PLEASE SEE UPDATE #3.
I currently project Mitt Romney to receive no more than 1039 delegates total, 105 short of a majority needed for a first-ballot nomination.
This is the second in my series of diaries on the GOP delegate math. Number 1, in the wake of the CO, MN, and MO votes is here.
After the result today in Maine, along with the PPP poll (admittedly a small sample size), Mitt still doesn't have the delegates. He got 11 for his 200 vote victory in Maine. Lol.
A few more results like this and I will start writing about Santorum's math. (Euler number. Get it?)
The first diary was based on the premise that Mitt had enough money, organization, and support to pull out wins in "winner takes all" states that were otherwise blue or bluish and on the coasts. I'm not sure I can make this assumption anymore.
The first diary then assigned a pro-rata share of the remaining delegates to Mitt based on his current polling (except in IN and VA, where Santorum and Gingrich are not on the ballot).
Here's how it stands right now for Mitt (based on the delegate projection most favorable to Romney):
IA - 12, NH - 9, FL - 50, NV - 14, CO - 13, MN - 6, ME - 11, Superdelegates - 17, Hunstman endorsement - 2.
So far: 123. 134.
Constructing a scenario where Mitt outperforms his current polling in open primaries and blue states performs at that level in closed primaries and red states, here are the projections for the winner take all states:
AZ - 0
PR - 20
MD - 37 Winner-take-all by district.
DC - 16
WI - 39 Winner-take-all by district.
DE - 17
NJ - 50 Winner-take-all by district.
UT - 40* (219 total from winner take all states)
That brings him up to 342 353 meaning he needs 791 802 from the remaining proportional states and superdelegates to get a first ballot nomination.
I give Mitt IN and VA due to ballot failures. That adds IN - 27 + VA - 46 (assuming none for Paul which is not realistic, but we're constructing a really good outcome for Mitt here). That leaves Mitt with 426 415.
Giving him 33% (rounded up) of the delegates (and thus substantially outperforming his polling nationwide) in the proportional states yields these results:
MI - 10*
WA - 14
Super Tuesday:
AK - 8
GA - 26
ID - 11
MA - 13*
ND - 10
OH - 21
OK - 14
TN - 19
VT - 6
WY - 10
KS - 14
VI - 3
GM - 3
AL - 16 (Running total: 198 proportional delegates)
AS - 3
HI - 6
MS - 13
MO - 18
IL - 23
LA - 9
TX - 52
CT - 9
NY - 31
PA - 24*
RI - 6
IN - 9
NC - 19
WV - 10
NE - 11
OR - 9
AR - 11
KY - 14
CA - 57 (Running total: 532 proportional delegates)
MT - 9
NJ - 17
NM - 7
SD - 9
MR - 3
For a total of 577 from the remaining proportional states. The total is now 1003 992. We must deduct 81 from states I mistakenly believed were true winner-take-all. 922 delegates only.
Assign all 117 of the remaining Superdelegates to Mitt.* The total is now 1039 1120 1109, or 35 24 105 delegates short of a first ballot nomination.
*There are numerous assumptions in this diary that are unlikely, but they are designed to provide a margin of error. For example, he will unlikely get all of the remaining superdelegates. He is not polling at ~33% nationally at the moment. That is the point. If he does worse, naturally, he can't do it either.
My back of the envelope calculation is that he needs to get roughly 40% of the remaining delegates and all of the superdelegates to win the nomination on the first ballot. So, if the Superdelegates split evenly as well, he needs to get closer to 45%.
He is simply not performing this well at the moment.
Of course, a fortiori, Gingrich, Paul, and Santorum are even further away from a first ballot nomination.
4:33 PM PT: N.B. That the 11 delegate estimate from Maine is CNN's, not mine. With 39% of the vote, he barely beat... Ron Paul. Remind you of Iowa? Hmm.
7:44 PM PT: Georgia poll today:
Gingrich 35, Santorum 26, Romney 16, Paul 5.
8:03 PM PT: Using the latest polls, where available, here are the results up to and including Super Tuesday, roughly, for Mitt.
AZ - w/t/a 29 (+29)
MI - 10 (+29)
WA - n/a (+29)
AK - n/a (+29)
GA - 12 (+15)
ID - n/a (+15)
MA - n/a (+15)
ND - n/a (+15)
OH - 16 (+10)
OK - 13 (+6)
TN - 16 (+3)
VT - n/a (+3)
So, three more than I'm projecting just based on the scenario I'm suggesting. In reality, he will probably do better in MA due to organization and in ID due to the Mormon vote there, which is quite strong, but probably do worse in Alaska and North Dakota, but who knows.
8:17 PM PT: MD, WI, and NJ are also not "true" winner take all primaries.
Therefore, we should deduct another 81 delegates from Mitt. Ouch.
8:29 PM PT: RCP isn't counting Superdelegates the way other sites are and currently have Romney at 98 delegates. There are 2043 left on that count. Romney needs therefore to win about 52% of those remaining to get a first-ballot nomination. I have him at at slightly higher number, closer to 50%.
With less winner take all states available, he absolutely has to start getting majorities.
8:31 PM PT: There are only 122 winner-take-all delegates left. If Mitt gets all of those, and I'm counting the Superdelegates right, he needs 898 out of 1797 from the other states, or, just about exactly 50%.
Mon Feb 13, 2012 at 9:26 AM PT: Thanks to Brad De Long for the linky!
Mon Feb 13, 2012 at 9:41 AM PT: More chaos: whoever wins Arizona will effectively lose 29 votes due to the state having too early of a primary with winner-take-all results, just like Florida. If it's Mitt, that means he'll have lost 79 delegates to that rule, and that could very well be the difference in a first-ballot nomination.
Mon Feb 13, 2012 at 2:27 PM PT: And it appears Romney did not win Maine at all, with a Paul stronghold yet to vote in a rescheduled caucus.
