Preliminary 2016 Senate Race Rankings - No Tossups

[Smash255]

Since you said no toss ups, I put "tilt" ratings where I would normally put toss ups.

Alaska: Likely R - Since Murkowski will actually be on the ballot next time, it's hard to see her losing, but in a potential three way race it could be possible.
Arizona: Lean R - If McCain doesn't retire, he'd get beaten in the primary anyway.
Arkansas: Lean R - If Mike Beebe runs it's tilt D. But if he doesn't it's likely R. So I split the difference.
Colorado: Tilt D - What happens with Udall this year will speak volumes for Bennet. Fortunately for Bennet, he will benefit from presidential year turnout.
Florida: Tilt R - Lean R if Rubio doesn't run for president, but it's looking like he will.
Georgia: Lean R - Likely R if Isakson runs, tilt R if he retires and Dems get Nunn to run again.
Illinois: Lean D takeover - Hard to see how Kirk survives considering he barely beat a horrible opponent in a GOP wave year. He has no groundswell of popularity, he got hammered for lying about his war record in 2010 and won clearly as the lesser of two evils, not as an above the fray moderate. That said, Dems could snatch defeat from the jaws of victory here by nominating another horrible candidate (it's Illinois after all).
Indiana: Lean R - Coats could retire, even if he doesn't it's possible he could be beaten. Ellsworth would've given him a tough race in a non wave year.
Iowa: Likely R - Assuming Grassley doesn't change his mind. If he retires, tilt D.
Kentucky: Likely R - Whether or not Paul runs for president. Though Steve Beshear running would make it tilt R.
Louisiana: Likely R - Too much in flux here to say anything, but Dems have slim odds.
Missouri: Tilt D takeover - Roy Blunt is not popular at all, and Hillary should do much better here than Obama. It feels weird putting it as tilt D, but yeah, no toss ups.
Nevada: Tilt D - So much depends on who runs. Democrats would be best without Reid, and Sandoval would put the Republicans in control if he ran (lean R minimum). In generic D vs. generic R, the Democrat has the edge though.
New Hampshire: Tilt R - This will be hotly contested, but NH is quirky enough that Ayotte could survive even in a good Dem year.
North Carolina: Tilt R - I think Burr survives by the skin of his teeth, even if Hillary narrowly wins the state.
Ohio: Lean R - Portman could be primaried, which would make this tilt...R I guess? This race should be competitive, but Portman starts with the edge.
Oregon: Likely D - Safe if Wyden runs for re-election.
Pennsylvania: Tilt D takeover -  Toomey will be in huge trouble during a presidential year with higher turnout. Since I have to pick a winner, I'm giving it to Sestak by the skin of his teeth.
Wisconsin: Lean D takeover - Johnson will probably be a one termer. He's done basically nothing he needs to do to get re-elected in a blue state. His only hope is another Republican wave dragging him over the finish line or a crappy opponent.

Everything else is safe. Overall net seats is D+4.

Specter survived in 2004 and 1992 when PA went Dem. What's the difference? And Kirk is shown to be able to hold his own based on early polls.

Nearly every incumbent holds their own in early polls.

Then why are we quick to call them gonners? I get that their states are left leaning, but Heitkamp won when her state went overwhelmingly red? Or why did Brown only lose by high singles when MA went overwhelmingly blue? I'm pretty sure Toomey and Kirk are what loons would call RINOs, so moderacy is no issue. And who is to say that those early polls aren't genuine strength?

Toomey doesn't have the suburban Philly base Specter had.  Kirk hasn't done anything to hurt or help his chance in 2016 (though his health issues could still hurt him).   However, the Dems will have a stronger candidate (almost impossible not to) in a less GOP friendly environment than 2010 in a Presidential year.