Why young men are described as most at risk, whether that holds up, and how what happens to them reaches everyone else. Updated 2026-10-08.

10 checked findings · 3 measured on young men · 89 leads waiting to be checked

New here? How to read the library · how I check sources · start with the first entry.

Who’s at risk, and why

Among under-30s, 10% of men and 7% of women scored as problem gamblers; across all ages it was 3%

men 18-30: 10% · women under 30: 7% · all respondents: 3% (PGSI 8+)

Measured on young men Measured on: US registered voters; subgroups men 18-30 and women under 30 · 18+; subgroups 18-30 (men), under 30 (women) · USA

What the number is of: Share of each subgroup with PGSI 8+. Overall margin of error +/-3.5 pts; subgroup sizes not published (much wider)

Study type: Poll: live phone (520) and text-to-web (281), 17-20 Aug 2024, n = 801

Source: FDU Poll (2024). 'FDU Poll finds online betting leads to problems for young men.' Fairleigh Dickinson University, 19 Sept 2024 · Checked: press release, 2026-10-05

In young Australian men, more frequent sports betting predicted later problem gambling

OR = 1.63 [95% CI 1.22, 2.20] per one-step increase in frequency ('64% increased odds'; table shows 1.64)

Measured on young men Measured on: Young men in the Ten to Men longitudinal cohort; n = 265 (n = 215 controlling for prior severity) · 18-25 · Australia

What the number is of: Odds of a higher PGSI risk category per one-step rise on a 5-point frequency scale (never / 1-2 a year / few a year / 1-3 a month / weekly+)

Study type: Longitudinal cohort; predictors at Wave 3 (Jul 2020 - Feb 2021), PGSI outcome at Wave 4 (Aug - Dec 2022); ordinal logistic regression

Source: Mancini, Brett, Heirene, Fisher, Nevill & Mitrou (2025). Predicting problem gambling in young men: the impact of sports gambling frequency and internalizing symptoms. J. Gambling Studies 41(3), 1119-1144 · Checked: full text, 2026-10-05

In the same young men, higher anxiety went with LOWER odds of later problem gambling; higher depression with slightly higher odds

anxiety OR = 0.89 [0.79, 0.99] per point (11% lower odds) · depression OR = 1.10 [1.01, 1.20] per point (10% higher)

Measured on young men Measured on: Same as previous row (Ten to Men, n = 265) · 18-25 · Australia

What the number is of: Odds of higher PGSI risk category per one-point rise in anxiety / depression score

Study type: Same longitudinal model

Source: Mancini, Brett, Heirene, Fisher, Nevill & Mitrou (2025). Predicting problem gambling in young men: the impact of sports gambling frequency and internalizing symptoms. J. Gambling Studies 41(3), 1119-1144 · Checked: full text, 2026-10-05

Problem-gambling severity is higher among men and among 18–24s — reported separately

All adults 2.4% (95% CI 1.9–2.8) · of past-12-month gamblers: all 4.0%, men 5.3%, women 2.7%, 18–24s 10.4%

Partly: young or male, not both Measured on: General population; breakdowns are of people who gambled in the past 12 months · 18+; band quoted 18–24 · Great Britain

What the number is of: Of past-12-month gamblers (only 2.4% is of all adults). Sex x age cell NOT published in report text

Study type: Push-to-web official statistics survey, ~20,000 adults, 4 quarterly waves, Jan 2025 – Jan 2026

Source: Gambling Commission. Gambling Survey for Great Britain, Annual Report 2025 (Official Statistics) · Checked: PGSI chapter + full print rendering, agreeing, 2026-09-21

Boys whose friends gamble are far more likely to gamble themselves

of boys whose friends gamble: over 8 in 10 (84%) gamble · of boys whose friends don't: under 2 in 10 (17%)

Partly: young or male, not both Measured on: US boys · 11-17 · USA

What the number is of: Share who gamble, among boys whose friends gamble (84%) vs boys whose friends don't (17%)

