Many things about this campaign have been unfair, but are no less real for being unfair. One is that the primary calendar is not evenly distributed. As I have argued for months now, the front-loading of states favorable to Ted Cruz in the first week of March is both an opportunity and a challenge for Cruz. Now, the hour of peril is here. If ever there was a time for Cruz’s projections of millions of non-voting true conservatives to materialize, this is it. But if Cruz doesn’t have a very good showing on March 1 and 5 (more below on what that might look like), his plausible path to the nomination will go off to John Kasich-land, and he will be faced with the choice of staying on as a factional/vanity candidate or getting out for the good of the conservative movement. It may not be fair that Cruz faces this test earlier than Marco Rubio does, but it has been built into Cruz’s primary strategy and the nature of the calendar from the very beginning.

I have said for some time now – ever since my first choice in this race, Bobby Jindal, got out – that while I prefer Rubio to Cruz and think he would be a better general election candidate, I would without hesitation support Cruz if Rubio’s campaign no longer had a path forward, and I’d happily go to war behind Cruz in the fall. I had always assumed that Cruz would be in this race to the very end, and he’ll have money and supporters enough to do so if he chooses. With other time-wasting candidates (Kasich and Carson) still onstage, I’m in no hurry to get rid of one of our two remaining good choices. But by the morning of March 6 at the latest, if the Cruz campaign isn’t clearly in the driver’s seat, it will be time to throw in the towel if there’s to be any chance of stopping Trump and saving the conservative movement and the Republican Party.

Cruz, Batting Right

Cruz and Rubio are all but tied in the popular vote and delegate count so far, with Cruz a bit ahead in both, and so superficially it might seem that Cruz is the best bet to be the last man standing against Donald Trump. But that ignores some basic facts about the primary electorate and the structure of the race.

The immediate reason why Cruz faces a harder road than Rubio is that the voting so far has turned out to be more or less consistent with the things that Cruz’s skeptics have warned for months: That his appeal to the most faithful conservative base voters would not translate well with the rest of the party, and would not be enough alone to build a winning coalition. If you look at the exit polls, you see a picture of a candidate who is still reaching only one faction of the party – a wider faction than pure social conservatives like Rick Santorum or Mike Huckabee, but a faction nonetheless.

Let’s start with Non-Evangelical-Christian voters:

IA: Trump 29, Rubio 26, Cruz 19, Kasich 4
NH: Trump 37, Kasich 18, Rubio 11, Cruz 8
SC: Trump 30, Rubio 22, Kasich 19, Cruz 13

Demographically, the backbone of the Cruz movement is self-identified Evangelical or Born-Again Christians. That is indeed a vital constituency in the Republican Party, and they made up 62% of the vote in Iowa and a staggering 72% in South Carolina – two states that should have been perfect testing grounds for Cruz’s theory of the electorate. But Cruz has simply not been competitive outside that group of voters – and that’s a big problem after March 5, as fewer and fewer states look like Iowa or South Carolina.

Geoffrey Skelley looked at this back in November, and you can see more data in this 2012 Pew report. Skelly estimated at the time that “about two-thirds (64%) of the total delegates in states with contests on or before March 8 will come from states with electorates that may be at least 50% white evangelical.” Using the actual numbers for the first three states and Skelley’s figures (drawn from 2008 and 2012) for the others, we get a calendar that is not all friendly turf for Cruz, but generally pretty good through March 5:

You will note that Evangelical turnout exceeded Skelley’s figures for each of the first three states, by 4-8 points, undoubtedly due in part to Cruz and his impressive turnout machine. So we can expect those numbers to rise from where they are projected. Also Skelley doesn’t count non-white Evangelicals, who have been a non-factor so far as the primary turnout has been at least 96% white in each of the first three states. But turn the page to the post-March 5 calendar, and the picture gets a lot worse for Cruz:

Only two, maybe three, states remain after March 5 where Cruz could hope for Evangelical Christian turnout on the order of Iowa, let alone South Carolina, and there is a very long stretch from March 6 through the end of April in which Evangelicals are likely to be a minority in nearly every state’s primary electorate. Rubio has shown he is relatively competitive with non-Evangelical voters, and might have some hope not only of picking up the prior supporters of Jeb and Christie but possibly Kasich as well, if he can get Kasich to drop out. Cruz has yet to show he can do that. Which means that if he wants to make his move, now is the time.

