Ten states or one county, which loan book actually survives a bad year

Two funds I've been reading take opposite positions on the same page of their material and both call it risk management.

The first has around 140 loans across eleven states and says the spread is the protection. If one metro corrects, it's 9 percent of the book, and the other ten states keep paying. Nothing in the tape is more than 2 percent.

The second has 38 loans inside about a two hour drive of the office. Their argument is that they've been lending in that market for fourteen years, they know which streets sell in 40 days and which ones sit for six months, they've foreclosed there before and know exactly what the process costs and how long it takes, and that a foreclosure timeline is a state-by-state fact you learn by doing it rather than reading about it. Spread across eleven states, they say, means eleven sets of rules and eleven markets you don't know.

I hold land for long stretches and I'm used to thinking one market at a time, so the second argument lands easier for me. But I can also see how one bad county turns a 38 loan book into a real problem while the same event barely shows up in the 140 loan book.

Curious where the room lands, because I don't think there's a clean answer here.