Cue the Wagner music and get ready for the smell of napalm in the morning. Unless this weekend’s KKK/David Duke controversy dramatically shakes up the race, Donald Trump’s political Blitzkrieg is poised to break through the GOP’s front lines.
The polls are pretty clear that Trump will win most of the Super Tuesday states, and it will not be close in many of them. He has a huge lead in Massachusetts and a large one in Vermont, which we should expect given how demographically similar the GOP electorates are in those states to New Hampshire. Cruz leads comfortably in his home state of Texas, although not by a large enough margin to prevent either Trump or Rubio from garnering a lot of delegates. The other states, with the possible exceptions of Minnesota, Oklahoma, and Arkansas, look poised to go for Trump by 10-15 points in each.
Most troublesome for those desperate to stop him, Trump looks to be settling in at a higher level of support than in the early stages of the race. If the polls are accurate, Trump should win with at least 38 percent in Virginia, Georgia, Tennessee, and Alabama. Even should he lose the three states mentioned above as toss-ups, he will still nearly equal or slightly exceed the 32.5 percent he got in South Carolina. Narrowing the field seems not to have shrunken his lead, as many had expected.
I will use three measures to analyze Super Tuesday’s impact. First, who wins where (or finishes second) and by how much. Second, how many delegates does each person get? Third, which candidates still have viable paths to the nomination after the results.
I go through my state-by-state analysis below, but the bottom line is that Trump will sweep the night by any measure. I expect him to win at least seven states and perhaps as many as ten. He will win most of those states by ten points or more, and some by over twenty. Cruz should finish second in three states and third or worse in five. Rubio should finish second in every state except the three Cruz finishes second in, where he will finish third. My best guess is that Trump earns 270 delegates, Cruz 184, Rubio 126, Kasich 10, and Carson 2. Outside of Texas, however, Trump should lead with 215, followed by Rubio at 108 and Cruz at 102.
Most importantly, there will be only two candidates with viable paths to the nomination, Trump and Rubio. Kasich is only getting some votes from moderates and independents; he will to be able to steal somewhat conservative voters from Rubio or movement conservative voters from Cruz. Indeed he barely seems to be even campaigning for them.
Cruz’s lack of a path forward stems directly from his inability to persuade Republican voters outside of the base to back him. He will trail badly among moderates, who still cast 15-20% of the votes in GOP primaries. And he trails Rubio among somewhat conservatives not backing Trump.
Once the race leaves the South, these two groups will consist of larger shares of the vote than they will in most Super Tuesday states. That means Rubio will do better and Cruz will do worse in crucial states like Florida, Ohio, Michigan, Illinois, Arizona, and Missouri. Given that Ohio, Florida, and Arizona are statewide winner take all states, and that Illinois and Missouri have substantial winner-take-all components, that means Cruz will only be a spoiler who enables Trump’s eventual victory should he remain in after March 8.
Cruz could argue that Rubio should drop out because his and Kasich’s voters are likelier to hate Trump than his voters, so only he can hold enough of the base to put together a winning coalition. But it’s hard to see how either of those two men give up the chance to win based on speculation when each knows his home state is soon to vote.
Where this leaves the GOP, besides in crisis, is beyond my ability to predict today. But one thing I do know: Apocalypse Now ended by finding Kurtz.
For those interested in my predictions, please read the analysis below.
The sheer number of states voting means there are many more variables to consider than on any other day so far. So let’s walk through the states one by one (in order of delegates available) and see what is likely to happen and how that impacts the delegate count.
Texas (155 delegates) – The Big Kahuna is likely to be the only significant brake on the Trump war machine. Cruz is ahead in every poll, and comfortably in most of them. Last night Cruz predicted he would win with under 50%, and that almost certainly will happen.
In the four prior races, Cruz out polled his final average by about 2-4 points. Trump (except in IA) also out polled his average, by 4 points in NH and NV and less than a point in SC. Finally, Rubio out polls his average by the largest of any candidate. Except in NH, where his last minute debate gaffe cost him 4 points from the average, he has out polled his final average by 3.5, 5 and 6 points. So each of the leading three men should do better than their final polling averages in each state.
