On Tuesday, Republicans will hold their first winner-take-all primaries of this election season. Republicans in Ohio and Florida will award their entire delegates—66 and 99 delegates, respectively—to the winner of the statewide vote, while Republicans in Illinois, Missouri, and North Carolina will allocate all the delegates from each congressional to district to the winner in that district. Many writers have spilled much ink on the significance of these primaries, but how much significance do they really have?

Clearly, two fairly large states throwing their entire delegation behind a candidate will have some effect. However, some pundits are arguing that it could all but end the race in Donald Drumpf’s favor, while Marco Rubio and John Kasich are counting on come-from-behind wins in their home states to keep their campaigns afloat. Laying aside our ideas of who should win and putting on our well-worn math nerd hats, how realistic is it to think that the primaries on Tuesday will change the entire race?

First, horrifically inaccurate as polling has been this cycle, it’s worth taking a look at the polls in the five states that will hold primaries or caucuses on Tuesday. In Ohio, Kasich and Drumpf have been exchanging leads, and in the most recent poll were precisely tied. Cruz, meanwhile, surged upward to within the margin of error of the polls, to make it a three-man race. In Illinois, it’s a faceoff between Drumpf and Cruz, with the most recent polls putting the two in an effective tie. Missouri is the same, and in North Carolina Drumpf holds first with Cruz a distant second. Florida, the crown jewel of the March 15th states, is also shaping up to be a Drumpf win, with the most recent poll putting him 20 points above Cruz, who slipped above Rubio to take second place.

The outlook for stopping Drumpf on Tuesday, then, is bleak. Kasich can’t do better than tie the New York philanderer, and that’s in Ohio, his home state. Rubio is even worse, not just losing by more than 20 points, but not even managing to be the first runner-up against Drumpf. Those who’ve pinned their hopes of stopping Drumpf on Rubio and Kasich taking Florida and Ohio’s delegates from him have little cause for hope.

Are the pundits right, then, that March 15th will end the race and ensure that Drumpf wins the nomination? Unsurprisingly given their record so far, no. The reason is that there’s something even bigger than the 165 delegates at play in Ohio and Florida. Cruz has been pleading for Republicans to unify to stop Drumpf for weeks now, because for weeks it has been clear that Cruz is the only one who could possibly beat Drumpf outright, without the rancor of a contested convention. If Kasich and Rubio were to win their home states and continue their vanity candidacies, it would mean that anti-Drumpf Republicans remain divided. The most recent polling indicates that Kasich and Rubio’s voters would overwhelming vote for Cruz if their candidate dropped out, breaking for the Texan over Drumpf 72-17.

How much of a difference would that make? As an old Danish joke says, “it’s difficult to make predictions, especially about the future,” but we can use the past as an analog. Particularly, we can look at how the race would shape up if everything continues as it has, and each candidate stays in and takes the same percentage of delegates they have been. We can then compare that scenario to how the race will shape up if Kasich and Rubio drop out and their supporters break for Cruz in the way polling suggests.

So far, 1,071 delegates have been allocated, and Drumpf has taken 460, or 43%, to Cruz’s 370, or 35%. If we go back, though, and assume that all the nonviable non-Drumpf candidates never entered the race and their support shifted to Cruz as polls indicate, that total shifts dramatically. Instead of 460, because of delegate allocation rules, Drumpf’s total falls to 261, or 24%, and Cruz’s rises to 785, or 73%. This is counterfactual, of course, but it does have the potential to tell us what the race would look like without Kasich and Rubio: Drumpf would take ~24% of the delegates and Cruz ~73%.

We have a simple model, so let’s apply it to possible results on Tuesday, and see what happens. If Rubio and Kasich win their home states and otherwise results follow polling, Rubio and Kasich will stay in, and the final result will be

Drumpf: 460 (current) + 531 (43% of the remainder) = 991 delegates
Cruz: 370 (current) + 427 (35% of the remainder) = 797 delegates

In this scenario, neither Kasich nor Rubio has a significant number of delegates, and unless Rule 40 were changed, would not be eligible to receive votes at the convention. If Rubio and Kasich manage to pull off their upsets, this plan works, and Drumpf is stopped well short of the 1237 delegates required to win outright. The convention would be contested, and the Republican Party would come out the other side bruised and battered, but likely without having committed the political suicide that a Drumpf nomination would represent.

What about the worst-case scenario? What if Drumpf takes every state on Tuesday, and sweeps out of the beginning of winner-take-all with a crushing victory? Are the pundits correct that the race would be over and Drumpf the de facto nominee? Well, in that case Rubio and Kasich would likely drop out, allowing Cruz to unify the anti-Drumpf vote—remember, that means Cruz should start taking ~73% of the delegates and Drumpf ~24%. Would it be enough to make up ground on Drumpf’s huge lead?

Drumpf: 460 (current) + 165 (OH and FL) + 301 (24% of the remainder) = 926 delegates
Cruz: 370 (current) + 906 (73% of the remainder) = 1276 delegates

That’s right. Even if Drumpf sweeps the primaries on Tuesday, as long as the outcome knocks Kasich and Rubio out of the race, our (admittedly simplistic) model still predicts not just that Drumpf fails to win a majority, but that Cruz clinches and avoids a contested convention. Now, the takeaway isn’t that Cruz will clinch if Rubio and Kasich drop out—the model doesn’t have that kind of granularity—but that it’s more than possible to stop Drumpf even if he wins on March 15th, and that we’re more likely to stop Drumpf if Rubio and Kasich lose their home states.

That’s huge. That destroys the entire purpose of the “strategic” voting that people like Mitt Romney have been advocating. The chief hurdle to stopping Drumpf isn’t the chance that he’ll win Florida and Ohio, it’s the reality that Kasich and Rubio’s presence in the race gives delegates to Drumpf. In Texas alone, for example, Rubio gained next to nothing and gave Drumpf 48 delegates by holding Cruz under the winner-take-all threshold. In Maine, Rubio gained nothing and kept Cruz from breaking the winner-take-all threshold, thus giving delegates to Drumpf. Even the lead Drumpf would amass by winning the first (and largest) winner-take-all states wouldn’t be enough to withstand the united Republican Party.

There’s even better news for those who oppose fascism, though, and a bigger refutation to the establishment GOP’s strategic voting: even if Drumpf wins all five March 15th states and Rubio and Kasich continue their vanity candidacies, Drumpf still only reaches 1156 delegates in this model—not enough to clinch the nomination. It’s nearly impossible for Drumpf to win a majority outright at this point, unless his performance improves dramatically.

In other words, the idea of Kasich and Rubio staying in to prevent Drumpf from winning a majority is fundamentally flawed. Instead, all they’ll do is make it more likely for Drumpf to win. Conservatives in Florida and Ohio who want to stop Drumpf should have a twofold goal: trying to stop Drumpf directly by voting against him, and stopping Kasich and Rubio. It is only if those two are forced out that anti-Drumpf Republicans have the best chance of winning at the convention or any chance of beating Drumpf outright.