Skip to content
Get started

The research behind Lessen.

Each part of Lessen comes from published research. This page lists every source, what it found and where it falls short. None of these studies tested Lessen itself, and Lessen doesn't diagnose or treat anything.

Every source was checked on October 19, 2026. Three labels help you read it:

Cuts the other way
A finding that limits the ones above it.
Abstract only
We could read only the paper's summary, and it leaves part of the study unclear.
Secondary copy
We read a faithful copy, not the official original.

The real number

In Lessen
Lessen reads your record from your connected books and shows it next to your own guess.
In short
Many bettors remember their results as better than they were. Not all do, and how far off they are varies.

  1. Only about 4 in 100 sports and race bettors guessed their last 30 days right, and about 2 in 3 thought they lost less than they did.

    What it found
    A survey went to 40,000 customers of an Australian sports and race wagering website, and answers were checked against their account records. Of 514 people who reported their 30-day net result, 21 (4.09%) were within 10% of the real figure and 333 (64.79%) underestimated their losses. Of 652 who reported how many bets they placed, 48 (7.36%) were within 10% and 454 (69.63%) underestimated their number of bets.
    Its limits
    One Australian operator, and only a small share of the 40,000 invited answered. People with smaller real losses made bigger errors in both directions, so this study does not show that the biggest losers guess worst. Recall covered 30 days.

  2. About 1 in 3 online bettors remembered their results as better than they were, and people with gambling problems were further off.

    What it found
    Customers of bwin.party's online service estimated their results for regular and live sports betting and for online casino, over the past 3 or 12 months, and their estimates were compared with their betting records. Depending on the period and game, between 34% and 40% on average remembered their results as better than they were, either losses smaller or wins bigger. The size of the gap was consistently linked to self-reported gambling problems, but the direction of the error was not.
    Its limits
    Volunteers from one European operator. The summary does not give the sample size; a later paper describes it as 2,259 bwin players. Only the size of the error, not its rosy direction, was linked to gambling problems.

  3. The typical player guessed their monthly losses a little too low, and people who spent more had a harder time guessing right.

    What it found
    Of 17,742 Norsk Tipping (Norway) online gamblers invited, 1,335 estimated their overall win or loss for May 2015, which was compared with their player-card records. The median player underestimated losses by 105 Norwegian kroner, about 15% of their average monthly loss, and players who spent more had more trouble estimating. Using a lenient rule (error smaller than one average month's loss), 74% counted as correct, 9% underestimated losses or overestimated wins, and 17% overestimated losses; sports bettors had the highest share of unfavorable estimates (16%).
    Its limits
    Only 7.5% of those invited answered, and high spenders were oversampled. One state-owned operator in Norway. Funded by Norsk Tipping, and both authors report consulting for gambling companies. Estimated and actual spend correlated at r = .35.

  4. Players guessed their monthly deposits fairly well on average, but people who deposited more were more likely to guess too low.

    What it found
    From a dataset of 1,516 customers of a European online gambling operator, 639 who had deposited in the past 30 days were analysed. They estimated their 30-day deposits fairly accurately overall, but the higher the amount deposited, the more likely they were to underestimate it; younger gamblers tended to overestimate. Telling people whether they had over- or underestimated did not lead to any additional significant change in deposits.
    Its limits
    One operator, and deposits rather than losses. A balancing source: memory is not always far off on average. Telling people they had guessed wrong added nothing measurable here.

Back to the list of sections

Plans you make once

In Lessen
Lessen asks you to plan ahead while you're calm: a weekly cap, and what you will do when a heads-up comes. Lessen can't make you keep the plan.
In short
Across hundreds of tests, plans written as 'if this, then that' helped people reach all kinds of goals. In gambling they have rarely been tested, and there is no field trial yet.

  1. In 94 tests, people who made 'if this happens, then I will do that' plans reached their goals more often.

    What it found
    Pooling 94 independent tests, making an if-then plan (of the form: if situation Y happens, then I will do X) had a positive effect of medium-to-large size on reaching goals, d = .65. The plans helped people start working on a goal, shield it from distractions, step back from a failing course of action and save energy for later goals.
    Its limits
    The tests covered many kinds of goals, and the paper's summary does not mention gambling. d = .65 is an average standardized difference, not a percentage. A newer review of 642 tests found a wider range of effects.

  2. A newer review of 642 tests found these plans help, and help more when written as 'if this, then that.'

    What it found
    Across 642 independent tests, if-then plans worked for thinking, feelings and behavior, with effects from d = .27 to d = .66. Effects were larger when the plan used an if-then format, when people were highly motivated to reach the goal, and when they rehearsed the plan.
    Its limits
    Not specific to gambling. Effects vary by outcome, from small to medium-large.