Cruz leads the RCP average 37-29, with Rubio on 18. My guess is that Cruz will get between 36 and 43 percent, Trump 28-33, and Rubio between 19 and 24. My precise guess is Cruz 41, Trump 29, Rubio 23, Carson 4, Kasich 3.
Texas gives the winner of each Congressional District (it has 36 CDs) two delegates and the runner up one. It does give a CD winner all three delegates if that person receives 50% or more, but Cruz is highly unlikely to reach that mark. I looked at all the 2012 races with CD level data that had three candidates or more. I found a candidate needed to get at least 41 percent before he could win even one CD with 50%, and usually he needed to break 44 percent to win a significant number of seats with a majority. Thus, I do not think Cruz will win any CDs with a majority, although it is possible that if he gets into the low 40s that could change.
I also do not think Cruz will win all 36 districts. We know that in any state CDs differ: even Newt lost two CDs to Mitt in 2012 when he won Georgia with over 47%. Trump did best in SC in the African-American-dominated 7th CD, which makes sense given that we know he does best with economically downscale whites who tend to live in close proximity to economically downscale minorities in the South. Texas has eleven majority-minority CDs, and to the best of my knowledge none of them has pockets of upscale whites in small cities tucked inside their boundaries. Thus, I think Trump will win these eleven districts, with Cruz finishing second in each. Cruz will win the other 25.
Rubio has been doing better in upscale, suburban areas, so it is possible to think he will beat out Trump for second in CDs with high numbers of upscale voters. I’ve identified seven such districts (2, 3, 7, 22, 26, 32, 35), and I predict he will finish second in each.
Finally, Texas awards 47 delegates (44 base plus the 3 RNC officials) proportionally to all candidates getting 20% of more. Rubio might not make the threshold, but I think he will barely squeak over.
Combining all of the above leads me to predict the following delegate breakdown: Cruz 82, Trump 55, Rubio 18. If Rubio does not make the threshold, he won’t get 11 delegates from the statewide allocation and he is not likely to finish second in more than one or two CDs. If everything broke Cruz’s way – say, he wins with 45 percent and Rubio does not make the threshold, it is possible he could win a few seats outright and get over 100 delegates. If the race is closer than I think, then expect him to lose a couple more CDs to Trump and perhaps get as few as 70 delegates.
Georgia (76 delegates) – Trump leads the Peach State by double digits in every poll taken since February 23, and there is no reason to think he won’t win going away. Rubio and Cruz are locked in a battle for second, with Rubio slightly ahead. My ranges are Trump 36-42, Rubio 23-27, Cruz 22-25, Carson 6-8, Kasich 5-8. Best guess: Trump 40, Rubio 25, Cruz 25, Carson 6, Kasich 5.
Georgia’s delegates allocation system is the same as Texas’, with one exception: the statewide winner gets the three RNC “superdelegates”. It has fourteen CDs.
Rubio’s strength closely tracks Romney’s so far, so I think he will win the two upscale Atlanta metro seats Mitt took from Newt in less favorable circumstances. Rubio will finish second in places where Mitt also finished second, which tend to be more middle-class suburban seats. I think Rubio will get ten delegates from Congressional Districts overall.
Cruz has been running best where there are high concentrations of very religious evangelicals, and those voters tend to be found in rural northern and southern Georgia. I don’t think Cruz will win any CDs, but he will finish second in the 2d, 3d, 8th, 9th, 12th, and 14th. That gives him six CD-level delegates. Trump will get the remaining 26.
Based on my best guess scenario, Trump will get 16 delegates from the statewide share, with Rubio and Cruz getting 9 apiece. That gives a statewide total of Trump 42, Rubio 19, and Cruz 15.
Tennessee (58 delegates) – The only poll we have for Tennessee has Trump ahead by a lot, with Cruz in second and Rubio in third. This is another state with large numbers of downscale whites and religious evangelicals and pockets of upscale whites in the major metros. In that sense it is a slightly more downscale and religious Georgia, so I expect it to turn out similarly with Trump and Cruz doing a bit better and Rubio doing a bit worse.
