The Wonders of Crowd Estimation
Long-time PJM readers will recall I've done a lot of crowd-size estimates. Now, the whole science of crowd estimation seems to be controversial, especially since in general the people who do crowd estimates for the legacy media seem to insist their methods are proprietary; that's why I've always made an effort to be very open about it.
Sometimes, though, it's easy. If you fill an arena with 5000 seats, you pretty well have to figure that's a crowd of about 5000 people.
Except if it's an Obama rally.
As noted by the blog battlegroundwatch, crowd sizes get weird in Wisconsin.
Here's the crowd in the 5000-seat BMO Theater, from the Battlegroundwatch article:
That looks like a pretty good crowd. People at the event said there were some empty seats, but as the BGW folks said, it's fair to say it was pretty full. Ergo, since it's a 5000-seat venue, about 5000 people.
Once it got to Politico, is was 18,000.