  3. In a lab betting game, getting worked up led to riskier bets, but not for people who had made an 'if-then' plan.

    What it found
    In a first experiment, an unpleasant mood made people without a plan more willing to take risks, but not people who had an if-then plan for unpleasant moods. In a second, which used a gambling task, getting worked up led to riskier betting in the comparison group, while an if-then plan that pointed attention to the chance of winning kept people aware of the risk and protected them from the effect.
    Its limits
    Short lab experiments with a gambling task, not real-money sports betting, and the summary does not give sample sizes. A 2022 review notes that if-then planning has rarely been tested in gambling, so there is no gambling field trial yet.

Back to the list of sections

A wait before loosening a limit

In Lessen
Deposit limits at your sportsbook come with waits like these. Lessen's protection plan helps you set them. Lessen itself can't stop a bet.
In short
Regulators build the same kind of wait into deposit limits: 7 days in Australia, 24 hours plus a second yes in Britain, and in New Jersey not until the current limit period ends.

  1. In Australia, online betting sites must lower a deposit limit right away, but must wait 7 days before raising it.

    What it found
    Measure 6 says limits must be binding and customers must be prompted to set a deposit limit at sign-up. It states: 'A request by a customer to decrease their deposit limit must be applied immediately' and 'A request by a customer to increase their deposit limit must not be applied until 7 days after the day the request was received.' Customers must also be prompted to set and review their limit at least once a year.
    Its limits
    Applies to licensed online betting providers in Australia through state and territory laws. The government server did not load for us, so we read the Internet Archive's saved copy of the same file (captured 12 April 2026). Tasmania's current government summary also says customers 'cannot increase a limit on the spot'.

  2. One Australian territory's rules say a betting site must lower a limit right away and must wait at least 7 days to raise it.

    What it found
    New section 1.20E(2) requires a licensee to set or lower a deposit limit immediately when asked, and section 1.20E(3) says a licensee 'must not increase the deposit limit for a person's betting account with the licensee until at least 7 days after the licensee receives a request to do so.' The statement explains the reasoning: lowering a limit reduces the risk of harm so it happens at once, while raising it increases risk so a cooling-off period applies.
    Its limits
    This is the explanatory statement for the 2019 change that put the national framework into ACT law. It describes the rule as made in 2019, which may differ from the current consolidated regulation.

  3. In Britain, a betting site must lower a limit right away, but can raise it only after at least 24 hours and a second yes from the player.

    What it found
    RTS requirement 12D: 'Customer-led limits must only be increased at the customer's request, only after a cooling-off period of at least 24 hours has elapsed and only once the customer has taken positive action at the end of the cooling off period to confirm their request.' It adds that 'customer-led reductions to limits must be implemented immediately' unless a technical failure prevents it.
    Its limits
    Applies to online gambling operators licensed in Great Britain. The British wait, 24 hours plus a confirmation, is much shorter than Australia's 7 days.

  4. Secondary copy

    In New Jersey, online sites must lower a limit by your next log in, but can raise it only after the current limit period ends.

    What it found
    Subsection (n) requires internet and mobile gaming systems to let patrons set responsible gaming limits, including daily, weekly and monthly deposit limits, and says: 'Any decrease to these limits shall be effective no later than the patron's next log in. Any increase to these limits shall become effective only after the time period of the previous limit has expired.'
    Its limits
    Read from the Cornell Legal Information Institute's copy, not the official New Jersey Register. The rule sits in the chapter on internet and mobile gaming, and we did not confirm how it applies to online sports betting accounts. The wait lasts until the current limit period ends, not a fixed number of days.

  5. A UK review said gambling blocks you can turn off any time are 'a light switch rather than a lock.'

    What it found
    Of eight UK bank card gambling blockers, three could be switched off immediately, one after 24 hours and four after 48 hours; the report says over a third can be toggled on and off at will, 'making them a light switch rather than a lock', and recommends every blocker be built around a time-released lock. Among surveyed people affected by gambling who saw a cooling-off period as important, 48 hours was the most common answer but 58% wanted it longer, and one bank with a 48-hour block reported that as many as a third of its users had turned it off and then gambled.
    Its limits
    About bank card blocks, not betting caps. The survey was of people affected by gambling problems, not a random sample. The 'at least 48 hours' wording comes from the university's press release of 7 July 2020; the report itself recommends a time-released lock.

  6. Secondary copy

    When a bank let people pick how long to wait before they could undo a gambling block, the average choice rose to 10 months.