My ranges are Trump 38-43, Cruz 24-29, Rubio 21-26, Carson 6-8, Kasich 3-6. My best guess is Trump 42, Cruz 27, Rubio 22, Carson 7, Kasich 3.
Tennessee follows Texas’ system except that candidates can take home all the delegates statewide or in a CD only if he gets 66% or more. No one will pass that threshold. It has nine CDs.
It’s possible Rubio could win either the Nashville-dominated 5th or the inner-Memphis seat (CD 9). Romney carried both seats in 2012, but he did not lose the state by 20 percent. I think Rubio will have to settle for second in each seat to The Donald. Cruz could win in the rural, central 4th, 6th, 7th, or 8th, but he failed to win similar seats in SC when he trailed by a smaller margin. Expect Trump to win all nine CDs, getting eighteen delegates. Rubio finishes second in two and Cruz finishes second in seven (Rubio could nudge Cruz out for second in the Knoxville-dominated 2d).
I expect Trump to walk away with 31 delegates, Cruz with 17, and Rubio with 10. If Rubio does not break 20%, then add five delegates to Trump and three to Cruz.
Alabama (50 delegates) – The Alabama polls are the most surprising to me of any of the states’. This is a place with one of the highest concentrations of devout evangelicals in the country, a state that went for Mike Huckabee in 2008 and Rick Santorum in 2012. Despite that, the candidate who has so far been running best among very religious evangelicals – Ted Cruz – is running a poor third in almost every poll. And that was before Jeff Sessions endorsed Trump.
Not coincidentally, it is also the state where Ben Carson is running strongest. Clearly Carson’s support is eating into Cruz’s and depriving the Texas Senator of the votes he needs to potentially cross the 20% threshold needed to obtain a share of the 29 delegates allocated via the statewide vote.
Alabama also uses Texas’ system. It has seven CDs.
My ranges are Trump 36-44, Rubio 20-26, Cruz 15-20, Carson 8-10, Kasich 4-6. My best guess is Trump 43, Rubio 24, Cruz 19, Carson 9, Kasich 3.
Rubio could narrowly win Romney’s best district, the Mobile-based first. All others will go for Trump. Rubio will finish second in the 1st, 2d and 3d (Montgomery and Birmingham) and Cruz will finish second in the other four.
Delegate count: Trump 33, Rubio 13, Cruz 4. If Cruz breaks the threshold, subtract five from Trump and two from Rubio and add seven to Cruz.
Virginia (49 delegates) – My, how Trump will like winning a state with precincts within the Beltway. He won’t win those precincts – Rubio will win Fairfax, Arlington, and Alexandria, with Kasich doing his best here too – but he will ride support from the areas outside the Beltway to a comfortable win.
My ranges here are Trump 39-45, Rubio 25-32, Cruz 15-20, Kasich 5-7, Carson 4-6. My best guess is Trump 41, Rubio 30, Cruz 18, Carson 5, Kasich 5.
Virginia awards its delegates proportionally according to the statewide vote with no floor. That should give Trump 20, Rubio 15, Cruz 9, Carson 2, and Kasich 2.
Oklahoma (43 delegates) – The Sooner state might as well be the Snoozer State. It is one of two states (Mississippi is the other) that awards one delegate each to the top three finishers in each of its five CDs, provided each person gets 15%. Thus, winning Oklahoma gives the leader a tiny, inconsequential delegate lead. Barely worth your time to campaign here.
Twenty-eight delegates are allocated based on the statewide vote with a 15% floor.
As one might expect for a state that borders Texas, Oklahoma is one of Cruz’s stronger states in the polls. Nevertheless, he is behind Trump by a significant margin.
Based on Huckabee’s and Santorum’s strong showings here, I think Cruz will do much better than the polls suggest, but he will still fall short by single digits to Trump.
My ranges are Trump 30-37, Cruz 26-33, Rubio 22-28, Carson 4-6, Kasich 2-4. My best guess is Trump 36, Cruz 30, Rubio 25, Carson 5, Kasich 4.