Yet while Cruz has bet everything on Evangelicals, he has yet to consistently beat Trump with them, either:

IA: Cruz 33, Trump/Rubio 21, Kasich 1
NH: Trump 27, Cruz 23, Rubio 13, Kasich 11
SC: Trump 33, Cruz 27, Rubio 22, Kasich 5

The trendline has not been positive, in part because Rubio bounced back in South Carolina. If Cruz is going to roll up a big delegate score on March 1 and 5, he will need to turn that around, and fast.

Let’s look at a second split, this one ideological, to explain further why Cruz’s campaign has failed to expand beyond his niche and why Rubio is more likely to inherit the “establishment lane” supporters whose candidates have dropped out. There is some overlap here with the religious lanes, but it is also telling:

Somewhat Conservative voters:

IA: Rubio 29, Trump 24, Cruz 19, Kasich 2
NH: Trump 36, Kasich 14, Rubio 12, Cruz 9
SC: Trump 35, Rubio 25, Cruz 17, Kasich 6

Moderate voters:

IA: Trump 34, Rubio 28, Cruz 9, Kasich 7
NH: Trump 32, Kasich 28, Rubio 8, Cruz 4
SC: Trump 34, Rubio 23, Kasich 21, Cruz 7

The “Somewhat Conservative” voters have been 43-45% of the vote each of the first three states, whereas the Moderate bloc was much larger relative to the Very Conservative voters (27% each) in New Hampshire than in Iowa or South Carolina (40% and 38% Very Conservative vs 14% and 17% Moderate). That will vary as well as we go along, and if you’re familiar with the map, you know that the Very Conservative voters will get scarcer as we get outside the South (although more of them who are not Evangelicals can be found in the Mountain West states). Either way, these voters aren’t interested in Ted Cruz.

Exit The Voters

I decided to do a little more with the exit poll numbers and run an unweighted average of the first three states, then divide the Trump/Cruz/Rubio/Kasich exit poll figures into an out-of-100 percentage, so I could pinpoint a little better which voting blocs were each candidate’s strongest and weakest so far (unfortunately, this requires taking Kasich seriously as if he were a real candidate, but as long as he stays in, there are certain types of voters that he alone appeals to). First, by voter demographics – the strong ones are in red, weak in green:

Some of this is unsurprising. We covered religion already, and that’s just one of several areas where Kasich is the mirror image of Cruz, strongest with the types of voters least interested in Cruz and vice versa, while Rubio is the mirror image of Trump. Trump, of course, does best with the least educated and least religious voters, with men, and with middle-aged voters at the prime age for economic insecurity for blue collar workers, 45-64. Rubio does best with the most educated voters and voters age 30-44; he would probably also do best with urban voters if we had data for South Carolina (Rubio won the cities in both Iowa and South Carolina but got killed in New Hampshire’s smaller cities).

Where Cruz has done best is with youngest voters, who may respond best to his calls for principled governance; he has also capitalized on wooing the younger, libertarian-leaning Ron Paul voters, plus campaigns that spend a lot on turnout and digital outreach will tend to overachieve with younger voters.

We covered the ideological splits already; Cruz has dominated the Very Conservative voters, winning them in Iowa and South Carolina but significant margins. Kasich does best, and Cruz worst, with independents, which is mixed news for both since the blue states later in the calendar have more closed primaries, but also fewer of the kinds of Republicans that like Cruz. Rubio has done best with the voters who decide late, whereas Trump has tended to just coast on his early, built-in support – an encouraging sign in the sense that it means Trump’s coalition is not growing with momentum voters, but a discouraging sign in that his early leads can be hard to overcome and his longtime supporters nearly impossible to dislodge.

