Georgia On My Mind

On Runoff Risks And Suburban Strategies

I don’t know if anyone has noticed, but I haven’t put out a House forecast in a while, for a simple reason – I know if I put out my forecast, based on the current Generic Ballot, it would get resoundingly mocked and made fun of as just another wishcast from a dumbass Democrat who can’t stop predicting Democrats to do well. The worst part is, it wouldn’t even be an unreasonable claim, given the history I have in predicting US elections, and I know this. It’s why I mostly save my flag plants for rare races now – say, Josh Shapiro or Brian Kemp, two candidates I fundamentally can’t see losing.

I avoid doing this with, say, Blake Masters – a candidate who has shown literally nothing to justify this conservatism from me, and who the GOP are probably triaging now that the Senate Leadership Fund is pulling money – because I know that I just can’t do that and have any credibility. I don’t say any of this to play some woe is me song – it would be played on the world’s tiniest violin if I did – but as a statement as the bleeding obvious.

I bring up all this history for a reason – I don’t have the credibility to plant my flag without a hell of a lot of evidence in 49 states. But 49 is not 50, and the one state that I seem to be able to call is, fortunately, the one that has a lot of people panicking right now. And so, listen up – Democrats are clear favourites to win Georgia’s Senate seat, and there’s little reason to think that will change.

…

In 2020, the source of all of my Georgia optimism came down to one fact – Stacey Abrams lost by 1.4% in 2018 while losing white voters by 48%, and Democrats routinely lost white voters by the high 30s or low 40s in polls. In the end, white turnout was higher than the polls were banking on, and while Biden only lost Georgia whites by 40%, the fact that 64% of the state was white meant it was a nailbiter.

What happened between 2020 and January 2021 was simple – Black turnout held up (and might have even risen, in raw numbers), and white turnout didn’t, so the electorate got substantially Blacker. The Black share of the electorate rose from 29% in November to 32% in January, and that meant that Jon Ossoff won while losing whites by 46% - or, more precisely, getting 27% of the vote.

Now, we’re seeing a pattern in the good polls this year – Warnock is exceeding, and in many cases quite so, the white win number he needs. So, why is it a two point race, on average? Pollsters either aren’t getting representative Black samples and have the GOP at 22% of Blacks or whatever (Emerson), aren’t pushing their Black undecideds (AJC/UGA), have samples that are less Black than 2020 (CBS/YouGov) – or have Warnock winning by 5%+ (Marist & Quinnipiac). Monmouth is sort of in the middle, with a 30% Black sample and a 4% Democratic lead on their weird not-really-a-vote-intention-but-actually-functionally-is-one shit.

What’s basically happening is when the polls that won’t happen get released – Emerson and their just constant, horrific inability to poll non-white people, or AJC/UGA with their oversampling of Republicans and lack of willingness to push Black undecideds – the media freaks the fuck out. When CBS/YouGov and Marist and Monmouth come out with reasonable electorate screens (and, honestly, including CBS/YouGov in any “reasonable” category feels wrong, which I’ll get into in a bit), then Warnock has a lead, and a big enough one to survive a risk of a runoff.

“Ah, but he’s only at 45% in Monmouth, and 47% in Marist, how can you say he’ll avoid a runoff?” I can hear you ask, and the answer is very simple – the relevant metric here is the size of the lead, not the vote share. This is a state with 1 third party candidate, which means that the runoff rule is essentially this – does [insert leading candidate here] have a lead bigger or smaller than the projected Libertarian vote? In 2020, Shane Hazel got 2.4% in November, which deprived David Perdue what would have, in any other state, been his victory. Warnock’s lead right now in 538 is just under that number, but that’s with shit like Emerson and fucking AJC clogging it up with bullshit.

For those trying to come up with polling trutherism, this ain’t the state to do it – the poll miss in 2020 was less than a point, using the 538 average, and RCP had a Trump lead, so depending on which source you want to use, there’s no real poll unskewing you can and should be doing here. Oh, and if anything, CBS/YouGov is almost assuredly wrong, but too long on Democrats. Their poll – somehow – only has 27% of the electorate being Black, a number that would require Georgia to get whiter in a midterm, which is not how any of this works.

The reason Ossoff and Warnock won despite underrunning Biden in the white parts of the Atlanta suburbs and exurbs is simple – the share of the electorate that is rural and white falls off when Georgia turnout falls. Be it runoffs or midterms, the share of the vote that comes from what you might call “Greater Atlanta” – including places whose historical identities are very much their anti-Atlanta natures – falls off. In the 2008 runoffs, the statewide turnout was 57% of the previous, November turnout. In Forsyth (59.9%), Cobb (60.5%), and Gwinnett (58.5%) – three left-trending suburban or exurban counties – the share of votes cast was above state average. The counties that border Florida? They had turnout that year of 50.4%.

