Nate Cohn

Nate Cohn @Nate_Cohn

Over all that puts Trump underwater in states worth 412 electoral votes with RVs, compared to 310 electoral votes among 2018 voters — PolitiTweet.org

Nate Cohn @Nate_Cohn

Hadn't mapped out our estimates for Trump's approval among all registered voters, but they're quite similar to Civiqs' estimates today at least. Seems like UT (where we have Trump at +1) is the only disagreement on whether Trump is up or down https://t.co/2pC17Mh1dR — PolitiTweet.org

Nate Cohn @Nate_Cohn

@otis_reid i am not saying this is true, but it is at least possible that something like issues with hispanic voters / misestimation of turnout could hurt dems in, say, california without it indicating that this broader pattern would exist — PolitiTweet.org

Nate Cohn @Nate_Cohn

@otis_reid could just be intercept shift, but it is possibly worth noting that we don't really have pro-Dem bias (D+0.5) in the states that were biased towards clinton in 16. we just have more significant pro-GOP bias in the states where polls were biased towards trump — PolitiTweet.org

Nate Cohn @Nate_Cohn

@otis_reid if you prefer, the coefficient for 2016 bias is .309 in upshot polls and .652 in non-upshot polls, so a little less than half the effect here — PolitiTweet.org

Nate Cohn @Nate_Cohn

@otis_reid no, but easy enough. looks like there's some of the pattern here. smaller n per state, of course. https://t.co/kiuPi17QSC — PolitiTweet.org

Nate Cohn @Nate_Cohn

@bwreed @econjared @jimtankersley @jbarro it's possible — PolitiTweet.org

Nate Cohn @Nate_Cohn

@econjared @jimtankersley it's been my view, though i know people who know more about econ stuff than me disagree @jbarro — PolitiTweet.org

Nate Cohn @Nate_Cohn

@jimtankersley @econjared rate of change matters? there's a reason we care about election year growth not 4-year growth. if he could hypothetically get some kind of a deal--even a bad one--and end the tariffs, he'd presumably get a nice election year stimulus out of it, no? — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @NaturelsLit: Elephant caught throwing away litter into a trash can at a safari outpost https://t.co/2tR44EVG4w — PolitiTweet.org

Nate Cohn @Nate_Cohn

@aaron_strauss i guess i mislabeled the y-axis. it's trump net-approval in wisc. (say, .8 in '16) minus USA (say, -2.1) or positive 2.9 — PolitiTweet.org

Nate Cohn @Nate_Cohn

Maybe a clearer way to show it: how much better Trump's approval is in Wisc. than the US in Civiqs. As you can see, Trump's edge there has persistently been above the his relative edge from '16 there (the horizontal line) But it's narrowed a bit tighter over the last few weeks https://t.co/JrbBQYfh7E — PolitiTweet.org

Nate Cohn @Nate_Cohn

But over the last month, the WI-USA gap in Civiqs has fallen the lowest level yet. Will be interesting to see if it lasts, especially if/when they show a rebound in Trump approval(when Trump is strong nationally, he's really strong in WI in their data) https://t.co/XoxnjtulAE — PolitiTweet.org

Nate Cohn @Nate_Cohn

The Civiqs data is one of many things that didn't make the cut in my Electoral College piece from last month. But Civiqs, on average over the long term, has had Trump approval a net-5 points better in Wisc. than nationwide--worse than the estimate in our piece. — PolitiTweet.org

Nate Cohn @Nate_Cohn

Obviously that's sort of facially good news for Democrats, but I'd draw your attention to another change that offers some good news for Democrats, if confirmed, in Wisconsin. There, Trump's approval is at -9, which is about as bad as Civiqs has ever had it — PolitiTweet.org

Nate Cohn @Nate_Cohn

The Civiqs national approval tracker--which is fairly novel and has produced some reasonable results--seems like it's getting more attention in recent weeks, perhaps because they now have Trump underwater in states worth more than 400 electoral votes https://t.co/1EuA0odq2A — PolitiTweet.org

Nate Cohn @Nate_Cohn

proper link https://t.co/UsBwE5jSj2 — PolitiTweet.org

Nate Cohn @Nate_Cohn

A striking shift here from Pew on conservatives and presidential power file:///Users/205938/Downloads/PP_2019.08.07_Pres-Congress_FINAL.pdf https://t.co/wLyhcfSrge — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @mattmxhn: you can pretty much mark the exact 'shoreline' of the blue wave in houston suburbs https://t.co/9oDNe3J6zi — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @maggieNYT: Epstein dead, per veteran NYPD reporter Larry Celona >> https://t.co/e2b6OfEuzG — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @juliamacfarlane: Breaking: Jeffrey Epstein committed suicide overnight at MCC Manhattan, the federal lockup where he had been held pend… — PolitiTweet.org

Nate Cohn @Nate_Cohn

@ElectProject @rp_griffin @RachelBitecofer @gelliottmorris i might have misunderstood what you were asking for, but that's the educational attainment of times/siena poll respondents weighted to RVs, and then the subset of validated voters. the use of ACS data generally reduces the educational attainment of RVs in our estimates v CPS — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @PollsterPatrick: Obviously, the Iowa State Fair is letting in too many people with landlines. https://t.co/xEC7glzCVu — PolitiTweet.org

Nate Cohn @Nate_Cohn

this doesn't necessarily indicate that winning Iowa --> being top competitor, it is simply to specify the means by which it happens (you could choose Harris today, for instance, in part because you didn't think she could win Iowa) — PolitiTweet.org

Nate Cohn @Nate_Cohn

If Biden could choose to any candidate to win Iowa and emerge as his top competitor, who would it be — PolitiTweet.org

Nate Cohn @Nate_Cohn

@ElectProject @rp_griffin @RachelBitecofer @gelliottmorris (and i should note that the differential turnout rate holds even if you drop the education weights) — PolitiTweet.org

Nate Cohn @Nate_Cohn

@ElectProject @rp_griffin @RachelBitecofer @gelliottmorris so to me this begs the question of the CPS-based RV weights, but the very similar differential turnout among RVs by education lends some credibility either way — PolitiTweet.org

Nate Cohn @Nate_Cohn

@ElectProject @rp_griffin @RachelBitecofer @gelliottmorris well we're not representative of the US, and we're weighted to a CPS/ACS-based estimate for education of RVs. in any case, we're 37.5% with a degree among RVs; 40.8% among '18 voters — PolitiTweet.org

Nate Cohn @Nate_Cohn

@ElectProject @rp_griffin @RachelBitecofer @gelliottmorris yup i'm just using the CPS weight — PolitiTweet.org

Nate Cohn @Nate_Cohn

RT @amyewalter: As IA state fair kicks off this week @Nate_Cohn picks up on something I've also been thinking about a lot: what happens if… — PolitiTweet.org