Pokies? Lotto? Sports betting? Which forms of problem gambling affect Australians the most?

Gambling, especially sports and race betting, is a hot political issue at the moment.

This is largely due to the recommendations from a 2023 report from a nonpartisan federal government committee, chaired by the late Peta Murphy, called You Win Some, You Lose More.

This report recommended “the Australian government, with the cooperation of the states and territories, implement a comprehensive ban on all forms of advertising for online gambling”.

This has led to lots of debate and controversy.

Recently, Peter V’landys, head of the NRL and Racing NSW, claimed lotteries were more harmful than race and sports betting combined, citing independent statistics.

Let’s explore the relative harm of different types of gambling and see if this claim holds up.

Australians love a punt

Gambling is widespread in Australia, with more than half of adults engaging in at least one form each year.

According to the latest national data, lotteries are the most common type (40% of Australians buy a ticket annually), followed by race betting (17%), pokies (16%), scratchies (15.7%) and sports betting (9.6%).

However, the popularity of a gambling form doesn’t necessarily reflect its harm. Different gambling activities have distinct characteristics.

Two key factors mean that some gambling forms are more harmful than others: the speed of gambling and bet size.

Pokies allow for frequent, small bets, with spins every three seconds. Race and sports betting can involve much larger sums and betting that is relatively fast, but still slower than pokie spins.

Sports betting, in particular, is getting faster with in-play betting and microbetting.

Lotteries, on the other hand, are much slower-paced.

People typically spend a small amount on tickets and wait for a draw to find out if they’ve won.

Although it’s possible to spend a lot on tickets, people tend not to, unlike with faster gambling forms.

The average spend on pokies among the 16% who play them is around $4,782 per year, compared to an average spend on lotteries of $377 per year. These are averages. Most won’t spend these amounts but some will spend far more, which raises the average amount.

V’landys’ claim about lotteries being more harmful than race and sports betting was based on “independent statistics”.

He said that of 100 people seeking help from a gambling hotline, 70 had issues with pokies, 15 with lotteries, eight with race betting, four with sports betting, and three with casinos.

We were unable to verify these figures – if anyone has the data, we’d love to see the research to assess them.

However, we do have publicly available data.

What the data say

The NSW GambleAware website’s 2020-21 report shows that of 2,886 people seeking help, 73.3% identified pokies as their primary form of gambling, while only 13 people (less than 1%) listed lotteries. Race betting accounted for 13.1%, and sports betting for 7.9%.

These patterns were consistent with previous years.

People who experience problems also usually take part in more than one form of gambling, as the NSW report showed.

When these secondary gambling activities were considered, sports betting was cited by 35.5%, race betting by 33.5%, pokies by 19.5%, and lotteries by 13.7%.

What we discovered

The best evidence on gambling problems and harm comes from large-scale prevalence studies, typically commissioned by governments and conducted by independent researchers.

These studies offer high-quality insights into how each gambling form contributes to problems.

While one prevalence study is great, our team recently combined data from seven national and state-based prevalence studies. This resulted in a very high-quality dataset that we can use to study this question.

In our analysis, we used statistical techniques to show how strongly each gambling form is associated with problems.

These techniques give us regression coefficients, which are just numbers that tell us how strong the association is. A higher number means a stronger association between that form and gambling problems.

The most problematic form was pokies (coefficient = 0.147), followed by casino games (0.136), sports betting (0.068) and race betting (0.038).

Lotteries, with a coefficient of 0.001, were the least problematic and were not statistically significant even in our large sample.

As you might guess from such a low number, there’s very little relationship between lotteries and gambling problems.

What about prevalence?

Prevalence matters too – while pokies were most strongly associated with problems, the number of people participating in each gambling form is also important.

Let’s consider an analogy – a car that gives out a lot of exhaust fumes. That car is harmful, but if virtually no one owns one, then it’s not going to account for much pollution.

The same idea applies for gambling forms. If a gambling form is very harmful but very few people do it, it doesn’t account for many problems in the population.

It works the other way, too – if there is a very clean type of car that many people drive, they also won’t add up to much pollution.

Similarly, if we have gambling forms that have very little association with problems, it won’t add up to many problems in the population, even if lots of people take part.

The regression coefficients tell us how problematic each gambling form is. Prevalance tells us how many people do it.

When we combine these two bits of information, we can work out the degree of problems in the community that come from each form.

When we did this, pokies were responsible for 52-57% of gambling problems in the community.

Sports and race betting each contributed 9-11%, with a combined total of around 20%.

Lotteries accounted for just 0.1-1% of problems.

Even if we include scratchies as part of lotteries, this only adds another 2-5% of problems, still far below sports and race betting.

The real issue

What’s the takeaway?

Lotteries are widely played but are not typically associated with much harm.

Sports and race betting, despite having fewer participants, are more harmful due to their faster pace and the potential for large, frequent bets.

Lotteries involve slower betting and lower spending, making them much less risky.

If we aim to reduce gambling harm in our community, the focus should be on pokies, which are widespread in pubs and clubs outside WA, casino games and race and sports betting.

These forms have features that make them far more harmful than slower-paced gambling like lotteries.