Rubio will do best in the Oklahoma City CD (5), perhaps even winning it. But it won’t make a difference – he’ll get the same delegate he would have gotten with half the vote.
Delegate count: Trump 16, Cruz 15, Rubio 12.
Massachusetts (42 delegates) – Just in case you think The Donald is a Southern phenomenon, the Bay State is here to prove that wrong. Polls show Trump running away with the vote here, perhaps on track to get his first absolute majority.
Massachusetts awards its delegates proportionally according to the statewide vote with a 5 percent floor.
My ranges are: Trump 45-52, Rubio 15-21, Kasich 14-19, Cruz 10-12, Carson 2-3. My best guess is Trump 48, Rubio 20, Kasich 18, Cruz 11, Carson 2.
That translates into a delegate count of Trump 21, Rubio 8, Kasich 8, Cruz 5.
Arkansas (40 delegates) – Arkansas is another Southern state with large numbers of downscale and religious whites. The only poll we have is a month old and this should be discounted.
Arkansas awards 12 delegates via Congressional Districts (three each in four CDs) and 28 via the statewide vote. If a candidate gets 50% in a CD or statewide, he gets all the delegates in that jurisdiction. Otherwise, the winner in each CD gets two delegates and the second place finisher gets one, and the statewide delegates are allocated proportionally to everyone with at least 15% of the vote.
Without any real data, this state is hard to predict. I will guess that Trump ekes out a narrow in over Cruz with Rubio trailing badly. Ranges: Trump 33-39, Cruz 31-39, Rubio 15-20, Carson 6-9, Kasich 2-4. My best guess is Trump 37, Cruz 35, Rubio 18, Carson 6, Kasich 2.
I am awarding Cruz CD 4, which borders Texas. Trump wins the other three, with Cruz in second.
Delegate count: Trump 18, Cruz, 16, Rubio 6.
Minnesota (38 delegates) – Minnesota is a caucus state. It went for Romney in 2008 and Santorum in 2012. The only poll we have is from January, so even though it showed Rubio in the lead one ought not to put any faith in it.
Minnesota wins the award for the most convoluted delegate allocation system of the evening. It will award 24 delegates, three each, through caucus votes in each of its eight Congressional Districts. The three delegates will be allocated proportionally to anyone with 10 percent or more, but you start with the first person and round to the nearest whole number. In practice, this means that the top three finishers will get one delegate apiece unless someone wins 45% or more. I doubt anyone will do that, but if anyone will it will be Cruz in the rural 7th.
Fourteen delegates are allocated proportionally via the statewide total with a ten percent floor.
I can’t even venture a real guess as to the popular vote, although I will note that candidates with backing from pro-life groups do well here. I would guess this is a place where Cruz or Rubio could steal a win. Give Cruz, Rubio, and Trump 25-30% each, with Carson possibly breaking the ten percent barrier and earning one delegate.
Delegate count: Cruz 13, Trump 12, Rubio 12, Carson 1.
Alaska (28 delegates) – Alaska is a caucus state. It awards delegates proportionally based on the statewide vote with a 13% floor. There are no recent polls.
In prior years, Huckabee and Santorum got 22-30% of the vote. I would expect Cruz to get in that range. Romney won both years, which is a good sign for Rubio. But Trump does well everywhere, and with the oil price crash wreaking havoc on the Alaska economy I expect him to do very well indeed.
I’ll give Trump a narrow win here, with Rubio and Cruz getting about a quarter of the vote each.
Delegate count: Trump 11, Rubio 9, Cruz 8.
Vermont (16 delegates) – Vermont’s Republican Party is tiny, but not as moderate as Massachusetts. Santorum received 24% here, suggesting Cruz could do OK. But he is still not likely to break the 20% floor required to get any delegates.
The one poll from earlier this month has trump way ahead, Rubio in second, and Cruz and Kasich fighting for third. There’s no reason to think this isn’t how it will end up, which means only Trump and Rubio will break the threshold. Give Trump 2/3 of the delegates (11) and Rubio one-third (5).
Colorado and Wyoming each hold caucuses, but there is no biding presidential preference poll in either, so these events have no predictable influence on the race.