Lean R to Likely R.
Trump has more pull in GA than most people give him credit for.
Trump's numbers are worse in Georgia than any surrounding state.
Abrams is leading in polling and Kemp is a nutcase. Lean D.
Is she pulling in >50%? If not it's meaningless.
That's a literally meaningless benchmark in polling. Undecideds should not be expected to all go for Kemp.
It's not a matter of undecideds--Georgia is a 50%+1 rule state and Dems don't win runoffs in Georgia--or Abrams (or any Democrat) to win they'll have to avoid one.
The democratic coalition is higher turnout that it historically and this is a wave year. The idea that she would necessarily do much worse in a runoff is unsupported.
https://uselectionatlas.org/RESULTS/state.php?fips=13&year=2008&f=0&off=3&elect=6
why does everyone always bring up 2008? That was 10 years ago, and GA has changed in those 10 years. No one is looking at MO and saying "MO was close in 2008, we should be able to win it in 2020".
Because the same reasoning for why a 2018 runoff would be competitive applies to 2008. Wave year, Democrats more enthusiastic, etc. etc.
There's also the fact that Democrats have vastly underperformed in Georgia special elections.
But
1. The Democratic floor has grown
2. The D electorate has become more white
3. The voters who are more likely to vote in midterms, college educated people, are now more D than R
These are incredibly important changes that have occurred in GA. While many special elections that were held on off-days were weak for Democrats, Runoffs have been stronger, overall, for Democrats.
There is a large difference between an election that no one knows is occurring, and a runoff that will decide the governorship. One brings out enthusiastic voters, or regular voters, the other is a rather unknown quantity.