How We Sampled California: Random and Snowball Sampling Across the State
A score is only as honest as the people behind it. We stratified our sample to mirror California itself, then used random and snowball sampling to reach players other surveys miss.
By Chloe MarshMay 21, 2026
From Survey to Score: Weighting the Eight CAC Pillars Out of 100
Survey data on one side, hands-on testing on the other. Here is exactly how the two become eight weighted pillars and one number in a green poker chip.
By Ryan CarriganJune 4, 2026
The CAC team weighs in
The analysts behind the study talk through what this chapter means for a California player.
The crypto-versus-other payout gap is real, not noise: t = 3.31, p = .006, with a large effect. Banking model genuinely separates the field.
It's the single sharpest discriminator in the whole dataset. If you only looked at one pillar, payout would tell you the most.
Region, by contrast, barely moved the needle — eta-squared of .0035. That's what justifies one statewide score instead of six regional ones.
Which is reassuring for a reader: a player in Sacramento and one in San Diego can trust the same number for the same casino.
And we publish confidence intervals, so you see the precision behind each figure rather than a number floating on its own.
Keep reading
Our full 67-page methodology and dataset: a stratified survey of 4,217 verified California players aged 21+, the eight-pillar weighting model, complete data tables and statistical analysis behind every score on this site.
Download the full study (PDF, 67pp) →