(CNN) Judge Brett Kavanaugh is more likely than not going to win Senate confirmation to join the Supreme Court.
After all, most nominees to the Supreme Court win confirmation. And with at least 50 Republican senators able to vote, President Donald Trump's pick doesn't need to win a single Democratic vote.
But is it possible Kavanaugh doesn't get confirmed? The answer to that question is yes. It's likely to happen only if the public turns against his confirmation, though, as they did with Robert Bork in 1987.
Last year, I fine-tuned a model that tries to explain the factors behind how each lawmaker voted in Senate confirmations since Bork and then tries predicting how Supreme Court future confirmation votes will go. It came within one vote of correctly predicting how many votes Neil Gorsuch would get during his Senate vote. We shouldn't expect the model to be this good every time. Still it gives us a good idea of what to expect.
The model looks at the ideology of each senator and nominee, how often each senator votes with his or her party on votes where at least 50% of Democrats vote one way and 50% of Republicans vote another, the party of each senator, the qualifications of the nominee, the president's popularity and the nominee's popularity. (No polling was done on Anthony Kennedy's popularity before his confirmation.)
We already know or can estimate the values of most of these variables. Trump's approval rating is, for example, in the low 40s, and Kavanaugh is seen as a very conservative and a very well-qualified nominee.
What we don't know is how the public is going to react to the nominee. I reran the model with the known variables and then plugged in different popularity ratings for Kavanaugh. The model suggests that the only way Kavanaugh goes down is if he ends up as unpopular as Trump.
With a net favorability rating of between -10 and -15 percentage points, Kavanaugh would be predicted to get 49 votes. That's, of course, one short of the 50 he needs. In this scenario, Sen. Susan Collins of Maine, who is the most moderate and has voted against her party most often, is projected as the Republican senator who'd vote against Kavanaugh.
It's not too difficult to imagine why this would occur. Given how Trump seems to manage to polarize nearly everything he touches, it seems at least somewhat possible that his Supreme Court nominee ends up as unpopular as he is. Furthermore, Democrats and liberal groups are likely to be more active during this confirmation process than during Gorsuch's. This could help rally the Democratic base and bring up some swing voters.
Put another way, Kavanaugh needs to be unpopular enough that he won't be seen as an electoral liability for moderate Democratic senators to vote against. These include red state senators up for re-election such as Joe Donnelly of Indiana, Heidi Heitkamp of North Dakota, Joe Manchin of West Virginia and Claire McCaskill of Missouri.
On the other end, Collins needs to feel electoral pressure to vote against Kavanaugh.
Keep in mind, Democratic senators and liberal groups were able to pull off such a feat with the Bork nomination. Bork's net favorability ended up at -11 points nationally. Bork finished the nomination process so unpopular because Democrats and liberal groups went after him for his views on the Constitution, such as privacy and the right to an abortion. They also brought up his views on executive privilege, which could be the cornerstone of a Democratic argument against Kavanaugh.
The obvious difficulty for Democrats is that it's unlikely Kavanaugh will end up as unpopular as Bork was or Trump is. Among those who form opinions of Supreme Court nominees, their feelings are usually positive.
Kavanaugh wins confirmation under most realistic scenarios.
If he has a net favorability rating equal to the average nominee since 1987, he'd be expected to win somewhere in the area of 54 to 56 votes. Red-state Democrats will feel pressure to vote for Trump's nominee in this instance. Collins will feel safe going with her party. (Something she was more likely to do in 2017 than in any previous year of her Senate career.)
Perhaps the most probable scenario is for Kavanaugh's net favorability rating to end equal to Gorsuch's. He'd probably win somewhere between 51 and 53 votes in this case. Again, he'd take all the Republicans and a few Democrats.
Now, the model mis-predicts by four votes on average. So given the relatively close vote margins, it's conceivable that Kavanaugh wins confirmation even if he is really unpopular. He could also be voted down if he is relatively popular.
Still the point remains: Democrats' best chance to sabotage the nomination is by making Kavanaugh disliked.