To get to the astronomical number of cases allegedly spread because of the Sturgis Motorcycle Rally, the researchers analyzed “anonymized cellphone data to track the smartphone pings from non-residents and movement of those before and after the event,” notes Newsweek. “The study then linked those who attended and traveled back to their home states, and compared changes in coronavirus trends after the rally’s conclusion.”

Essentially, the researchers assumed that new spikes in cases in areas where people went post-rally must have been caused by those rally attendees, despite there being no particular evidence that this was the case. The paper, which has not been peer-reviewed, failed to account for simultaneous happenings—like schools in South Dakota reopening, among other things—that could have contributed to coronavirus spread in some of the studied areas.

The researchers also assumed a $46,000 price tag for each person infected to calculate the $12.2 billion public health cost of the event—but this figure would only make sense if every person had a severe case requiring hospitalization.

The results of the IZA paper “do not align with what we know,” South Dakota epidemiologist Joshua Clayton said at a Tuesday news briefing.