Rubio’s NH momentum had stalled even before the debate. If anything, the establishment seems to be coalescing around Kasich, although not enough to give him a convincing showing.
So I’m willing to stand by my projection for the D’s but after that debacle of a debate last night I would modify the GOP side thusly:
Trump 28%
Kasich 22%
Christie 12%
Bush 10%
Cruz 9%
Uh, yeah. Not going to even consider that one without a credible cite.
So you’re thinking almost all of Rubio’s votes will go to others? Wow, even I don’t believe he will drop so precipitously.
ETA: My guess is Rubio stays somewhere in third place.
I too have done this without consulting any polls (I would have done except that the debate scrambled things a lot, I suspect):
Trump 30.0
Kasich 18.4
Bush 10.5
Rubio 9.8
Cruz 8.3
Christie 6.6
Fiorina 4.1
Carson 2.7
Pataki 2.3
Paul 1.1
Other 6.2
TRUMP-35%
KKKa$ich-17%
Jeb!-14%
Rubio-12%
Cruz-11%
Christie-9%
Carson-1%
Fiorina-1%
Dems:
Sanders-55%
Clinton-45%
Slight revision to above:
Trump 35%
Kasich 15%
Cruz 12%
Christie 12%
Rubio 10%
Rubio 10%
Rubio 10%
Rubio 10%
Bush 6%
Carson 5%
Ha!
Alright, I’m giving myself a B this time. Predicting the Rubio non-bounce - before his disastrous debate - was a stroke of genius. I picked the top three correct, just not in the right order. I really thought Cruz would do better, in terms of picking up the far-right vote that is there in NH and turning them out.
Had Rubio gotten 6/10ths of a point of Cruz’s support, I would have gotten the top eight in the exact right order. Giving myself an A-.
It might be a slow morning for me, but it looks like if you give Rubio 0.6 points from Cruz, you’d have Trump, Kasich, Rubio, Bush, Cruz, which is not the correct order.
I see the numbers are slightly different from when I posted that. Here is what looks to be the final tally:
Trump 35.3
Kasich 15.8
Cruz 11.7
Bush 11.0
Rubio 10.5
Christie 7.4
Fiorina 4.1
Carson 2.3
So, yeah, now if I take 0.6 from Cruz and give it to Rubio, it becomes (slightly different from what you said):
Trump 35.3
Kasich 15.8
Rubio 11.1
Cruz 11.1
Bush 11.0
Christie 7.4
Fiorina 4.1
Carson 2.3
So I would have to change it based on these final numbers to “If I took 0.8 from Cruz and gave 0.6 of it to Rubio and 0.2 to Bush, I would have the top eight in the right order.” More complicated, not as impressive, alas; but still indicative of being very close.
Also, here were my numbers compared to the actual (mine/actual):
Trump 30.0/35.3
Kasich 18.4/15.8
Cruz 8.3/11.7
Bush 10.5/11.0
Rubio 9.8/10.5
Christie 6.6/7.4
Fiorina 4.1/4.1
Carson 2.7/2.3
I called Kasich’s second place finish, and that it would be by a comfortable margin over Bush/Rubio/Cruz, all three of whom would be tightly bunched. My biggest miss was Cruz, but for the bottom five (Bush, Rubio, Christie, Fiorina, and Carson) I averaged less than a half percentage point off of their actual numbers. I’m sticking to my A-, dammit! 
WTF, Bush is still relevant? Along with Kasich being 2nd and Rubio dropping to 5th, this has been a most unpredictable NH
Well, Trump did much better than I thought with 35.4% but at least I had Kasich in second though he didn’t do as well as I expected.
The only ones I got close on were Cruz and Bush. Even Rubio didn’t fall as far as I [del]hoped[/del] thought he would.
Yeah, guess the exact numbers are still in flux a bit…