Doug Jones’s victory in Alabama was a historic one. Look no further than the fact that this was the first statewide victory for Democrats in nine years and that becomes apparent. As far as top offices go, we haven’t won a gubernatorial election since 1998 or a senatorial election since 1992. Those previous victories were based on ancestral ties to the now dead super-conservadem wing of the party, of which Dick Shelby, the 1992 winner, was a remnant. While the state thankfully no longer promotes George Wallace facsimiles to the national stage, Jones’s victory marks the first time Alabama has chosen to elect a candidate that could be considered socially moderate in a race where social issues took center stage.
I took the opportunity to see how Jones’s electoral returns fit into history. There are several ways to do this, but one of the more interesting is showing the last time a Democrat outperformed Jones (percentage-wise) in a race for Senator, Governor, or President. It turns out that nine counties are so morally depraved that Parker Griffith, whose double-turncoat status made him popular with just about nobody, managed to outperform Jones in his 2014 gubernatorial bid. However, all of these but Walker County, home to some Birmingham exurbs, have populations under 35k.
Indeed, there is a very high correlation between county populations and the historical significance of Jones’s performance. While Jones needed to outperform Hillary by roughly 15% in order to win, he managed to win while hitting that 15% mark in only 12 of the state’s 67 counties (constituting 47% of the state’s population). No Democrat has matched his percentage in a major race in any of the state’s eight most populous counties since Don Siegelman won the Governorship in 1998. Nine counties gave Jones better results than any Democrat since at least Shelby in 1992. Six of those are members of the ‘black belt,’ five tiny and one small (Dallas), which suggests that the black community was paying close attention to this race. The only one of the nine that didn’t vote heavily for Jones was affluent Shelby, which witnessed Jones’s second-best improvement over Hillary’s numbers from last year at 19%, turning a 72%-23% loss into a mere 56%-42% loss. That the state’s wealthiest county did this is a clear indication that the ‘educated suburban’ voter really flocked to Jones (although his largest outperformance of Hillary came in middle class Lee, where he turned her 36%-59% defeat into a 57%-41% victory).
Jones’s two most historically significant victories came in the counties of Alabama’s two largest cities, Birmingham (Jefferson) and Montgomery (Montgomery), where he did better than any Democrat in a major race since 1978. That year for governor, Fob James won all but two counties in the state, and carried Montgomery County 81%-18%, while Jones managed a 72%-27% victory. Democrats took an average of 52% there from 1980 though 2016. More eye-popping, however, is Jones’s 68%-30% victory in Jefferson, appreciably better than James’s 64%-34% margin. That same year, Howell Heflin won there 94%-6%, due to the fact that his opponent was representing the Prohibition Party. Meanwhile, Donald Stewart, running in the special election for the other Senate seat that day, fell to a virtual tie in Jefferson. It is possible that Jones’s performance there is the best ever by a Democrat in a competitive top-level race.
While I have no hope of further political victories in Alabama (legislative gains next year would of course be welcome), it is encouraging to see signs that the state isn’t totally dead. The state’s black community, which as a whole is one of the most conservative in the country, seems to be alarmed enough at the nearly overt racism of today’s GOP to take political action; Mobile and Madison (Hunstville) can still back a Democrat under the right circumstances (a first since 1998); and the trend of educated voters having a moral compass that can actually swing their vote continues into the Deep South.
P.S. I don’t mean to minimize the importance of downballot state races, but data isn’t as readily available for those races.