Lots of new followers since @politico. FYI: all my elections research is getting relaunched soon at my new Center @NiskanenCenter including an update to my 2020 prez forecast + my initial House & Senate ratings. Keep in mind, my modeling gives you conditions @ election time. That
means I don't do (or need to do) weekly or hourly updates like the other sites. And as I said in my 2018 July 1 Midterm forecast, my initial ratings always seem radical, but my expectation is, like 2018, they will reflect the mainstream forecast by the late fall. For Ex: my House
ratings when they 1st came out looked loony tunes, but pretty much everyone knew these districts were hot to trot for Ds by Sept/Oct . As I said back in '18, the senate was hard forecast bc it had to competing, prevailing winds running dead into each other. A massive fundamentals 
advantage for Ds on a map that almost could not be more hostile. It was clear that the demo evolution in NV & AZ were far enough along to flip, but TX? GA? Keep in mind, what made my July 1 2018 TX senate prediction radical was that I believed the state would be competitive. At
the time even very supportive @LarrySabato thought that was a bit outlandish. But on the other side, almost all of the map was either still early into a long-term demo realignment (TX, GA) or it was getting VERY far into a long-term realignment towards the GOP (TN, MO, ND, IN).
Now, I'm a political scientist, and a new one, and also a woman, so I am happy to admit I am learning every day. I am also proffering a new theory, and I would have been shocked if it was perfect. Or if it ever achieves perfection. Indeed, if you expect analysts to be perfect, to
never make mistakes, to never get a call wrong, you are going to be sorely disappointed bc such a feat is impossible. And if you are the type of person to dismiss an analyst's entire body of knowledge bc they make a mistake, even a really big 1 (say like missing Trump's win) then
I'm not sure your brain is well-suited to the world of analytics anyway. I knew there was NO WAY that TN senate was competitive for Ds and said so, unequivocally, on July 1st- when I advised them to shift those resources into TX instead. TN was PVI R+14. Of course, a freak
election w the right circumstances can override a party advantage like that, we just saw it in AL senate & KY Gov, but these have become exceedingly rare & in most cases, require a extenuating circumstance (child molester opponent, opponent that is taking away people's health
care). MD, MA, WVA: these 3 states are still continually electing statewide office holders from the minority statewide party- but they are exceptions, not rules. As for ND, MO, and IN, all 3 had shifted, on average about +5 more R points during the tenures of their D incumbents
due to the long term realignment that has lower edu, more rural states moving right, and higher edu, more developed states moving left. The 3 incumbent D senators COULD survive but it would take a huge Dem turnout surge AND a subsequent R turnout roll back. Dems had pulled back
their turnout in the 2010 & 2014 Midterms under the Obama Admin, & in VA in 2017 the surge WAS a bit uneven, mostly coming from Ds. So, I reasoned it was possible that Rs would roll back turnout. However, Trump's base mobilization strategy, a strategy that every single GOP
consultant would advise against, BTW, bc it goes against every conventional wisdom of the status quo about how to win elections in America, worked. I'll show you just how well in @newrepublic analysis. The GOP turnout decline did not come, and thus, IN & MO join ND in flipping &
unprecedented turnout surges in GA & TX fell short, but also represent two of the best waged campaigns of the cycle- taking two red states Ds had no business of almost winning, ESP IN A MIDTERM, & nearly flipping them. That leaves FL, FL, FL. The data I've looked at shows GOP
turnout in FL was, well, amazing. And thus far I haven't seen evidence that Gillum underperformed Nelson- at least not in exit poll data. I will be looking more closely at FL. Florida has a unique quality, one that makes it continually break in favor of the GOP: old people. In
Great Realignment, old people have been moving right. They are afraid of the changing world around them, and find the party's current platform attractive. And where 40% of voters under 40 show up to vote, 80% of voters of 60 do. But what really is wrong w FL is the D Party's piss
poor strategy, both messaging and GOTV (see other thread).