    What it found
    After Monzo, working with GamCare, extended the longest cool-down on its gambling block (the time between switching the block off and being able to gamble) to a full year, the average cool-down people chose rose from three months to ten, and 41% chose the full 12 months. 8,500 customers used a 'note for future you' shown if they try to turn the block off.
    Its limits
    Usage numbers from the bank, reported by a charity partner. Not an independent evaluation, and not evidence that people gambled less overall. It shows that many block users choose long waits, not that long waits work better. GamCare's site blocks automated readers, so we read the Internet Archive's saved copy (29 December 2024).

Back to the list of sections

Your own numbers

In Lessen
Heads-ups and your weekly review show your real numbers.
In short
Showing people their own numbers has a small, mixed track record.

  1. Across 11 studies, showing people feedback about their own gambling helped by a small amount.

    What it found
    Reviewing 16 interventions from 11 studies, the authors found a small, statistically significant effect in favor of personalized feedback, d = 0.20. Feedback worked best for people with more severe gambling problems, with gambling education and with motivational interviewing. It worked less well in person without motivational interviewing, or when people were told their score on a gambling-severity questionnaire.
    Its limits
    Many of these interventions compared people with others, which is different from feedback on your own history. The effect is small.

  2. In the week after reading a message about their own play, online players cut back more than players who got no message.

    What it found
    Of 69,631 eligible Norsk Tipping online players, 17,452 were randomly selected and 5,528 took part. Compared with a no-message group, all five groups that got a message (feedback on their own play, a comparison with other players and/or a recommendation) significantly reduced their gambling. In the 7 days after reading, half of message readers cut their theoretical loss by at least 42%, against 36% for control players. Comparing people with other players did not work better than feedback on their own play alone.
    Its limits
    Reading the message was voluntary, which the authors say made the comparison with control players difficult. Over 30 days, theoretical loss rose slightly in the message groups. Comparing people with other players did not appear to help sports bettors. Norway, one operator.

  3. Players who used a feedback tool about their own play spent less time and money gambling than similar players who did not.

    What it found
    The study compared 1,015 online gamblers at a European site who got feedback from mentor, a tool that gives personalized feedback, with 15,216 similar players matched on age, gender, how long they had played and theoretical loss. Those who got feedback spent significantly less time and money gambling than the matched players.
    Its limits
    Not randomized: players chose to use the tool, so they may differ from controls in ways matching cannot fix. Data came from a gambling operator.

  4. Abstract only

    Swedish online players bet less on the day they read a message about their own betting, and in the week after.

    What it found
    Among 7,134 Swedish online gamblers who played at one of five ComeOn Group sites between July 2019 and January 2020, those who received personalized feedback text messages about their actual gambling wagered significantly less money on the day they read a message and seven days after.
    Its limits
    We could read only the summary, and it does not say how the comparison was built, so this is not proof that the messages caused less betting. Short follow-up. Operator data.

  5. Abstract only

    Casino players who had lost more than they thought, and then saw their real numbers, bet and lost less over the next 3 months.

    What it found
    649 players in a casino loyalty program were asked how much they had won or lost over three months while using their loyalty card, and were then shown their player-account data. Players who had underestimated their losses did not feel they had cut back, but their account data showed they significantly reduced both the amount wagered and the amount lost over the next three months.
    Its limits
    Casino loyalty players, not online sports bettors. We could read only the summary. The key result is for the people who had underestimated, and no randomized comparison group is described, so it shows change after feedback, not proof of cause.

  6. Cuts the other way

    Many bettors opened their monthly betting statements and some said they helped, but their betting did not change when statements arrived.

    What it found
    Since July 2022 Australia has required online wagering operators to send activity statements to active customers. In a survey of 1,647 customers of two operators (85% male), matched to their account data, 57.2% opened statements at least sometimes; of those, 17.6% said statements decreased their gambling and 0.8% said they increased it. There were no significant changes in betting frequency or deposit amount around the time statements were sent, and reading statements was not linked to better recall of spending or net result.
    Its limits
    Statements are monthly operator messages, not timed heads-ups, so this does not test Lessen's design directly. Still, it is a warning that a recap on its own may not change behavior. The first author reports past funding from gambling operators and government bodies.

  7. Cuts the other way

    A pop-up showing players their losses did not help them keep to their limit, and about half could not remember what it said.

    What it found
    In two studies at a gambling venue, with 124 and 109 players, people played a simulated gaming machine with a money limit. Adding feedback about the money and credits they had lost to the limit-reached pop-up did not help them keep to the limit, and about 50% of players in both groups could not recall what the pop-up said.
    Its limits
    Simulated slot machine, small samples, single sessions. The lesson is about message design: a message people do not read cannot help.

