Dataset and Research Scope

Researchers pulled transaction records through Open Banking connections that flagged payments to operators outside the regulated framework, then matched those records against known licensed operator activity for the same accounts, which produced a clear view of how individuals moved funds across different gambling channels without relying on self-reported behavior.

The sample covered repeated transactions rather than one-off events, and analysts tracked activity both before and after the first detected payment to an illegal site so that any shifts in licensed operator usage could be measured directly from the data.

Key Overlap Between Licensed and Unlicensed Activity

Figures show that 91 percent of individuals who recorded repeated transactions with suspected illegal operators also maintained accounts and activity with licensed operators during the same period, and this overlap remained stable even after the initial illegal transaction occurred.

Licensed operator spend showed no substantial decline following the first engagement with an unlicensed site, which indicates that the two forms of gambling operated alongside each other for most users in the dataset rather than one replacing the other.

Problem Gambling Severity Index Scores

Among the 179 individuals, those with available Problem Gambling Severity Index scores displayed elevated risk levels, and 45 percent recorded a score of 8 or higher, the threshold that signals problem gambling, whereas only 6 percent of users who confined activity to licensed operators reached the same score range.

The higher concentration of elevated PGSI scores appeared consistently across the group that interacted with illegal operators, yet the data does not isolate whether the illegal activity itself contributed to the scores or whether pre-existing risk factors drove the choice to use both channels.

Transaction Patterns Observed

Transaction volumes and frequencies varied across the 3,426 entries, with some accounts showing clusters of small payments to illegal sites alongside steady licensed operator deposits, while others displayed larger single transfers that coincided with continued licensed play without interruption.

Timing analysis revealed that the majority of users continued licensed activity within days of the first illegal transaction, and the overall spend distribution across licensed channels remained comparable before and after the shift to unlicensed sites began.

Implications for Substitution Theory

Conventional assumptions that illegal gambling simply displaces regulated play receive limited support from these figures, because the data demonstrate parallel usage rather than a zero-sum transfer of activity from one sector to the other, and the persistence of licensed operator engagement points to supplementary rather than replacement behavior.

Observers note that the 91 percent overlap rate challenges models that treat illegal and licensed markets as direct competitors for teh same customer spend, and the absence of measurable substitution effects suggests that risk factors associated with illegal sites may compound rather than offset exposure to licensed environments.

Regulatory Context and Data Limitations

The UK Gambling Commission compiled the dataset through Open Banking APIs that captured direct bank transfers, which excludes cash or cryptocurrency routes that may operate outside this visibility, and the 179 individuals represent only those whose transactions triggered automated flags for suspected illegal operators.

While the study provides transaction-level granularity not available through survey methods, it does not capture demographic details or motivations, and researchers therefore present the results as descriptive patterns rather than causal conclusions about why individuals engage with both channels.

Conclusion

The October 2026 release supplies regulators and researchers with transaction evidence that 91 percent of repeat illegal gambling users also remain active with licensed operators, that licensed spend does not drop measurably after illegal engagement begins, and that problem gambling scores appear at markedly higher rates in the combined-use group. These documented patterns establish a baseline for further monitoring of cross-channel activity and for evaluating whether future interventions alter the observed overlap or risk distributions.