List Senators up in 2016 from most vulnerable to least vulnerable

[IceSpear]

1) Kirk (R-IL)
2) Johnson (R-WI)
3) Toomey (R-PA)
4) Reid (D-NV)
5) McCain (R-AZ) (Because of both primary and general danger, otherwise he'd be lower)
6) Ayotte (R-NH)
7) Rubio (R-FL)
Burr (R-NC)
9) Bennet (D-CO)
10) Blunt (R-MO)
11) Murkowski (R-AK) (Because of primary danger)
12) Portman (R-OH)
13) Coats (R-IN)
14) Isakson (R-GA)
15) Lee (R-UT) (Because of primary danger, though possibly general if Matheson runs)

The rest are irrelevant barring unexpected retirements, exceptional/terrible recruits, and/or freak circumstances (which, to be fair, probably will occur in at least a couple).

[IceSpear]

Are these lists exclusively to be based on the General Election? Because someone like Brian Schatz is perhaps the safest under that criteria, but under criteria based on a primary election... not so much.

No, they can be for primaries too. That's why many are including Murkowski and Lee in their lists. I personally don't see Schatz vulnerable to another primary challenge, I know he barely survived that primary challenge, but he just got elected with 70% of the vote.

Plus he just won the primary and the election in his own right (as opposed to being just an appointment), and Abercrombie is gone. All of Hanabusa's ammunition against him is gone now.

[IceSpear]

Bennet
Toomey

The delusion is real.

So, when exactly is the PADP going to stop desperately searching for an alternative to their presumptive nominee?

Who cares? The PA Dem establishment is still bitter that Sestak upended Specter in 2010. It has nothing to do with candidate quality, which he already proved by far outperforming expectations in 2010.

[IceSpear]

Bennet
Toomey

The delusion is real.

So, when exactly is the PADP going to stop desperately searching for an alternative to their presumptive nominee?

Doesn't really matter. Considering Democrats' worst-case scenario was a 2-point loss, saying Colorado is more vulnerable than NH, FL (in an open race), or especially PA is blatant hackery.

Why is a 2-point loss Democrats' worst-case scenario? Republicans won the combined House races in Pennsylvania in 2014 by 11 points, and Arlen Specter beat Joe Hoeffel by 11 in 2004. Yes, 2010 was in a very good environment, but Toomey has incumbency now, he's perceived as more moderate now than he was then, since he's been very vocal about reaching across the aisle, and Joe Sestak has become a has-been after being inactive for six years.

Florida is very unlikely to be an open seat. Ayotte is definitely vulnerable in a Democratic wave, since her state is prone to large swings, but she's leading Democrats' dream nominee Maggie Hassan in early polling and overperformed significantly in 2010 -- I think she should be fine in a neutral year.

Colorado is more vulnerable than it looks because of the paucity of Republican offensive opportunities -- if 2016 is a good Republican year, Colorado will be targeted with more offensive money than any other opportunity over the past few cycles, and Bennet will be very vulnerable. (Of course, in a good Democratic year, it might just fall off the list of competitive races, as states like North Carolina, New Hampshire, and Ohio -- perhaps even Pennsylvania -- will in a good Republican year).

(not to mention saying that Reid is "objectively" more vulnerable than a largely unknown senator deep on enemy turf)

Yeah, it's better to be largely unknown than hated.

To further add, Sestak was never exactly well-liked in the PADP anyway, and they don't really have anybody who could compete with him. I'm not counting him out.

I'm not counting him out either, but I'm not going to count him as a super-strong candidate that starts off favored against a respected incumbent Senator in a swing state.

I wouldn't say a 2 point win is Toomey's ceiling, but your examples for why it isn't are pretty lousy. So the GOP swept a bunch of House seats that Dems didn't even make a token effort to compete for in a low turnout midterm wave? Yeah, not exactly indicative of a lot. The Arlen Specter comparison is also fairly irrelevant. Specter was a long time respected incumbent who still had appeal to many people who are now fairly solid D voters, particularly around Philly and the suburbs.  Toomey, as a mostly unknown first termer, could only dream of matching that appeal. In addition, the country has become much more polarized since 2004. In short, the examples you provided aren't exactly good for any kind of extrapolation. An equivalent example would if Democrats said Hillary, Sestak, and all other statewide Dems were locks to win by double digits due to Wolf's double digit win even in this tough year. It just doesn't make any logical sense.

Let's compare to Colorado, where the GOP's strongest possible candidate ran a top notch campaign against a guy who spent the entire election talking about how his opponent wanted to ban birth control and condoms, all during an enormous low turnout midterm Republican wave. He won by 2 points. At least one of those factors won't be present in 2016 (low turnout midterm), and Bennet will probably learn from Udall's horrific campaign as well, so that would remove that factor. To say he's more vulnerable than Toomey is pure hackery. And just for the record, I wouldn't say Sestak is favored either. I'd say it starts as a pure toss up, but I'm not sure how one could call CO anything other than lean D (or at least tilt D) at this point.

[IceSpear]

2014: "Democrats' worse case scenario in Colorado is a one-point win. Udall is a stronger candidate than Bennet and 2014 will not be as bad a year for Democrats as 2010. Udall will win."
2015: "Democrats' worse case scenario in Colorado is a two-point loss. Bennet is a stronger candidate than Udall and 2016 will not be as bad a year for Democrats as 2014. Bennet will win."

Nobody said that first one. Everybody acknowledged Gardner was a much stronger candidate than Buck, even though Udall seemed stronger than Bennet, clearly that wasn't the case. Realistically, could Bennet possibly run a worse campaign than Udall did, short of having an affair, hiding money in a freezer, or pulling a Casey/Warner and going out to lunch? And while it did seem like 2014 couldn't be as bad as 2010...well, you won't see many Democrats making that mistake again in 2018. Still not comparable to a presidential year, where even if Republicans do well, they'll still have to contend with much higher turnout among Democratic leaning demographics.

Oh, and nobody's saying Bennet will "definitely win" either, just that he's clearly not near the top of the most vulnerable list.