The top of the list — where Trump is doing better than what we’d expect from the state’s partisanship — is generally populated by two types of states. First, there are blue states like Rhode Island (No. 1), Hawaii (No. 4) and Massachusetts (No. 5); it’s interesting that these strongly Democratic states haven’t gotten more anti-Trump. Second, there are Southern states like Mississippi (No. 2), Louisiana (No. 3) and Alabama (No. 6). The fact that Trump’s net approval rating so closely matches partisanship in these states may be because they are very inelastic, meaning they are home to few swing voters. Specifically, in these states, evangelical whites are likely to be staunchly Republican, and black voters are likely to be loyal Democrats. So it wouldn’t be surprising if all the Republicans simply approve of Trump and all the Democrats disapprove, with few independents left to move the needle.

Meanwhile, the bottom of the list — where Trump is underperforming partisanship — is filled mostly with non-Southern red states like North Dakota, Utah and Kansas. (New Hampshire is a notable exception; read on.) This may seem like a big problem for Trump, but it’s probably not. He simply has more room to fall in red states than in purple or blue ones. Indeed, if there’s anywhere that Trump can afford to lose fans, it’s in these states; they are so conservative that there’s no real danger of their going blue in 2020. (This also suggests that Trump’s low approval ratings nationally aren’t as bad a sign for him as they might appear. If Trump is disproportionately unpopular in safely Republican states but his popularity roughly matches partisanship in swing states, his low national ratings shouldn’t have much of an impact on the Electoral College.)