I made a chart. On average, blue states have seen steadily declining coronavirus infections since mid-April. On average, red states have... not. 
People asked to see the same chart with NY, NJ, and CT broken out. If anything I'd say this makes the point more strongly: red and (non-NYC) blue states were on similar trajectories until about 3 weeks ago. Then blue states started falling while red states rose.
People are also asking for similar charts for hospitalizations and deaths. My source, covidtracking.com, doesn't have good enough data on hospitalizations to make a meaningful chart—a number of states aren't reporting data at all and others seem spotty.
Here is a chart of deaths, which does not show a divergence between red and blue states. However, note that deaths (and more importantly official reporting/classification of deaths) can lag covid diagnoses by several weeks.
If red states began to see a big spike in covid infections ~3 weeks ago, as my previous charts suggested, we wouldn't necessarily expect to see a corresponding increase in death rates by now.
A common response to my first chart is that the recent rise in red states might reflect more testing rather than more cases. Here's a screenshot showing states with highest (right) and lowest (right) positive rates. I've marked them by 2016 vote status. coronavirus.jhu.edu/testing/testin…
Three of the states with the biggest recent increase in cases—AL, AZ, and SC—are also the three states with the highest rate of positive test results. FL, TX, and NC are also in the top 10. It is very hard to believe their recent increase is an artifact of increased testing.
The 15 states with the highest summer temperatures (KY, MO, TN, NC, KS, AZ, SC, AL, GA, AR, MS, OK, FL, LA, TX) all went for Trump in 2016. In this chart I've broken those states out into their own category. This suggests the recent spike is linked to high temperatures.
