Numerical tidbits

The crowd that attacked the U.S. Capitol on January 6 didn't amount to very many people. 

The Washington Post farmed out video footage to Carnegie Mellon’s Informedia Project and came up with a count of 9,400 on the west side (Capitol steps and inauguration scaffolding) where the main breach occurred. Figuring in estimates of numbers on the other sides of the building, there may have been an additional 10,000 people wandering around (and some getting inside). 

To put this in some context, the crowd amounted to about half the average pre-pandemic game attendance watching a San Francisco Giants game at Oracle Stadium.

If the Justice Department eventually charges some 450 people for various offenses ranging from trespassing, though violence against the cops, to conspiracy to overthrow the election, that number will amount to about 2.3% of the crowd. A lot of insurrectionists are going to walk -- have they learned anything?

• • •

The Treasury Department today stated definitively that Paul Manafort passed 2016 Trump campaign polling data to a Russian intelligence agent. No more hints -- they just out and said it.

The Treasury Department stated it very matter-of-factly in designating Kilimnik, saying only that “during the 2016 U.S. presidential election campaign, Kilimnik provided the Russian Intelligence Services with sensitive information on polling and campaign strategy.”

It almost went unnoticed. But now, there’s no real way to deny where this points: Kilimnik giving polling data to Russian intelligence services nearly completes the link that was always missing from the scandal — connecting the Trump campaign to the Russian effort to damage Clinton.

The question remains: how useful is campaign data -- granular breakdowns of voter attitudes and impulses -- in targeting for maximum electoral effect?

I'm not sure we ever measure that. Every pollster and campaign consultant is motivated to tout their findings and the tactics they suggest that derive from them as the winning formula. And when an election is decided by a razor thin margin, they might be right and they might be wrong. I'm sure Russian spooks popped some corks!

And when elections are not close (most of them), it doesn't require any private polling data to figure out which voters and which tactics were needed.

But we're sure to go on polling and modeling and trying to find an edge. And sometimes it may help. But I distrust our capacity to identify which times.