Back to the list of sections

Limits set ahead of time

In Lessen
Your weekly cap is a limit you set ahead of time. Lessen shows how much room is left but can't stop a bet, so for a hard stop, pair it with your sportsbook's deposit limits.
In short
People who set limits in advance often go on to bet less, though most of this evidence comes from watching players, not fair tests.

  1. A short message got more sports bettors to set a deposit limit, and those who set one went on to bet less.

    What it found
    In a pre-registered trial with 26,560 eligible customers of four Australian online sports and racing betting sites, 161 (0.71%) of those sent messages set a deposit limit within 5 days, compared with 3 (0.08%) of the comparison group. Customers who set limits significantly cut their average daily bet, how much it varied, their net loss and how intensely they bet, compared with customers who did not.
    Its limits
    The messages were randomized, but setting a limit was not, so the drop among limit-setters is a link, not proof. Fewer than 1 in 100 people set a limit even after a message.

  2. Among the heaviest online players, those who set their own money limits were betting less a year later than those who did not.

    What it found
    Using anonymized data on 49,560 players who had placed at least one wager with the operator Kindred, the study asked whether voluntarily setting money limits was linked to spending over one year. Among the top 10% most gambling-intense players, those who had set limits gambled significantly less money a year later than those who had not; there were no differences by age or gender.
    Its limits
    Observational: players chose whether to set limits, so this is a link, not proof that limits caused the drop. One operator. Two authors were affiliated with the company neccton GmbH.

  3. Among the heaviest online players, setting a money limit was followed by less spending for casino and lottery players.

    What it found
    On the Austrian site win2day, every player must set time and cash-in limits (no more than 800 euros a week) when opening an account. Tracking voluntary limit-setting by 5,000 players over 3 months and focusing on the 10% most intense players, voluntary spending limits had the largest significant effect on later spending among casino and lottery players, and poker players' spending fell after setting a time limit.
    Its limits
    Observational and self-selected; casino, lottery and poker rather than sports betting; inside a system where some limit was already mandatory.

  4. Most players in Norway said they liked having a loss limit, and very few went to other sites after hitting it.

    What it found
    After Norsk Tipping introduced a mandatory global loss limit across its games, 2,352 of its players were surveyed. Three-quarters knew about the new limit, two-thirds knew how to set limits, four-fifths had a positive attitude toward the global loss limit, and very few said they gambled with other operators after reaching their limit.
    Its limits
    Opinions and self-report, not measured betting, and the limit was required, not chosen. It shows people accept limits, not that limits work.

  5. Cuts the other way

    Asking new players to set a deposit limit got more of them to set one, but did not change how much they lost.

    What it found
    New customers of a publicly governed Finnish gambling operator (4,328 whose preferred game was online slots) were randomly given a deposit-limit prompt at sign-up, before or after their first deposit, or no prompt, and followed for 90 days. Prompted groups set limits far more often, but their net loss did not differ from the no-prompt group. Customers who later raised or removed a limit lost more than limit-setters who kept theirs.
    Its limits
    Slot players, not sports bettors, and a prompt to set a limit of any size is different from a cap you design yourself. The raise-or-remove finding is a link, not proof, though it fits with rules that slow down loosening.

  6. Cuts the other way

    An early review found most players liked setting limits ahead of time, but the studies were mixed and had weak methods.

    What it found
    The review found 17 relevant publications, some not peer reviewed, mostly on limit-setting systems for gaming machines in Norway, a Canadian province and Australia. Most gamblers were positive about setting limits ahead of time, though non-problem and low-risk gamblers saw it as unnecessary for themselves. Findings on keeping to money limits and on spending were mixed, few people used time limits, and low take-up, card sharing and gambling outside the trials limited what could be concluded.
    Its limits
    Older review focused on gaming machines, written before most online account-data studies.

  7. Cuts the other way

    A review found limits people choose for themselves have little proof behind them, mostly because few people use them.

    What it found
    Updating earlier reviews with a decade of trials, online limit-setting studies and Norway's required limits, the authors agreed with earlier reviews that limits people choose for themselves have little proof behind them, because few people use them and the studies have weaknesses. They found possible benefits in required limits, some uses for voluntary ones, and possible risks in required limits across all games.
    Its limits
    Griffiths and Auer published a reply (opens in a new tab) disputing parts of this review (International Gambling Studies, 2022, 22(3), 337-343).

