I learned something today. That's always a good thing.
The difference between the black vote counts and turnout numbers cited by Matt Smyth in the Augusta Free Press and by Larry Sabato in the papers cited in my earlier post today is that Smyth is, according to Sabato, citing results from "the traditional sample black precincts that date to Ralph Eisenberg in 1960s which show a lower percentage for GOP candidates; they are highly concentrated and more organizable for Democrats." The higher numbers cited by Sabato come from " exit polls, that supposedly include blacks from suburbs (higher income, slightly more Republican)."
As Larry says, " Both have flaws. Take your pick!"
And, just to confuse things... here's an exit poll from 1997 that puts the black vote for Gilmore at 14% (a number higher than Smyth's 11% and lower than Sabato's 18%).
What's that old saying about lies and statistics?