Seems to me that it's fluid depending on what's going on - and since the shutdown was blamed on Trump/repubs, January identification will reflect that. If the poll Bou posted was taken 5 months ago in Aug, would the result not be different then when more people identified as Republican than now when more people identify as Democrat (and yes, the Dem advantage is much greater)? January 2018 has repubs at a very low 22 and I can't remember what was going on then that was worse than the shutdown. I'm sure it'll swing, and I guess what's important is what the breakdown is come early November 2020.
R I D
2019 Jan 2-10 25 39 34
2018 Dec 3-12 26 39 32
2018 Nov 1-11 28 39 31
2018 Oct 15-28 29 36 35
2018 Oct 1-10 28 39 30
2018 Sep 4-12 26 44 27
2018 Aug 1-12 28 43 27
2018 Jul 1-11 26 41 30
2018 Jun 1-13 27 43 29
2018 May 1-10 26 43 29
2018 Apr 2-11 24 45 29
2018 Mar 1-8 23 45 29
2018 Feb 1-10 28 42 27
2018 Jan 2-7 22 44 32
2017 Dec 4-11 25 46 27
2017 Nov 2-8 25 42 30
2017 Oct 5-11 24 42 31
2017 Sep 6-10 29 40 30
2017 Aug 2-6 28 41 28
2017 Jul 5-9 25 45 28
2017 Jun 7-11 26 42 30
2017 May 3-7 29 40 28
2017 Apr 5-9 25 44 28
2017 Mar 1-5 26 42 30
2017 Feb 1-5 31 37 31
2017 Jan 4-8 28 44 25