Back to the list of sections

In control, without quitting

In Lessen
Lessen Mode keeps you betting with a weekly number. If you ever want a full break, Quit Mode counts your days. You can switch any time.
In short
In treatment studies, people who chose to cut down often did about as well as people who chose to quit, though quitters bet on fewer days. Many changed their goal along the way.

  1. People aiming to cut down did not bet more money than people aiming to quit, and almost half changed their goal at least once.

    What it found
    In further analyses of a randomized trial of brief motivational treatment, almost half of participants switched their goal at least once. Over 25% kept a steady goal of quitting their most problematic type of gambling, almost 20% of quitting all gambling, and about 10% of gambling in a controlled way. People aiming to quit gambled on significantly fewer days, but goal choice was not related to dollars gambled, dollars per day or how well people felt they met their goal. The authors conclude the findings do not support the idea that quitting goals are better than cutting-down goals.
    Its limits
    Goals were chosen by the participants, not randomized. Quitting goals did mean fewer betting days, so cutting down did not match quitting on every measure.

  2. In a small study of women in treatment, those who chose to cut down did about as well as those who chose to quit.

    What it found
    Women with pathological gambling in a cognitive-behavioural treatment program chose either abstinence or controlled gambling as their goal. Those choosing controlled gambling showed comparable improvement: among treatment completers, 89% of the abstinence group and 82% of the controlled-gambling group no longer met diagnostic criteria for pathological gambling at the 6-month follow-up.
    Its limits
    Women only, a preliminary study, goals chosen by the women rather than randomized, and the percentages count only people who finished treatment. The summary does not give group sizes.

  3. Years after treatment, both people who quit and people who gambled in a controlled way kept doing better.

    What it found
    63 of 120 pathological gamblers who had completed behavioural treatment years earlier were reassessed with interviews and standard questionnaires. Both abstinent and controlled-gambling outcomes were associated with continued improvement in social functioning and psychological measures, while people whose gambling stayed uncontrolled did not improve; the authors concluded that abstinence is not the only possible outcome and that controlled gambling was not a temporary state followed by a return to uncontrolled gambling.
    Its limits
    An old study (1991) of a small follow-up group (about half of those originally treated); not randomized.

  4. In a program that aimed for controlled gambling, about half of everyone who started was below the gambling-disorder line a year later.

    What it found
    89 pathological gamblers received cognitive-behavioural treatment aimed at controlled gambling. Counting everyone who started (intent-to-treat), 63% had a DSM-IV score of 4 or less at the end of treatment, 56% at 6 months and 51% at 12 months; among those who completed treatment the figures were 92%, 80% and 71%. Few variables predicted who would do well with control rather than abstinence.
    Its limits
    No comparison group, so it shows controlled gambling can be a workable goal for some people, not that it beats quitting.

  5. Of people who had gotten over a gambling disorder, 9 in 10 still gambled some in the past year.

    What it found
    In an Australian community twin survey of 4,764 people, 104 had a lifetime history of pathological gambling and 44 of them had no symptoms in the past year ('recovery'). Ninety per cent of the recovery group had taken part in some form of gambling in the past year, and the authors conclude: 'Controlled gambling appears to be a popular road to recovery in the community.'
    Its limits
    A community survey, not a treatment study; 44 people in the recovery group; 'recovery' means no past-year symptoms, and these people were not necessarily following any plan.

Back to the list of sections

When betting happens

In Lessen
Lessen finds your own riskiest hours in your record and sends a heads-up before them.
In short
Across all bettors in New Jersey, evenings were the busiest time to bet. These reports describe everyone, so Lessen goes by your own hours instead.

  1. In New Jersey, most sports bets were placed between 9 a.m. and midnight, and 6 to 9 p.m. was the busiest time.

    What it found
    Analysing operator data on about 17.3 million sports bets (about $849 million) placed in New Jersey in 2018, the report found about 87% of bets were placed between 9 a.m. and midnight, with the largest share from 6 to 9 p.m. (24.1%) and the next from 3 to 6 p.m. (18.7%). More than half of in-game bets were placed between 6 p.m. and midnight, and high-intensity bettors' average bet was highest from 9 p.m. to midnight ($87).
    Its limits
    Covers only the first five months of legal betting in New Jersey, and two operators' data were excluded. A state-commissioned report, not peer reviewed. It shows when bets are placed across everyone, not each person's own risky hours, and it does not report days of the week. Despite the 'Calendar Year 2019' title, the data are 2018 bets.

