Reply #31: evidence?

[OnTheOtherHand]

WTF do you mean by a "reliability rate," and what is your evidence over "these last several years"? (Or was that a change in topic?)

Presumably you're talking about the table in the January 2005 evaluation report. Actually, this table was discussed even in the report itself: controlling for urban/rural, there is no statistically significant difference between hand count and other precincts, so there is nothing more to explain. Steve Freeman doesn't like it, but it's true.

I note that you don't even mention lever machines. It blew my mind when I saw that Andi Novick filed an amicus brief about why New York should keep its lever machines, 'supported' by statements about the exit polls -- when lever machines had the biggest discrepancies of all in 2004. Do you think that the lever machines were massively hacked in New York in 2004? I admit that I don't for a minute think that Kerry's high-teens lead in the polls surged to 31 points on Election Day, but massive lever hacking brought the margin back down to 17 points.