I offer the poll not as an argument against doing it, just as a data point on what the future might look like. If ever there were an issue where public opinion is worth ignoring on moral grounds, this is it. To borrow Mike Lee’s analogy between Roe and Dred Scott, it would be repulsive to reason that slavery shouldn’t be outlawed because it might prove politically unpopular. Let justice be done though the heavens fall.

And the heavens may fall for the GOP. The numbers aren’t a moral caution, they’re a caution about the long-term practicability of overturning Roe. If criminalizing abortion (in red states at least) triggered a ferocious backlash that swept Democrats to power, how long would it be before they had five votes on the Court again to reinstate abortion on demand? How much other policy damage would be done by the Democratic majorities elected by the pro-Roe backlash? Just last year, by one vote in the Senate, Democrats barely avoided a hard reckoning about how quickly seismic policy changes can be undone when they’re unpopular with the public. The GOP may face the same reckoning.

But hey. Thanks to Roe itself, this matter is now out of the hands of everyone except Trump and the Court’s nine justices — and of course Susan Collins and Lisa Murkowski. (Imagine their reaction to these numbers when they see them.) We are bystanders on too many matters in our alleged democracy.

Even Republicans are only tepidly in favor. Poll junkies might note that “KFF” stands for Kaiser Family Foundation, an outlet that’s been relentlessly pro-ObamaCare and might be expected to have unusually rosy numbers for the left on whether abortion should be repealed. And it’s true, other outlets have showed the spread to be narrower on this question. In 2013, for instance, Gallup found the public split 29/53 on whether Roe should be overturned, down from 25/66 in the mid-2000s. Not great, but not as bad as Kaiser. Other reputable pollsters, though, have seen spreads closer to KFF’s. A snapshot of Pew’s data over the years:

Pew’s numbers as of December 2016 are nearly identical to KFF’s numbers today. There’s also reason to wonder if the smaller divide spotted by Gallup in 2013 hasn’t grown today. That’s because at the time Gallup saw an unusually high number of “no opinion” responses to the question about Roe, which they noted was “largely the result of a growing percentage of young adults aged 18 to 29 expressing no opinion.” Five years later, as the Obama era has transitioned to the Trump era, that ambivalence may no longer obtain. Young adults, particularly young women, lean heavily towards the Democrats. Even if they’re still in the “no opinion” column, it probably wouldn’t take much Democratic messaging to push them firmly into the “save Roe” column. And there will be a lot — a lot — of Democratic messaging about this in the months ahead, roughly 10 times more hysterical than their messaging is on every hot-button issue when they’re losing.

Here’s another number from Pew’s data worth paying attention to. We’ve gotten used to partisan splits on key issues mirroring each other in polls — if Democrats support something 80/20, Republicans opposing it 20/80 is usually a safe bet. Not this time.

If we’re expecting an equal number of enthusiastic Republicans to hold off the Democratic hordes in a clash over Roe, we might have to adjust our expectations. Especially since, unlike KFF, Pew has Republicans also in favor here of keeping Roe at least partly intact (i.e. not completely overturning it). Gulp.

There are things Republicans can tell themselves about KFF’s numbers to steel their resolve to nuke Roe. One point is that major changes to any important policy are usually unpopular just because the public fears disruption. “Status-quo bias” and “loss aversion” are real things. The party should expect an initial wave of anger if Roe is tossed followed by a cooling-off period in which Americans gradually acclimate to the new political reality. Even ObamaCare finally grew popular-ish last year when the GOP was finally on the brink of blowing it up. That’s because the same status-quo bias that had hurt the law for so long after it was passed finally swung around and began to work for it as O-Care itself became the status quo. Surely support for the post-Roe regime would rise over time.

It’s just that the GOP might have to lose the House, the Senate, the White House, and eventually the Supreme Court while it waits.

Another thing Republicans can tell themselves is that Democrats’ most apocalyptic fears about what the post-Roe reality will look like will be exposed as hysteria once we finally arrive at it. The analog is gay marriage: The numbers in favor of legalizing that have continued to rise even after SCOTUS legalized the practice as some of the doubters who feared a parade of moral horribles in post-SSM America felt relief that there’s no parade after all. (Yet?) Democrats will spend the next three months assuring the country that ending Roe will usher in the actual Handmaid’s Tale, replete with red dresses and white bonnets. In reality, blue states would quickly legalize abortion on demand. Purple states may also be stuck with legal abortion since Democrats might have enough leverage in state legislatures to block Republicans from changing the current abortion laws by statute. It may even be that the Roberts Court will compromise by modifying Roe rather than overturning it, holding that abortion must be legal within the first 12 weeks but that the states have carte blanche to regulate after that. (Which would make America’s abortion laws look more like those of our socialist friends in Europe.)

Once voters realized that abortion would still be legal in half the country or more and might conceivably still be legal everywhere in the first trimester, some of the opposition to overturning Roe would ease. But how much? And how soon? Would Americans adjust before or after they’d set in motion the electoral forces that would enable Democrats to reinstate Roe? Would Democrats even try to reinstate it once they had the votes on the Court to do so, knowing that status-quo bias might at that point favor the GOP’s post-Roe regime? These are the stakes.