  2. In New Jersey, more than half of bets placed during games were made between 6 p.m. and midnight.

    What it found
    Of more than 43 million sports bets placed in New Jersey in 2019, more than half of all in-game bets were placed between 6 p.m. and midnight, and almost two-thirds (65.5%) of bets placed between 9 p.m. and midnight were in-game bets. High-intensity bettors' average bet spiked in the 9 p.m. to midnight window.
    Its limits
    A state-commissioned report, not peer reviewed; describes all bettors in one state. Despite the 'Calendar Year 2020' title, the data are 2019 bets.

  3. When people who had just quit gambling slipped, it was most often in the evening, when they were alone and thinking about money.

    What it found
    Following 101 pathological gamblers who had recently quit, only 8% were entirely free of gambling during the 12-month follow-up. Relapses varied but happened most often in the evening, when the person was alone and thinking about finances; moods beforehand were as likely to be positive as negative, and the most common explanations were thoughts about winning and feeling the need to make money.
    Its limits
    People trying to quit gambling in general, not sports bettors specifically, studied about 20 years ago. It is about when slips happen, not about game schedules.

Back to the list of sections

Why people go it alone

In Lessen
Lessen is a private tool you can use on your own. It is not treatment, and it doesn't replace a counselor or the helpline.
In short
Most people with a gambling problem never get formal treatment. Common reasons are wanting to handle it alone, and shame.

If you need help now, call or text 988, or call 1-800-522-4700 or 1-800-MY-RESET.

  1. In two big US surveys, only 7 to 12 in 100 people who ever had a gambling disorder had gotten treatment or gone to Gamblers Anonymous.

    What it found
    Among people who had ever had pathological gambling in two US national surveys (21 people in one, 185 in the other), only 7% to 12% had ever sought formal treatment or gone to Gamblers Anonymous. Even so, 36% to 39% had no gambling problems in the past year, and about one-third had recovered without treatment.
    Its limits
    Data from the late 1990s and early 2000s, before legal online sports betting in the US. The smaller survey had only 21 people with the disorder. Formal treatment or GA leaves out self-help.

  2. In Ontario, about 7 in 10 people with a gambling disorder had never gotten help from a professional or Gamblers Anonymous.

    What it found
    Of 1,205 Ontario adults who screened positive for a possible gambling problem, 6% had ever used any help including self-help materials, and 3% had sought treatment when self-help materials were excluded. Help-seeking rose with severity: among the 57 people meeting criteria for pathological gambling, 53% had used some help including self-help and 29% had used treatment, and the authors note that 71% had never sought treatment from professionals or attended Gamblers Anonymous.
    Its limits
    The often-quoted 3% covers everyone who screened positive on a short three-question screen (824 of the 1,205 met only that screen), not people with a gambling disorder. The group with a disorder was small (57), and it was a telephone survey with a fairly low response rate.

  3. Across 24 studies, about 1 in 5 people with gambling problems had looked for any kind of help.

    What it found
    Pooling 24 representative population studies with data from 2010 onward, help-seeking of any kind (professional, non-professional or self-help) was 20.63% among people with problem gambling, 3.73% among moderate-risk gamblers and 0.27% among low-risk gamblers. The authors sum up: 1 in 25 moderate-risk gamblers and 1 in 5 people with problem gambling have sought help.
    Its limits
    Help-seeking was defined broadly (including family, friends and self-help), so formal treatment is lower still. Measures varied across studies and overall risk of bias was high.

  4. The most common reasons people gave for not getting help were wanting to handle it alone, shame, and not wanting to admit the problem.

    What it found
    Reviewing 19 studies from five countries (published since 1998) in which gamblers were asked why they had not sought help, the most commonly reported barriers were wanting to handle the problem oneself; shame, embarrassment or stigma; unwillingness to admit the problem; and issues with treatment itself. Lack of knowledge about treatment options and practical problems with attending treatment were also often reported.
    Its limits
    A review of reasons, not rates. It does not show that an app removes these barriers.

Back to the list of sections

Sources behind the guides

Our guides cite these sources too. Where a guide uses one of the studies above, that entry says which guide.

Patterns and risk in betting records

  1. Auer, M., & Griffiths, M. D. (2023). Predicting high-risk gambling based on the first seven days of gambling activity after registration using account-based tracking data. International Journal of Mental Health and Addiction, 22(6), 3397-3413 (published online 2 May 2023). (opens in a new tab)

    Cited in What your betting history already shows about you and Why the pull shows up at the same times.

  2. Finkenwirth, S., MacDonald, K., Deng, X., Lesch, T., & Clark, L. (2021). Using machine learning to predict self-exclusion status in online gamblers on the PlayNow.com platform in British Columbia. International Gambling Studies, 21(2), 220-237 (published online 8 November 2020). (opens in a new tab)

    Cited in What your betting history already shows about you and Why the pull shows up at the same times.

