To determine whether any party’s candidates have an advantage, data from the previous five years of contests with a statewide elective office on them can be used. This yields ten data points. For each election, the proportion of early voting compared to total registration and final turnout percentage may be computed to make a ratio of total turnout percent to early voting percent. An average of these total/early proportions can create a benchmark to forecast total turnout.
Democrats have averaged 39.26 percent total turnout while Republicans have averaged 43.59 percent. In terms of early voting over this span, those means respectively are 8.47 and 10.14. Thus, the ratio for Democrats, is 4.65; for Republicans, it’s 4.35. This shows in recent history that of those who vote Democrats in comparison to Republicans disproportionately don’t vote early, with early votes making up 21.5 percent of their total while for the GOP its early voters comprise 23 percent of that total.
At the same time, the early voting average higher Republican turnout of 1.67 percent is 2.6 times smaller than the average gap in total turnout that favors Republicans by 4.33 percent. With early 10/12/19 voting encompassing 13.16 percent of Democrats and 16.91 percent of Republicans, the gap more than doubled to 3.75.
Using the historical ratios, this means trouble for Democrats. That would imply a 64.68 percent turnout for Democrats and 73.56 for Republicans. Such lofty numbers won’t happen because of the trend to substitute early for election day voting, for which these ratios don’t compensate. However, comparatively these do point to a significant GOP advantage.
Of the three similar contests with similar recent past turnouts – the 2014 Senate runoff and the 2015 statewide contests which ranged from just under 42 to over 45 percent – the Democrat/Republican total turnout ratio was 0.934:1, while the extrapolated 2019 general election ratio is 0.879:1. In other words, Republicans typically enjoyed a 3.14 percent higher turnout in that range of total turnout but here they are on course for 3.34 percent higher in that range.
Keep in mind as well that the number of Republicans registered has surged in the past four years by some 103,000 while Democrats have dropped 78,000, so the 0.2 percent bonus becomes even more valuable along with the 3.14 base. And in the context of the last such elections, the difference becomes starker still.
Looking back at the 2015 elections the ratio of Democrats/Republicans voting early averaged 1.46:1 (the two highest of the ten); in 2019 so far, it is 1.06:1 (the lowest by far). In 2015, the actual vote average ratio was 1.57:1 or a mean roughly of 220,000 more Democrats voting than Republicans; the extrapolated number for 2019 assuming the same dynamics (including the same turnout range and that early voter proportions by party remain comparatively the same, meaning no differential change in the substitution effect) of 1.16:1 makes that gap fall to around 62,000.
That difference of 158,000 would have made GOP former Sen. David Vitter governor of Louisiana in the 2015 runoff by 15,000 votes. And Edwards doesn’t have as much of a margin for error this time. With these dynamics, unless Democrats change them by boosting election day turnout significantly compared to Republicans, this time he does lose.