Study type: Survey of 1,000+ boys, July 2025; method not stated on the press release

Source: Common Sense Media (2025). Betting on Boys: Understanding Gambling Among Adolescent Boys · Checked: press release, 2026-10-05

Worldwide, men were 3.4 times more likely than women to have a gambling problem, and younger adults 1.51 times more likely than middle-aged adults; the sex gap narrows in North America

men vs women 3.4x · young (18-35) vs middle (30-55) 1.51x · older (45-65) 0.80x · global PG prevalence 1.9% (Europe 1.3%, North America 5.3%)

Partly: young or male, not both Measured on: Problem-gambling prevalence samples in 21 studies; 156,249 participants (47.6% male) · Grouped 18-35 / 30-55 / 45-65 (overlapping) · 18 countries (Europe 52%, Asia 19%, Oceania 19%, North America 10%)

What the number is of: Ratio of problem-gambling prevalence: men vs women (all ages); young vs middle-aged (both sexes)

Study type: Systematic review and random-effects meta-analysis

Source: Dellosa & Browne (2024). The influence of age on gambling problems worldwide: a systematic review and meta-analysis of risk among younger, middle-aged, and older adults. J. Behavioral Addictions · Checked: abstract, 2026-10-05

Gambling problems predict later loneliness; loneliness does not predict later gambling problems (within-person)

problems→loneliness β = 0.069, p = .007 · loneliness→problems β = 0.001, p = .971 · escape→problems β = 0.077, p = .047

Measured on: General-population adults who gambled at least monthly, not problem gamblers; Norstat panel; 54.58% male · 18–75; mean 51.85 (SD 13.97); younger participants lost to attrition · Finland

What the number is of: Within-person standardized paths; analytic n = 612 of 1,530 at T1

Study type: Longitudinal RI-CLPM, 5 waves, Apr 2021 – May 2023

Source: Hagfors, Kaakinen, Savolainen, Vepsäläinen & Oksanen (2025). Lonely gamble. J. Behavioral Addictions 14(2), 972-981 · Checked: full text (PMC), 2026-09-21

Loot boxes are rare among games but near-universal in exposure: the few titles carrying them are the ones everyone plays

16.2% of games (75 of 463) · 71.2% of players playing a game with them (week of 16–22 Apr 2019)

Measured on: Concurrent players of the 463 most-played Steam games · Not reported (platform counts) · Global Steam

What the number is of: 71.2% = share of players; 16.2% = share of games

Study type: Longitudinal observational, 3,319 days, Mar 2010 – Apr 2019

Source: Zendle, Meyer & Ballou (2020). The changing face of desktop video game monetisation. PLOS ONE 15(5) e0232780 · Checked: full text (PMC + publisher), 2026-09-10

The people around them

About 4.7 people are affected for every person with problem gambling, once double-counting across shared families and friends is corrected; the widely cited 'six others' overstates it at population level

~4.7 affected others per person with problem gambling · low- and moderate-risk gamblers = 70% of harm 'vectors' · nearly 1 in 4 affected others are themselves at-risk gamblers

Measured on: Not reported in abstract: a reconciliation of published reports from gamblers and affected others · Not reported · Not reported in abstract

What the number is of: Affected people per person with problem gambling, population level

Study type: Network model reconciling gambler and affected-other reports, with clustering correction

Source: Browne, Tulloch & Woodbright (2026). From 'Six Others' to Edge Conservation: a network reconciliation of gambling harm to others. J. Gambling Studies. doi 10.1007/s10899-026-10521-3 · Checked: abstract, 2026-10-05

What helps

Most treatment-seeking patients who self-excluded on Sweden's national register gambled anyway

41 of 60 (68%) · of those, unlicensed sites 68% (28/41) · borrowed identity or land-based venues 22%

Measured on: Treatment-seeking patients at one regional gambling-disorder unit (Region Skåne, Malmö); 76% men · Adults 18+; not reported · Sweden

What the number is of: 41 of the 60 patients who had self-excluded (60 = 81% of 74 with data)

Study type: Retrospective chart review, Jan–Sep 2021

Source: Spelpaus self-exclusion breach study (Region Skåne). 85 included / 74 complete · Checked: full text, 2026-08-31

Numbers that don’t hold up

Figures that circulate in AI search summaries and shared posts, traced back as far as we could.