  3. Perrot, B., et al. (2018); Challet-Bouju, G., et al. (2020); Balem, M., et al. (2022). Three studies of player account data that measure chasing.

    They count, for example, three or more deposits within 12 hours, or a deposit less than an hour after a bet.

    Cited in The first 24 hours after going over your cap, What your betting history already shows about you and Why the pull shows up at the same times.

  4. Braverman, J., & Shaffer, H. J. (2012). How do gamblers start gambling: Identifying behavioural markers for high-risk internet gambling. European Journal of Public Health, 22(2), 273-278. (opens in a new tab)

    Cited in How betting risk builds, and how to spot it early.

  5. Allami, Y., Hodgins, D. C., Young, M., Brunelle, N., Currie, S., Dufour, M., Flores-Pajot, M.-C., & Nadeau, L. (2021). A meta-analysis of problem gambling risk factors in the general adult population. Addiction, 116(11), 2968-2977. (opens in a new tab)

    Cited in How betting risk builds, and how to spot it early.

  6. Ferris, J., & Wynne, H. (2001). The Canadian Problem Gambling Index: Final report. Canadian Centre on Substance Abuse. (opens in a new tab)

    The report that introduced the Problem Gambling Severity Index, the nine-question screen.

    Cited in How betting risk builds, and how to spot it early.

  7. Calado, F., & Griffiths, M. D. (2016). Problem gambling worldwide: An update and systematic review of empirical research (2000-2015). Journal of Behavioral Addictions, 5(4), 592-613. (opens in a new tab)

    Of 69 prevalence studies from 2000 to 2015, 23 used the South Oaks Gambling Screen and 21 each used the Problem Gambling Severity Index and the DSM-IV criteria, so the PGSI is one of the most used, not the most used.

    Cited in How betting risk builds, and how to spot it early.

  8. Shaffer, H. J., Peller, A. J., LaPlante, D. A., Nelson, S. E., & LaBrie, R. A. (2010). Toward a paradigm shift in Internet gambling research: From opinion and self-report to actual behavior. Addiction Research & Theory, 18(3), 270-283. (opens in a new tab)

    Cited in Apps that connect to your sportsbook.

Late nights and live betting

  1. Thorne, H. B., Rockloff, M. J., Ferguson, S. A., Vincent, G. E., & Browne, M. (2021). Gambling problems are associated with alcohol misuse and insomnia: Results from a representative national telephone survey. International Journal of Environmental Research and Public Health, 18(13), 6683. (opens in a new tab)

    Found a small but real link between gambling problems and insomnia.

    Cited in Why the pull shows up at the same times.

  2. Hamel, A., Bastien, C., Jacques, C., Moreau, A., & Giroux, I. (2021). Sleep or play online poker? Gambling behaviors and tilt symptoms while sleep deprived. Frontiers in Psychiatry, 11, Article 600092 (volume dated 2020). (opens in a new tab)

    23 regular online poker players. Sessions played short on sleep had more tilt, more hands and worse results.

    Cited in Why the pull shows up at the same times.

  3. Russell, A. M. T., Hing, N., Browne, M., Li, E., & Vitartas, P. (2019). Who bets on micro events (microbets) in sports? Journal of Gambling Studies, 35(1), 205-223. (opens in a new tab)

    A survey of 1,813 Australian sports bettors, not account data.

    Cited in Why the pull shows up at the same times.

Blocks and support

  1. Talbot, A., King, K., Tarplin, L., Victor, L., & Rodger, J. (2018, October). Evaluating online blocking software. Winning Moves Ltd for GambleAware. (opens in a new tab)

    Cited in How to take a break from sports betting and Lessen vs Gamban.

  2. Kantar for GamCare, Gamban and GAMSTOP (2026, June). TalkBanStop independent evaluation.

    Known only from trade press coverage. We could not find or open the report itself, and its 68% figure is what people said about themselves.

    Cited in How to take a break from sports betting and How to use Lessen well.

  3. Petry, N. M., & Weiss, L. (2009). Social support is associated with gambling treatment outcomes in pathological gamblers. American Journal on Addictions, 18(5), 402-408. (opens in a new tab)

    Cited in How to use Lessen well and The first 24 hours after going over your cap.

  4. Jiménez-Murcia, S., et al. (2017). The involvement of a concerned significant other in gambling disorder treatment outcome. Journal of Gambling Studies, 33(3), 937-953. (opens in a new tab)

    Cited in How to use Lessen well.

