If you’re suspicious of the result, I don’t blame you. The numbers come from the Our Principles Super PAC, one of the few outside groups that’s been bombing Trump with attack ads this cycle. The numbers here are exactly what you’d expect to see if you’re trying to motivate people to beat Trump, showing his nearest rival, Rubio, behind but not so far behind that all hope is lost. Most recent polls of Florida have Rubio trailing Trump by double digits — but not this one. Keep hope alive! Meanwhile, although the Miami Herald got a copy of the poll’s results, they weren’t given the crosstabs so that they could scrutinize the poll’s methodology. Hmmmm. Could be that this is little more than a dubious confidence-building exercise to try keep Rubio’s base, and undecideds, engaged.

But there’s reason to take it seriously too. Tarrance, the pollster that conducted the survey, is an established firm. And while it’s true that most recent polls of Florida show Trump far ahead, not all do. A survey taken late last month by AIF had the race Trump 34, Rubio 27, Cruz 17 — strikingly similar to what today’s poll shows. AIF used a stricter voter screen than other pollsters in Florida, looking to see if respondents had voted previously in the state instead of simply asking them if they intended to vote in the primary this year. Florida, remember, has a closed primary, so unless you’ve already registered as a Republican, you’re not casting a ballot on March 15th. That’s a problem for Trump, and that adds some credence to the idea that the race there is closer than thought.

Anyway. False hope or tight race? You make the call:

Like the AIF poll from last week — the one that found the tightest Florida race — Tarrance identified likely voters based on their voting history listed in the state’s voter file. Quinnipiac and PPP let respondents self-identify if they’re planning to vote. That may capture voters who are not registered Republicans and may not vote in Florida’s closed primaries, which shut out independents and don’t allow cross-over votes from Democrats. Trump has done better in states with open primaries because he draws support from independents and Democrats.

The new poll offered a glimmer of hope for Rubio in that he continues to have the most favorable rating among all the candidates. That gives him the most room to grow among undecided voters; according to the poll, 24 percent of respondents are still picking a candidate. Sixteen percent are leaning toward one, and 57 percent have already decided…

Asked if they would “never” vote for Trump, 32 percent of respondents agreed. Of those, 30 percent felt “strongly” about their position.

A lot of those supposed NeverTrumpers in Florida will come around by the time of the general election but obviously this isn’t a state where you want Republicans staying home. If Hillary Clinton wins all of the states that have been reliably blue in presidential elections over the past 20 years, she could clinch 270 electoral votes simply by adding Florida to her total. Florida is an absolute must-win, no questions asked, for the GOP this fall. If #NeverTrump is a major factor there, the party’s chances are hanging by a thread.

On the other hand, don’t underestimate how challenging even a “good” poll result like this one is for Rubio. The conventional wisdom is that Trump will lead among early voters because his supporters are more enthusiastic than everyone else’s and because he’s led steadily in the polls for ages. Even if Rubio catches up over the next 10 days and leads the polls by five points on election day, that might only be good enough to make the final result a toss-up between him and Trump given the early and absentee votes already banked. Meanwhile, he’s got Kasich and Cruz pulling votes from him here, probably on either side of him. In fact, if these Tarrance numbers are accurate, I’m skeptical that either one of them dropping out today would be enough to guarantee a Rubio win. To edge past Trump, he’d have to take nearly all of Kasich’s centrist voters. He’d have more of a margin of error if Cruz dropped out — winning two-thirds of Cruz-ites would probably do it — but even that seems like a tall order given how Cruz’s base is a mix of conservatives and populists. How many pro-Cruz populists prefer Trump to Rubio? If Trump pulled just six of Cruz’s 16 percent, that would be enough for him to hold on and win even given Tarrance’s Rubio-friendly results here.