  • “Young men 18-30: 10% problem gambling vs 3% of the general population”
    The numbers come from a real 2024 US poll, but the comparison is wrong. The 3% is everyone, all ages. Among under-30s it was 10% of men and 7% of women.
  • “Young men are 3.4 times more likely than women to gamble at a problem level”
    Real, but about men and women of all ages, from a 2024 meta-analysis of 21 studies. It reports younger adults separately and never measures young men as a group, and the gap narrows in North America.
  • “Young men at risk are twice as likely to have anxiety and 2.5 times as likely to have depression”
    We couldn't find a source for these figures. The one study we found that followed young men over time saw higher anxiety go with lower odds of later problem gambling, and depression with slightly higher odds.
  • “The highest-risk period is ages 16 to 22”
    No research source found. It appears on a treatment provider's web page with no citation.
  • “84% of boys who gamble report that their friends also gamble”
    Turned around. The survey found that of boys whose friends gamble, over 8 in 10 gamble themselves, against under 2 in 10 of boys whose friends don't.
  • “An average of 5 to 10 people are negatively affected by each compulsive gambler”
    The research version is the widely cited 'six others'. A 2026 analysis that corrects for the same people being counted twice puts it nearer 4.7, with the harm concentrated in families.
  • “Many young men use gambling as a way to escape or numb underlying anxiety or depression”
    Half right at best. In the one study we found that followed young men over time, higher anxiety went with lower odds of later problem gambling; depression went with higher odds.
  • “Many young men view gambling as the only viable shortcut to financial mobility when savings and career paths feel out of reach”
    Presented as a finding, but the only thing behind it is a YouTube video. It may well be true; nobody has shown it.

What we’re checking next

One question a week, in this order. Each answered question becomes a journal entry with its sources.

  1. Are young men actually most at risk? What does the data show when sex and age are measured together? (open)
  2. What pulls them in, and why now? The phone, always-on apps, sports-betting ads (the games get their own block, Q8–Q11) (open)
  3. Do loot boxes lead to gambling, or just travel with it? The gaming–gambling bridge: does paying for random rewards cause later gambling, or do the same people do both? (open)
  4. Skins, esports and in-game betting: how does playing turn into wagering? Tradable items, skin gambling sites, betting on matches (open)
  5. What does watching gambling do? Streamers gambling on stream, casino content on gaming platforms, betting influencers (open)
  6. Are free 'social casino' games a rehearsal for the real thing? Play-money slots and poker inside games and apps (open)
  7. Do young men gamble because other routes to financial mobility feel closed? Precarity, side-hustle culture, betting and trading as the 'only shortcut' (open)
  8. Does "handle it yourself" keep it hidden? Masculine norms and help-seeking (open)
  9. Loneliness first, or gambling first, in young people? (open)
  10. What does someone who bets or trades with limits actually do differently, and what does it protect? (open)
  11. Who else gets hurt, and how? Partners, parents, children, friends (open)
  12. What does being open about it do for trust? (open)
  13. What does it cost beyond the household? Work, debt, crime, health services, public spending (open)
  14. Which safeguards actually work? Deposit and loss limits, self-exclusion, cooling-off tools (open)

If gambling is costing you or someone close to you more than it should, the US National Problem Gambling Helpline is free and confidential: call or text 1-800-MY-RESET (1-800-697-3738), or chat at ncpgambling.org.