Summary: Both parties are 3-4 points below their long term (2004-18) average when EXCLUDING leaners. When leaners are included Dems are 1.5 below and Reps 1.0 below long term average. D leans are up 1.8 and R leans up 2.8 vs longrun./1 
Trends in party identification popped up over the weekend as a topic for debate. I last did a deep dive on this in February. Since then trends from 2017 have flattened a bit in 2018. /2
The "hot" part of this is whether fewer people are calling themselves Republicans and if this putative decline explains why approval of Trump stays at 85% or so among the remaining Republicans. /3
These are worthwhile points. But the changes & their effects are often exaggerated in my view. GOP support was down in 2017, but only by 3-4 points, which isn't nothing but isn't huge and doesn't explain continued Trump approval /4
Also note how much variation there has been in party ID since 2004. Recent changes are modest compared to the amount we have seen over the past 14 years. And the trends change direction so not all one way. /5
In the 6 months of 2018 data from Gallup at this point, Rep identification has stabilized at 25.3% without leaners or at 41.2% with leaners. Those are 3.7 points below the long run average without leaners and 1.0 point below w/leaners. /6
Dems have been essentially flat in 2017 & 2018, at 29.1, 3.3 points below long run, without leaners and 46.4, 1.5 points below, with leaners. /7
What has increased recently is the percentage of independents. Pure independents who don't lean to either party are up 2.3 from long average, while independents including leaners are up 6.8 from their long average. /8
Footnotes: My February deep dive here has more discussion. The conclusions would differ only a bit, primarily because of the apparent flattening of trends in Jan-June 2018. /9
medium.com/@PollsAndVotes…
medium.com/@PollsAndVotes…
Footnotes 2: See links at bottom of my Feb post for analysis by others and arguments on this topic. /end