Having said that, from what I’ve gauged on Twitter over the past 24 hours, March 16th is going to be an unholy bloodbath between Rubio fans and Cruz fans if Trump ends up winning in Florida by a margin that’s less than Cruz’s share of the vote. Rubio fans will blame Cruz for the defeat; Cruz fans will say Rubio had it coming after he insisted on contesting Texas. Recriminations galore! See this post last night for more details: Supposedly, the two conservative candidates should be colluding with each other right now to deny Trump the maximum possible number of delegates. Instead, by contesting a winner-take-all state which he has zero chance to win, Cruz is doing nothing more nor less than trying to guarantee Rubio’s defeat, which will probably knock him out of the race and force conservatives to unite behind Cruz. That’s good hardball politics, but it’s also incredibly foolish given that Trump winning Florida dramatically increases his odds of winning the nomination. Eliana Johnson ran the numbers on different scenarios that might play out on March 15th and found that if Trump wins Florida and Ohio (as well as 45 percent of all other delegates at stake that day, which seems feasible), he’d need to win just half of the delegates in all remaining states before June to clinch the nomination. Assuming Rubio and Kasich dropped out quickly upon losing their home states, Trump would simply need an even split with Cruz going forward to wrap things up — and most of those races will be run in blue and purple states, not Cruz country. Even if Cruz won a majority of the remaining delegates, Trump would get to the convention close enough to an overall majority that it’d be next to impossible to deny him. If your top priority as a voter is stopping Trump rather than electing Marco Rubio or Ted Cruz specifically, Cruz’s plan to compete with Rubio in Florida is a disaster in the making.

If, however, your top priority is electing Ted Cruz, then it’s arguably justifiable. The man’s not playing to elect “a conservative,” after all, he’s playing to win. He’s got a shot to kill off a formidable rival 10 days from now. He’d be a fool not to take it, especially after Rubio’s decision to compete in Texas likely cost Cruz the absolute majority he needed there to make Texas’s delegates winner-take-all. (Although re-read last night’s post to see why Texas and Florida aren’t perfectly analogous.) The hard question for Cruz fans is this: Who really benefits more from Rubio’s exit from the race, Trump or Cruz, if handing Florida’s delegates to Trump is a precondition of that exit? It’s important to understand that Cruz himself has virtually no path to clinching the nomination before the convention, even if he ends up in a one-on-one race with Trump. He might have had one if he had a shot at winning the windfall of delegates provided by Florida and Ohio — 165 combined — but he has no chance at those delegates. To get to a one-on-one, those delegates need to go to Trump, which means the very best-case scenario for Cruz is a brokered convention where he’s solidly in second place. Does that give him a much better chance at winning the nomination in Cleveland than he’d have if Rubio stayed in the race and the three of them fought it out until June? I honestly don’t know. I can see it both ways. At a brokered convention where the only remaining candidates are Trump and Cruz, Cruz could win by convincing Rubio’s delegates to break for him, enraging Trump fans. On the other hand, at a convention where Trump, Cruz, and Rubio are all still in the race, Cruz might emerge as the compromise candidate, the guy with some appeal to conservatives and populists. Trump fans might be less enraged at that since they prefer Cruz to Rubio, the only non-Trump alternative. And in a three-way race, with delegates spread more evenly among multiple candidates, Trump’s claim that he deserves the nomination because he finished first in delegates might seem weaker than it would in a two-man race. For instance, in a two-man race, we might get to Cleveland with Trump holding 48 percent of the delegates, Cruz holding 38, and the since departed Rubio and Kasich holding a combined 14. In a three-man race, we might see Trump with 40, Cruz with 32, and Rubio with 28. In which scenario does it seem “fairer” to make Cruz the nominee?

Long story short, I see more downside than upside to Cruz nuking Rubio in Florida. But then, while I’m pro-Cruz, I’m even more anti-Trump. As I said, though, there’s a fair chance that the results in Florida will resolve all of this. If Trump wins with 40 percent, Rubio finishes with 30, and Cruz takes 20, realistically Rubio wouldn’t have won even if Cruz had skipped the state entirely. Trump’s margin would have demanded too heavy a tilt towards Rubio among Cruz’s voters to make the outcome any different.