This is the premise of a lengthy report by Bill Galston and Clara Hendrickson recently published on the Brookings site. I recommend the report; it's well-written and has a great deal of useful data in it, summarized in a series of helpful tables. The tables generally compare a set of states they call the "northern tier" (IA, MI, OH, PA and WI) to another set they call the "southern tier" (AZ, FL, GA and TX). The are compared on 2018 results, including House, Senate and governor, as well as on time trend for these various offices. There is also a nice table on Obama-Trump counties in the northern tier and how many flipped D in the various races in 2018.
Galston's and Hendrickson's general argument is that these data--especially the 2018 data--suggest Democrats will likely have an easier time in 2020 expanding their electoral college coalition in the northern tier than the southern tier. That seems reasonable to me and their data do support that claim. I am less sure about the further implication they draw that Democrats need to decide on their geographic focus between the tiers and choose their candidate accordingly. This presupposes that the Democrats are going to go after one of these state clusters and not the other.
I don't believe this would be wise. Democrats need to put as many plausible states in play as possible to give them a variety of different paths to 270+. Putting all their eggs in one basket, such as the northern tier states, a strategy that Galston and Hendrickson appear to favor, would be a mistake on the Democrats' part.
Therefore the candidate that Democrats choose should be able to appeal to voters in both sets of states because that is how a Democratic candidate can maximize their chances of winning. And, it cannot be stressed enough, this is not just a matter of choosing the right candidate but of how that candidate chooses to run.
Words of wisdom from David Axelrod in a recent interview on the New Yorker site:
"I think that what is most important [for Democrats] is to not send the signals that were sent in 2016, which is, “We’ve got young people, we’ve got minorities, we’ve got women, so, you white working-class guys, we don’t really need you.” They believed it. They voted for Trump. And that is something that you can affect at the margins by addressing your message broadly, and I think Democrats should do that.
I think the country as a whole is restless on the issue of health care, whether it’s Medicare-for-all or some other prescription, as it were. I think people are eager for another round of health-care reform. I do think people think that there’s something wrong with our system right now, with this tremendous aggregation of wealth at the top while the majority of people are pedalling faster and faster to keep up. So I don’t think those issues are particularly radical. How you address them is another question."
And that is what we should really be worrying about.