GA Early Voting Update 11/07
Quickly - let’s look at what’s left. I have two ways to do this, the first is just let’s throw in any ‘18 voter that hasn’t voted yet. That puts us at a final electorate of 4.27mm with 50.5% Democratic score or a +42k win (above 50%). This is a testament to just how great the banking has been for non-18 voters for the Dems so far (if you want to look at it in a slightly different way, 49% of early voters were likely Biden voters in ‘20, 46% were likely Trump voters, and 5% are new voters since then, and they lean Democratic. So consistent with the current modeling assuming Warnock and Walker are trading votes
The second is to rely exclusively on turnout scores to see who is left. Anyone left with a 70+ turnout score + those who have already voted is the most pessimistic, an election day turnout of only 1.48mm and total turnout of 3.98mm with a final 49.1% Dem score or -73k below 50%. Non Dem-voters in this score are inclusive of Rep+Libertarian for the most part, so that’s probably looking at a runoff.
As we dip lower to 60+ on the turnout score, we get 1.84mm election day voters with a final Dem score of 49.5% or a -46k loss. And if we go all the way down to 50+ we get a final election day of 2.26mm with a final combined Dem Score right at 49.9% or a -13k loss. Really a jump ball at that point.
Of course, scores are just scores. For people that turned out to early vote, if you were Democratic leaning (>50% on the party score), your average turnout score was 81.9%. And if you were Republican leaning, your average turnout score was 84.3%. What does that mean in plain English? Well, about 227k more Democratic leaners turned out than what was expected for their group, and about 195k more Republican leaners turned out. In other words, Democrats (at least in early vote) are outperforming the score more than Republicans.
Illustration of why that could make a big difference - the highest likelihood turnout voters who have yet to vote are packed with Republicans. But as you go to a bigger electorate, it gets much more Democratic. So just as an example, in the strongest decile for Republican support, there are 417k people waiting to vote that are >70% on turnout score. You drop down to >60% and that only goes up to 445k, you drop to >50% and that’s only at 464k. In other words, the difference between a >70% electorate on election day and a >50% is a whopping 53% more voters, but it’s only 11% more of the strongest Republican voters.
Now do the same exercise on the Democratic side - there are 367k very strong Dems that are very likely to turnout on election day (>70% turnout score). You drop that to >60% and the new numbers is 450k, you drop it all the way to >50% and the final number is 525k. Where the best voters for Republicans only increased by 11% when you increased the electorate by half, the best voters for Democrats would increase by 43%. Most of the lower turnout voters aren’t super partisans (makes sense), but in a higher turnout scenario the people in the middle who lean Dem would go up a whopping 104% vs only 72% for those who lean GOP.
Now one final way to look at it, a poll last week reported that 60% of Republicans planned to vote early vs 70% of independents and Democrats. Taking those numbers at face value, we’d see a final electorate of 3.8mm that is 49.8% Democratic on the score, a -16k margin. So look no matter which way you look at it, I think we are going to see a very close election.
I hope to have some good updates tomorrow about how turnout looks in various parts of the state.