Analysis from the Manifold newsletter:

West Virginia: Glenn Elliott (D) vs. Jim Justice (R)

• Democrat + Other win chances (Manifold, 16 traders): 4%

• Democrat win chances per models: <1% (The Hill), <1% (538), 0% (Split Ticket)

• Expert forecasts: Solid R (flip) (Cook), Safe R (flip) (Sabato), Solid R (flip) (Inside Elections)

• Poll results (538): Justice+34 (more like Justice+10 vs. Manchin, but small sample size)

• Election result in 2018, a Democratic wave year: D+3 (Manchin vs. Morrisey)

• Election result in 2012, amid Obama's reelection: D+24 (Manchin vs. Raese)

• Recent West Virginia presidential election results: Trump+39 (2020), Trump+42 (2016), Romney+27 (2012) 

Manifolder commentary from a year ago:

• Conflux: Do people actually think [Manchin would] likely lose by more than 10? He won by 3 in 2018. Sure, that was a wave year, but a 13-point swing is a lot

• SemioticRivalry: I think Manchin is much less popular than in 2018 and this cycle is likely to be less Democratic, and he faced a weak candidate in '18,  all of which works in my favor. But there's another more important factor, which is that it's a presidential year. Holding everything else constant, you would expect a Democratic incumbent in WV to do about 10% worse in a presidential year than in a midterm, due to the very consistent trend of dramatically less ticket splitting in presidential years. Even if it was a very good D year and he was equally popular, I still think he may lose by around 10. 

• AndrewG: I mean, he's 75 and didn't even get a majority of the vote in 2018. I'd say there's a 30% chance that he runs, and a 30% chance of winning if he does.

Conflux commentary:

Joe Manchin’s West Virginia Senate seat hasn’t been held by a Republican since 1959. (To be fair, from 1959 to 2010, it was held by only one Democrat: Robert Byrd, the longest-serving senator in US history.) Its transition from blue state to solid red state came suddenly in the last few decades, amid Democratic opposition to coal combined with increasing racial/educational polarization. 

Manchin founded the Enersystems waste coal company in 1988, was elected West Virginia governor as a Democrat in 2006, and got elected to the Senate in 2010 after Byrd died. He was reelected in a landslide in 2012, winning tons of Republican votes. But as partisanship strengthened, he knew 2018 would be a close race — and was tired of the Senate. He literally said, “This place sucks.” But a few days before the filing deadline, he put his name in the race, and narrowly beat a weak opponent. In his last term, Manchin has held lots of leverage on key legislation, and proved that he’s to the right of the average Democrat, but far to the left of the average West Virginian senator. He’s surely the senator with the highest Wins Above Replacement.

But by now, the state is even redder, and Manchin (who’d had to spend over $75,000 on private security) is even more tired of Congress. He’s become an Independent, and flirted with running a third-party presidential campaign. His Senate reelection would have been an uphill battle — but his retirement assures defeat.

Who will replace him? Jim Justice, until recently the richest man in West Virginia. (He has paid millions in coal mine safety debt, a worrying phrase. Ugh, at least he has a dog?) “Big Jim” was a Republican, then he became a Democrat to run for governor in 2016, then he switched back to Republican.1 Now he’ll represent WV in the Senate.

But for key votes? In the words of a Democratic state delegate, “There’s an old saying that 80 percent of success in life is showing up, and Jim Justice rarely shows up.”

Manchin has apparently left the Democratic party, leaving open a run for an independent race. He has not yet declared his intent to run a third-party race, but he's certainly opened the doors to it!