  5. Schuler, A., Ferentzy, P., Turner, N. E., Skinner, W., McIsaac, K. E., Ziegler, C. P., & Matheson, F. I. (2016). Gamblers Anonymous as a recovery pathway: A scoping review. Journal of Gambling Studies, 32(4), 1261-1278. (opens in a new tab)

    Cited in How to use Lessen well.

  6. Ribeiro, E. O., Afonso, N. H., & Morgado, P. (2021). Non-pharmacological treatment of gambling disorder: A systematic review of randomized controlled trials. BMC Psychiatry, 21, Article 105. (opens in a new tab)

    Cited in How to use Lessen well.

  7. Marionneau, V., Ruohio, H., & Karlsson, N. (2023). Gambling harm prevention and harm reduction in online environments: A call for action. Harm Reduction Journal, 20, Article 92. (opens in a new tab)

    Cited in How to use Lessen well.

Slips, goals and habits

  1. Collins, S. E., & Witkiewitz, K. (2013). Abstinence violation effect. In M. D. Gellman & J. R. Turner (Eds.), Encyclopedia of Behavioral Medicine (pp. 8-9). Springer. (opens in a new tab)

    An encyclopedia entry.

    Cited in The first 24 hours after going over your cap.

  2. Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998-1009. (opens in a new tab)

    We could read only the summary. It reports 18 to 254 days for a new habit to become automatic; the study ran 12 weeks, so the longest times are projections.

    Cited in How to use Lessen well.

  3. Locke, E. A., & Latham, G. P. (1990). A theory of goal setting & task performance. Prentice Hall.

    Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57(9), 705-717. (opens in a new tab)

    Locke, E. A., & Latham, G. P. (2019). The development of goal setting theory: A half century retrospective. Motivation Science, 5(2), 93-105. (opens in a new tab)

    Goal-setting theory: a book and two reviews. Not about gambling.

    Cited in How to use Lessen well.

  4. Harkin, B., Webb, T. L., Chang, B. P. I., Prestwich, A., Conner, M., Kellar, I., Benn, Y., & Sheeran, P. (2016). Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence. Psychological Bulletin, 142(2), 198-229. (opens in a new tab)

    Cited in How to take a break from sports betting and How to use Lessen well.

  5. Matthews, G. (2007). The impact of commitment, accountability, and written goals on goal achievement. Dominican University of California.

    A conference talk, not peer reviewed: 149 of 267 people finished, over four weeks. Not about gambling.

    Cited in How to use Lessen well.

  6. Nelson, R. O., & Hayes, S. C. (1981). Theoretical explanations for reactivity in self-monitoring. Behavior Modification, 5(1), 3-14. (opens in a new tab)

    Cited in How to use Lessen well.

Apps and how people use them

  1. Eysenbach, G. (2005). The law of attrition. Journal of Medical Internet Research, 7(1), e11. (opens in a new tab)

    Cited in How to use Lessen well.

  2. Baumel, A., Muench, F., Edan, S., & Kane, J. M. (2019). Objective user engagement with mental health apps: Systematic search and panel-based usage analysis. Journal of Medical Internet Research, 21(9), e14567. (opens in a new tab)

    Cited in How to use Lessen well.

  3. Torous, J., Lipschitz, J., Ng, M., & Firth, J. (2020). Dropout rates in clinical trials of smartphone apps for depressive symptoms: A systematic review and meta-analysis. Journal of Affective Disorders, 263, 413-419. (opens in a new tab)

    Cited in How to use Lessen well.

  4. Linardon, J., Torous, J., Messer, M., Liu, C., Bell, I. H., Nicholas, J., & Goldberg, S. B. (2026). Association between user engagement and clinical outcomes in smartphone apps for depression and anxiety: A systematic review and meta-analysis. Psychiatry Research, 355, 116864. (opens in a new tab)

    Cited in How to use Lessen well.

  5. Yu, L. Q., Amato, M. S., Papandonatos, G. D., Cha, S., & Graham, A. L. (2024). Predicting early dropout in a digital tobacco cessation intervention: Replication and extension study. Journal of Medical Internet Research, 26, e54248. (opens in a new tab)

    Cited in How to use Lessen well.

  6. Baron, K. G., Abbott, S., Jao, N., Manalo, N., & Mullen, R. (2017). Orthosomnia: Are some patients taking the quantified self too far? Journal of Clinical Sleep Medicine, 13(2), 351-354. (opens in a new tab)

    Cited in How to use Lessen well.

Back to the list of sections

Spot an error, a retraction or a stronger study we should cite? Tell us. We'd rather fix the page than defend it.