I think it will be close.
The Kentucky governor's race was a very bad sign for Trump. There are a lot of parallels between Bevin and Trump. Both were elected as outsiders after defeating established politicians in the primary, and won upset victories in the general after consistently trailing polls. Bevin's personality and governing style are very similar to Trump's - impulsive, egotistical and needlessly confrontational, even with members of his own party. And those traits were the reason he was so unpopular and galvanized people into voting him out of office. On paper, Trump should be doomed when facing an electorate much less conservative than that of Kentucky.
However, what Trump has that Bevin really didn't is base that LOVES him, even if the rest of the country doesn't. No matter what, they will turn out to support him. This far out, Trump is trailing in almost every poll, but it could just be a repeat of 2016 where the Democrats lose thanks to a combination of complacency and death by a thousand cuts, considering that the four main candidates all have exploitable flaws of their own.
So does the Democrat manage to make it about Trump, or does Trump manage to make it about the Democrat? Both will try their hardest, and we'll see which one prevails. With the current polarization, it's anyone's game now.
You can't compare that race to 2020. Trump is far more popular than Bevan. Bevan had a 34% approval. Trump has consistently been between 42 and 47 for most of his presidency minus 2017. Trump's base as you pointed out is much larger and stronger. Trump wouldn't lose those rural counties or do as poorly as Bevan did. The suburban vote is a legitimate point but again this all comes down to who the democrats nominate. If Trump can successfully paint them as inept or radical he can do as well or better than he did in the burbs in 2016. I still believe this race is an uphill battle for the dems in the electoral college that is why I think it's lean Trump.
Two points here:
1. Since early 2018 Trump's average approval has generally bounced around between 40 and 43%.
2. You've made a great argument for why Trump will win the state of Kentucky. The country as a whole, however, is substantially less conservative than Kentucky, so you shouldn't be arguing that Trump would need an approval rating in the 30s to lose nationally.