The White Noncollege Difference

The Economist has published some very interesting numbers on Biden's performance relative to Clinton among various demographic groups (presumably, the number-crunching was done by G. Elliott Morris, who is their US politics data guy), The standout difference here is among white noncollege voters. As the chart below shows, the more white nonocollege the state is, the more Biden's performance is superior Clinton's in 2016.

That is a very beneficial pattern for Biden in terms of electoral college results. The article notes:

"Currently our model estimates that 41% of whites who cast ballots would vote for Mr Biden if the election were held today, whereas 51% say they will cast their lot for Mr Trump—a ten-percentage-point margin. In 2016 Mrs Clinton lost this group by 15 points. Mr Biden has improved his standing both among whites who have college degrees and the ever-watched group of those who do not. He polls four and six percentage points better than Mrs Clinton did among each group, respectively. Mr Biden is currently polling 11 points better than Mrs Clinton in states where working-class white voters make up the largest share of the electorate, and he is performing roughly six points worse in those states where they are the lowest share (see chart).

That improvement has a disproportionate effect on Mr Biden’s chances of victory. Whereas Mrs Clinton lost the election by small margins in Wisconsin, Michigan and Pennsylvania, we find Mr Biden with a slight lead in all three. He is also likely to beat Mrs Clinton’s margin in Arizona, which is likelier than any of the midwestern battlegrounds to tip the election."

The article also notes some underperformance by Biden among nonwhites and young voters relative to Clinton but this pattern appears to be less salient to electoral college results in their analysis. In the Nationscape data (80,000 cases since the beginning of the year), I find less of underperformance among these groups, but confirm the general pattern of Biden overperformance among noncollege whites with similar positive effects on Biden's electoral college results.

If this pattern continues, Biden is in very good shape.

https://www.economist.com/…/we-spent-the-lockdown-sorting-a…