Splits by what voters say they want in a candidate are less fixed categories, as they tend to reflect the candidate preference as much as they drive it. Trump, naturally, swamps the others on “tells it like it is,” “political outsider,” and immigration voters (the last suggests that Cruz maybe needs to accept that he’s not getting much more mileage out of the issue), and gets nowhere with “Values” voters. Rubio, unsurprisingly, wins handily with voters concerned with winning in November.

But if there’s a big alarm bell here for Cruz, it’s that his weakest showing on the issues is with voters concerned about the economy and jobs – the single largest issue category. Again, that partly reflects the fact that his principled message attracts voters who are more apt to cite terrorism, immigration or the size of government, but if I was advising Cruz’s campaign, I’d be worried about not attracting enough voters who think in terms of kitchen-table issues.

Where Is The Bar?

So because the map gets a lot weaker for Cruz after March 5, he needs to make his stand now. But what counts as enough of a stand? I don’t think artificial thresholds are a useful exercise, and because only one state in that stretch is really winner-take-all (Oklahoma, although the winners should sweep the bulk of the delegates in Texas, Georgia, Alabama, Virginia, Arkansas and Vermont), it is not realistically possible with the field as currently structured for Cruz to land an actual knockout blow of the sort he may originally have planned.

But Cruz needs to be looking at a couple of metrics:

1. Can he keep on Trump’s heels in the delegate race? I ran a simulation on RCP’s delegate counter yesterday and came out at Trump 303, Cruz 220, Rubio 170, Kasich 40 after March 1, and there’s any number of assumptions in that exercise, but it seems to me that if Cruz needs to be closer than that if he wants to show himself the strongest long-term contender. If he comes out of March 1 behind Rubio, or even closer to Rubio than to Trump, it gets a lot harder to argue that he is the right guy to close the sale.

2. Can he hold Texas? He’s been ahead in the polls in his home state, which has the second-largest delegate haul of any state (to California, which votes June 7), and Trump has yet to show any ability to close well. Cruz needs to win Texas and clean up the bulk of its delegates.

3. Can he beat Trump outside Texas? Both Cruz and Rubio need to prove they can do that, as Cruz has topped him only in Iowa, Rubio nowhere. With a raft of states more naturally favorable to Cruz, the time to start making that dent is now. With the field still split five ways, it’s not realistic to ask either to finish ahead of Trump in a large number of states, but if Cruz is going to keep “I beat Trump in Iowa” as a talking point after March 1, he’ll need to show it’s a trick he can repeat.

4. Can he consistently beat Rubio? Cruz in the last two states has more or less run even with Rubio, beating him after Rubio’s bad debate in New Hampshire, losing in South Carolina. If Cruz beats Rubio in most of the March 1 states, he can make the argument that time is running out for Rubio to come from behind him. But because those states are Cruz’s best turf, if he’s behind Rubio in as many states as he’s ahead, that’s a bad comparison for Cruz. And he might have trouble doing that – while the South is good for Cruz, parts of it are quite natural for Rubio as well, like Virginia, Tennessee and Georgia. Rubio was ahead of Cruz in the most recent Georgia, Virginia and Minnesota polls, and other March 1 states have old polling that shows neither with a major edge at the time.

5. Can he win his voters? Losing Evangelicals to Trump in both New Hampshire and South Carolina, and losing Very Conservative voters to Trump in New Hampshire, was a blow. A candidate like Cruz needs to protect his base and finish consistently first in his best demographics.

6. Can he broaden his base? That’s the real challenge here, and maybe the biggest question of all – no matter well Cruz does with Evangelical Christians and Very Conservative voters on March 1, if he continues to run a distant third or fourth with the rest of the primary electorate, it’s hard to see why anyone would consider him the logical option to take on Trump.

Like I said, for Cruz’s supporters none of this may seem fair. But then, 17 candidates started this race, and 14 of them would kill to have the opportunity that sits in front of Cruz over the next two weeks. If it’s going to happen, his will have to be Ted Cruz’s hour.