The Gambling Commission's shocking new betting survey is drawing heavy criticism - so why is it still being published?

• Read part one of our special report here

Later this week a bombshell new survey is expected to generate dark headlines warning that Britain is facing an explosion of gambling harm, with the rate of problem betting more than eight times higher than previously forecast.

The real problem, warn critics, is that the methodology of this survey is likely to significantly overstate the true level of harm, with potentially profound consequences for punters and a horseracing industry already reeling from the effects of affordability checks.

Even the Gambling Commission, which will publish the first annual report of the Gambling Survey for Great Britain (GSGB) on Thursday, has taken the extraordinary step of warning that it should not be compared to previous surveys and cannot be used "to calculate an overall rate of gambling-related harm" – in effect, an admission that the statistics set to be published are potentially unreliable.

The GSGB will measure gambling participation and problem gambling prevalence rates and replaces the NHS Health Surveys, which the Gambling Commission itself had described as once being 'gold standard' but now calls outmoded, as official statistics.

It is expected to report substantially higher rates of problem gambling compared with those NHS surveys. In November, during the experimental statistics stages of its development, 2.5 per cent of respondents scored eight plus on the Problem Gambling Severity Index (PGSI) screen – used to measure different levels of problem gambling – compared to the 0.3 per cent on the same screen reported in the Health Survey for England 2021.

Major doubts have been raised over the reliability of the figures with concerns they substantially overestimate the level of gambling participation and gambling harms.

The Gambling Commission has said the GSGB should not be compared to previous surveys or used to extrapolate a total figure for those suffering from gambling harm. But campaigners and elements of the media have already used provisional data to do just that and justify calls for tougher regulation of the gambling industry.

The new figures could have a huge impact on punters and horseracing. The problem gambling rate underpins policies such as affordability checks and provides a foundation for the thresholds being proposed.

A rise in the headline problem gambling rate figure could clearly result in campaigners and politicians calling for affordability check thresholds to be lowered from the rates proposed in the white paper.

Online bookmakers are also required by the Gambling Commission to conduct safer gambling interactions based on the problem gambling rate. An increase in the rate would seem likely to result in more gamblers being faced with intrusive questions by bookies trying to hit a quota.

Online bookmakers are required to conduct safer gambling interactions with customers based on the problem gambling rateCredit: Edward Whitaker (racingpost.com/photos)

The aim of the GSGB is to collect data on gambling behaviours, including participation and the impact of gambling "both positive and negative" from 20,000 individuals aged 18 years and over annually.

The new survey is principally conducted online, whereas the old NHS Health Survey was typically carried out through face-to-face interviews.

That is important – because academics argue that online surveys produce overestimates of problem gambling prevalence.

In 2022 Professors Patrick Sturgis and Jouni Kuha of the London School of Economics reviewed differences in prevalence survey estimates in the journal Public Health.

They wrote: "Surveys conducted online produce substantially higher estimates of problem gambling compared with in-person interview surveys. This is because online surveys, whether using probability or non-probability sampling, overrepresent people who are more likely to gamble online and to gamble frequently."

The Gambling Commission itself has previously acknowledged the issues with online surveys and also gambling-specific surveys over-recruiting gamblers – and heavily engaged gamblers at that.

Its 2020 consultation on "gambling participation and problem gambling prevalence research" noted that it was "important to ensure that, if moving to a gambling-specific survey, it does not attract an over-representation of gamblers or problem gamblers".

In February the Gambling Commission published a press release which said the development of the GSGB had been endorsed by Sturgis as "exemplary in all respects".

It quoted Sturgis as saying: “The Gambling Commission has engaged with a broad range of stakeholders and followed industry standards of best practice in developing a survey design that can be expected to yield high quality and timely estimates of gambling prevalence in Great Britain."

However, and adding fuel to those who believe it is partial in how it uses evidence and statistics, the Gambling Commission's press release did not mention one salient comment from Sturgis's independent review of the GSGB.

That comment was: "Until there is a better understanding of the errors affecting the new survey’s estimates of the prevalence of gambling and gambling harm, policy-makers must treat them with due caution, being mindful to the fact there is a non-negligible risk that they substantially overstate the true level of gambling and gambling harm in the population."

Sturgis also recommended that the commission take seven further steps to review the accuracy of the GSGB data.

Although the Gambling Commission did not reveal Sturgis's comments at the time, there was at least a recognition of them.

An email from the commission to Sturgis in January, taken from a Freedom of Information (FOI) request and published on the regulator's website, admits that estimates from the GSGB “might be on the high side”; and that “the focus should be less on the number itself but more on the patterns within the data and the trends going forward”.

More recently the Gambling Commission has acknowledged Sturgis's comments while pushing back against those raising concerns.

In a recent speech the regulator's deputy chief executive Sarah Gardner said "we note the risks he identified in having a new methodology and the caution that should be applied when seeking to draw precise conclusions".

However, Gardner added she had "seen some seeking to decry the GSGB before it has even been published in full", adding: "There is an obligation on all of us to use research, statistics and insight in a responsible way."

The Gambling Commission last week published guidance as to how the statistics contained within the GSGB should be used, and how they should not be used.

Among the ways the regulator said the data can be used was to assess future trends and changes in gambling behaviour, measured against the 2024 baseline.

