Good question. Two observations: * there are good reasons why they'd be shy about direct identification, * "The analysis stems from CVR data Clark County post

Good question. Two observations: * there are good reasons why they'd be shy about direct identification, * "The analysis stems from CVR data Clark County posted on its website. This data is publicly available for download. Links and archived links are provided here, and listed in detail in the final section of this document." which means that any data wonks (there are a few on HN) can download the same raw data and look to see if this pattern is present (as claimed) and unusual wrt to other election data dumps from prior years, other counties, other countries, etc.