarterly waves from January 2025 to January 2026.
  3. Helen Bryce, Head of Statistics, Gambling Commission, “Gambling Survey for Great Britain: 2025 results and three years of trend data”, Gambling Commission blog, 16 July 2026.
  4. Gambling Commission, “Gambling behaviour and harm among people from different ethnic groups”, statistical research release, 27 October 2026.
  5. Wheatley Institute, Brigham Young University, Gambling With Our Future, published 13 October 2026, national survey of 2,548 US adults fielded by Qualtrics with an oversample of adults aged 18 to 29, covering 16 gambling activities.
  6. House of Lords Liaison Committee, Gambling Harm: Time for Action: Follow-up report, HL Paper 52, published 17 September 2026.
  7. Murphy v. National Collegiate Athletic Association, 584 U.S. 453 (2018), decided 14 May 2018, striking down the Professional and Amateur Sports Protection Act 1992.
  8. Gambling Act 2005 (c. 19), remote gambling licensing provisions in force 1 September 2007.