> and it's especially good for republican voters.

This is true, but maybe not for the reasons you think. The number of people who self-identify as "Republican" has shrunk considerably over the past few years. https://www.brookings.edu/blog/fixgov/2018/06/14/trump-owns-...

Republican Party ID is up (at the expense of independents; Democratic ID is the same) from when Trump was elected.

https://news.gallup.com/poll/15370/party-affiliation.aspx

Huh! The Brooking's story actually cites this Gallup data. But it's a little out of date. Looks like Republican ID has sprung back up in the last ~6 months. Touché!

It's really noisy poll-to-poll; really, I'd be skeptical of anyone claiming that there is much of a meaningful trend based on it. Pew’s series which separately tracks partisans and leaners might be more useful for spotting trends, but they do take seem to have really recent data public in that series.

Trump's approval rating is really not doing OK. It's mediocre. Not Nixon or GWB-at-the-end-of-his-term bad, but an unbiased observer would not call being down 12 points "OK". https://projects.fivethirtyeight.com/trump-approval-ratings/

Yes, but I think it's perfectly normal given how politics works now. Everything is becoming more polarised.

> I think it's perfectly normal given how politics works now.

Yes, Trump's approval rating is dead average for President’s taking office after the end of Obama’s second term at the same point in their term of office.

But unless you reduce the domain so much that you compare Trump only to himself, it's not perfectly normal.