This is why Democrats won in 2021 – white turnout was not strong enough, even with some suburban reversion, to make it so that Republicans could overcome the tidal wave of Black turnout. In 2022, with both Warnock and Stacey Abrams at the top of the ticket, I struggle to see Black turnout being the problem, so what’s the way Democrats lose? Well, suburban reversion, I guess – but let’s be frank, Herschel Walker is the wrong candidate to induce that.

The Atlanta suburbs and exurbs are a story in two parts – a story of diversification on one hand, but more importantly a story of familiarity on the other. What led to Forsyth being whites only until the 90s was obviously racism, but the source of the racism wasn’t just that those residents at the time were irredeemably bad people, it was a lack of familiarity. We know that animus towards protected classes declines when people are exposed to them – just as people with gay family members are less likely to hold homophobic views, people who work with racial minorities are less likely to default to stereotypes about those groups. What happened when Black wealth started to truly be created, and Black people moved into the nice neighbourhoods north of Atlanta proper, those areas saw declines in racial animus as some of those residents got exposed to Black people more regularly and their worlds didn’t entirely collapse.

Now yes, these voters were still reliable Republicans through the 00s and even in 2012, when Romney swept Cobb and Gwinnett and won Forsyth by nearly 60%, but what’s happened since then is Republicans have moved further and further away from social liberalism and Democrats have made their social liberalism a much more prominent part of the message, which has meant that so many former Republicans haven’t been able to pull the trigger to vote for Republicans anymore. Dobbs will help accelerate the trend of pro-choice whites moving left, but it’s not just Dobbs or the American context that explains this.

The reasons Forsyth is moving left are the same that explains why places like outer London, wealthy pockets of Sydney, or the rich western parts of Greater Toronto are moving left too – it’s social liberalism being a greater priority for voters than economics. The voters we’re talking about here are not economic radicals – hell, they’re not even economical liberals, but the tendency of the global right to move towards right wing populism, and away from Romney-style tax cuts while avoiding talking about social issues means that these voters aren’t moving back to the GOP.

Throw in Herschel Walker’s undiagnosed CTE, multiple secret children, and the credible allegations of abuse he faces, and you have a recipe of a candidate that the Atlanta suburbs are going to give a rhetorical middle finger too, because Walker is the exact opposite of the candidate for the Atlanta suburbs. If the GOP were smart, they would have run someone who could appeal to the Atlanta metro’s band of socially liberal professionals – the people who play golf with the gay couple two streets over but really don’t want changes to how capital gains on stock profits are taxed – they’d win the state. But, they’re running Herschel.

Now, could Walker get enough rural turnout? Maybe, but it’s not likely in a midterm. Given the changes in educated whites since 2018, a 2018-style turnout of rural white voters wouldn’t be enough to win the state. I’d say he needs a 2020-style or better electorate, but even in the one poll that actually did that, he’s still down 2, so he really needs a steep decline in Black turnout – or to actually win 15% of the Black vote, which won’t happen. Why am I so willing to dismiss this possibility, despite the Georgia GOP running a Black candidate? Because Black voters aren’t fucking stupid.

The “Republicans run a Black guy and he’ll win with a totally different coalition than we always win with” stuff is almost always shit, and the one case it isn’t is South Carolina, where Democrats have not credibly contested either of Tim Scott’s runs and if/when they ever do so, his willing coalition won’t look much different than the generic Republican’s. A look at John James’ coalitions in Michigan don’t suggest any substantial strength with Black voters above Generic Republican – Trump got 30.3% in Wayne County in 2020, and James got 30.2% - so assuming that there will be minimal crossover of Warnock/Biden/Ossoff 2020-Walker 2022 Black voters is a perfectly reasonable assumption.

So what’s the intrigue left in the seat? Whether or not Warnock can hold a lead above the Libertarian vote share, essentially – does the Libertarian break 2%, and then can Warnock win by more than that? Plainly, I think the chances of Walker winning and hitting 50% on November 8th are functionally non-existent, and I think that Walker starts any runoff – regardless of the November results – as a substantial underdog. So it’s really just about timing.

I don’t have the intellectual arrogance, or the balls, plainly, to be too cocky about, well, anything much in the US these days, but I do have a good sense of Georgia. I do not think there is much chance of a Herschel Walker victory, either in November or in a runoff. I just, plainly, don’t. If you want to panic about Senate seats Democrats need, Nevada is the one. But frankly, I just don’t find anything worth panicking about in Georgia.