Pretty stupid to conduct a poll over five days when a presidential debate is being held on day four, no? Either do the whole poll beforehand so that you have a pre-debate snapshot of the race for comparison later or do the whole poll after the debate to get a sense of the new state of the race. As it is, there’s no way to draw firm conclusions from these numbers. Did Rubio and Cruz fail to get a bounce from the debate or had the bounce simply not happened yet by the time this poll concluded? Did Trump decline because of the debate or had he already declined beforehand and actually came back a bit afterward thanks to his performance?

Either way, it’s harder this morning to argue that that NYT/CBS bombshell last week showing Carson ahead of Trump for the first time nationally was an outlier. NYT/CBS had him ahead 26/22; WSJ/NBC has it 29/23. And even that margin understates how well Carson is doing at the moment. When you combine Republican voters’ first and second choice, Trump’s suddenly behind by 15 points.

That’s the first time any Republican has cracked 50 percent combined in one of WSJ/NBC’s polls. Note the trend lines too. Rubio jumped from 16 to 26 percent between July and September; Ted Cruz jumped from 17 to 23 percent in just a few weeks last month; and Carson has built on his support steadily in every poll since June. Trump, however, has been stuck in the 30th percentile since his campaign took off this summer and dropped more than even Rubio did in the course of October. If you’re looking for an argument that he’s plateaued for now, there’s your data. That doesn’t mean he can’t win — 25-30 percent in Iowa or New Hampshire could be enough if the rest of the field splits evenly enough — but you’d much, much rather have Carson’s trend lines, needless to say, than Trump’s.

On the other hand, here’s yet another poll challenging the idea that Republicans will break against Trump in the end because they see him as unelectable. When Republicans are asked who has the best chance of winning the general election, it’s Trump, not Carson, who leads:

The supposedly electable Jeb Bush, who’s expecting to win a head-to-head race with Trump if only he can figure out a way to make that race happen, is a dismal fourth place and in single digits. (Ted Cruz, who’s fourth overall in the poll, is fifth here with just six percent. That’s a bad omen if he ends up head-to-head with Rubio and undecideds start tilting based on perceived electability.) Trump also does much better than Bush when Republicans are asked whether they could see themselves supporting a given candidate for president. Trump splits 60/37 on that question, comparable to Rubio at 59/32. Bush splits … 45/52, the first time he’s been underwater in a WSJ/NBC poll. The reason you’re seeing big donors like Paul Singer and Sheldon Adelson swinging behind Rubio now may be because they’ve concluded from numbers like these that Bush wouldn’t necessarily prevail in a head-to-head final two against Trump or Cruz. The donor class is broadcasting its judgment about who the most electable center-right candidate is. And based on the polls, they’ve chosen wisely.

In fact, here’s one more bit of data for you. This comes not from WSJ/NBC but from YouGov’s poll yesterday. And this one was conducted after last Wednesday’s debate.

The GOP establishment is filled with rich, savvy investors. Looking at those numbers, who seems like the better investment to you? QED.

By the way, thanks to the WSJ/NBC poll and last week’s NYT/CBS survey, Carson now trails Trump in RCP’s poll average by just 1.3 points. If/when we see a poll out of New Hampshire showing Trump in trouble there, the media’s going to turn the news cycle into a full-bore Trump-crash collapse-apalooza.