Analysis of play among British online gamblers on slots and other casino-style games


The Report gleans what can be learned from one month of data (January, 2017) provided by each of thirteen operators offering online slots and casino-style games to customers in Great Britain. The data, aggregated to the level of individual account statistics for the month, relate to slots play on more than 1¼ million accounts and non-slots games played by about 785,000 people. A weakness of the data is that it is not possible to identify whether the customers of different games and of different operators are in fact sometimes the same individuals. Each customer of each game/ each operator has to be treated as a separate individual.

Spending by players was measured by their net losses over the month. For most, their expenditure was somewhat modest and comparable with what might be spent on many other leisure activities. 73% of slots and 85% of non-slots customers spent less than £50 during the month. However, a small proportion of customers, but still up to 32,000 individuals, lost in excess of £1,000. Given typical income levels and spending commitments in Great Britain, it is plausible that many of these will have experienced harm.

Stake sizes showed patterns very similar to those which have been observed in the offline environment notwithstanding that there are no regulatory maxima online. The large majority of stakes are at less than £1 for a spin. For slots, the proportion of plays where the stake exceeded £5 (the limit in land-based casinos) was less than 2%. Higher stakes were more commonly placed on online games similar to those offered at land-based casino tables. The stake exceeded £50 in about 2% of plays. On the other hand, a stake of more than £100 was much rarer. It was not possible from the data set provided to link stake sizes with player losses.

Number of active days in the month was also recorded for each customer (who had all played at least once to be included at all). For either category of game, a clear majority of customers played on only 1-2 days in the month. Those who played daily or near-daily accounted for a low proportion of customers but still represented more than 27,000 slots players and more than 11,000 users of non-slots games.

A comparison between month losses and the number of days of play by customer was provided by each operator. For the very heaviest losers over the month (loss> £5,000), there was considerable heterogeneity in how the client had incurred the loss. More individuals in this very high losses group had played on 5-10 days in the month than had played daily or near-daily. Operators should be alert to the risk of potentially severe harm being incurred as a result of play over only a few days. In fact, very harmful play might frequently be intermittent or of short duration because losses are so high as to be unsustainable.

The data provide a basic but useful insight into the pattern of play on online slots and casino websites. The Report recommends that the Gambling Commission seek richer data from operators in the future. Less aggregated data might reveal how gamblers play rather than just the final outcomes of play. Trajectories of gambling would be particularly useful to model. And data by time of day would be valuable for comparison with studies of gambling at land venues, which demonstrate a tendency towards more potentially harmful play late at night and in the early hours.

Problem with this document? Please report it to us.