The commission said the survey could be used "with some caution" to provide estimates of PGSI scores among adults but, rather confusingly, not to gross up the prevalence of problem gambling for the whole population until further work is completed. Neither can it be used to provide direct comparisons with results from prior surveys and as a measure of addiction to gambling.

That would seem a vain hope.

In April, leading gambling reform campaigner Lord Foster of Bath referenced the publication of the survey during a debate on gambling advertising in the House of Commons.

He added: "The gambling minister in the other place has already indicated that they are likely to show that 1.3 million people will classify as 'problem gamblers' and that a further six million are at risk. 

"If confirmed, these figures are far higher than those used to inform the government’s work on their white paper. This is a real cause for concern, further strengthening the call for action."

That same 1.3 million figure was quoted in an opinion piece in the Times newspaper by Liberal Democrat parliamentary candidate Edward Lucas.

Gambling operators believe such comments are only the beginning.

One betting industry source told the Racing Post: "Anti-gambling campaigners are lining up waiting for this research so they can demand even tougher affordability checks and outright bans." 

Last month Gambling Commission chief executive Andrew Rhodes was asked about the GSGB at a conference organised by law firm CMS.

Gambling Commission chief Andrew Rhodes has said he doubts the survey results will lead to stricter regulation

He said he doubted that the survey's results would lead to stricter regulation.

"There are people who are worried that because somebody might say something about what they think it means there will be a sudden lurch in regulation," he said. "That doesn't seem at all likely to me because we do understand the methodology."

However, analysts at investment bank Peel Hunt do not agree. They believe that the GSGB's figures will lead to calls for more restrictive regulation.

They said in a recent research note: "The commission did not seem to engage with the substance of what it was trying to measure and whether it mattered that it was reporting a measurement of a problem a multiple higher than had previously been thought accurate."

They added: "The new figures on gambling harm and prevalence are likely to feed into a narrative that gambling problems in the UK are much worse than previously thought and used as a lever to restrict activity further.

"It is only specialists who are likely to want to understand the detail and challenge whether this narrative reflects reality."

It is not yet clear what impact a new problem gambling rate far higher than those published through previous surveys will have on policy.

The Gambling Commission's own formal guidance does, however, link problem gambling rates for gambling activities to the number of customer interactions an operator must carry out.

Gambling operators fear that will lead to a leap in the number of punters who will need to be contacted, whether that be a phone call, encouragement to take up gambling management tools or setting time and spend limits.

As another industry source put it: "We are expected to interact with customers who might be at risk of harm in line with the percentage rates of at-risk behaviour.

"If that figure jumps, there will be a lot more customers we will have to interact with who may well not be at risk of harm at all, because we know that the survey itself is inaccurate, which will obviously irritate them and increase the friction on them.

"Not only that, operators would have to invest ten times more in the number of colleagues they have dealing with this when they are not actually solving any true problem because the problem doesn't exist."

Any talk of friction again brings the debate about affordability checks to the fore. Problem gambling rates have been used to inform the thresholds being used in the government's proposed financial risk checks.

There are huge concerns that a higher problem gambling rate under the GSGB will put downward pressure on those thresholds with consequences for not only punters, many of whom have been driven to the black market by affordability checks or away from racing altogether, but to the income of British racing, which is reeling from the impact of the checks already introduced and fears a further £250 million hit to its finances over the next five years.

Dan Waugh of analysts Regulus Partners said: "It seems likely that the GSGB will lead to a reduction in the thresholds used for affordability checks given that the commission claims to have calculated them by reference to the NHS’s far more conservative estimates of ‘problem gambling’ prevalence. It is also likely to lead to higher quotas for operators to conduct safer gambling interactions, which are pegged directly to the prevalence estimate."

The Racing Post contacted the Department for Culture, Media and Sport to ask if it was happy with the reliability of the GSGB but was told it would not comment until the report is published.

The Gambling Commission was also contacted for comment for this article but instead pointed the Racing Post to the guidance it released last week.

However, its chief executive told the CMS conference that he believed the GSGB was "the right way to go", although he admitted "all methodologies have limitations".

He added: "The commission has said all along this is a new methodology, it has been independently reviewed, it has got some limitations as the other one did, it keeps being refined but it will give us a much greater granularity, sample size and robustness than we can have by persisting with the methodology that has become outmoded."

The release of the GSGB annual report is, however, sure to reignite the debate about gambling and while the regulator says it may know how it should be treated, campaigners for reform are unlikely to pass up the opportunity to leap upon the publication of problem gambling figures which are almost certainly inflated.

Once the genie is out of the bottle it will be hard to put back in.

Read these next

How debunked stats and a selective approach to publishing research has misled the gambling reform debate 

Gambling Commission's reluctance to publish affordability survey results should be concerning to all 

Revealed: Gambling Commission survey demonstrated considerable opposition to affordability checks 

'The whole thing hasn't gone away' - wary response to Gambling Commission's six-month pilot for affordability checks 

Subscribe to Racing Post Members' Club Ultimate Monthly and get 50% off your first three months! 

Available to new subscribers purchasing Members' Club Ultimate Monthly using code MEMBERS24. First three payments will be charged at £24.98, subscription renews at full monthly price thereafter. To cancel please contact us at least seven days before subscription is due to renew.