...

Earlier this year, I called attention to some analysis by Nathaniel Rakich at FiveThirtyEight. He compared Warren's 2018 vote totals to those garnered by Clinton in 2016. With the enormous caveat that it is always tricky comparing a midterm to a presidential election, it was interesting what Rakich said about the 10 towns where Warren most outperformed Clinton: "They are all small towns in Western Massachusetts. With a few exceptions, they have incomes lower than the statewide average. Most of them have fewer college graduates than average as well. And Trump improved upon Romney's margin in all but one of them."

Conversely, the towns where Warren most underperformed Clinton were the wealthy suburbs around Boston, and suburban, moderate women Republicans and independents were the key to the Democrats' retaking the House last year. I wonder, though, how many such Republicans and moderates voted for Trump in 2016 because they told themselves he would mature into the job. Either way, it is not inconceivable that Warren could hold onto some of the Democratic gains from the midterms, while overperforming Clinton in rural areas of the country. And she could slip on a banana peel tomorrow.

Elections are not predictable events. Trump certainly looks vulnerable today and if the economy slows down, he could be subjected to a thumping by almost any Democrat. Bill Clinton look vulnerable in the summer of 1995 and Obama did in 2011. Democratic primary voters should select whom they want to carry their banner and not get into the business of electoral prognostication. If they want Warren, they should vote for Warren, and if they do, they may have selected the first nominee in this century who can actually beat an incumbent president. (emphasis mine)