Which is more likely?

[MT Treasurer]

Even if Marshall wins the R nomination for Senate, Bollier winning will be more likely than Gardner getting reelected.

[MT Treasurer]

Is Kansas somehow competitive or something?

Bollier is probably their strongest recruit of this cycle and the state isn’t as solidly Republican as Atlas makes it out to be, so yeah (although Kobach is obviously more likely to lose the general election than Marshall and especially Wagle).

[MT Treasurer]

You've gotta be kidding me Atlas Forum.

All of 1 OP & 9 "Gardner" poll voters = The entire contingent of Atlas Forum

No, but only on this forum is this kind of question asked and the way more strongly partisan state is chosen to be more likely to flip because of trends and good/bad candidate analysis.

Because as we all know, the presidential election results of one year are all that matters, and blue states have been more than willing to send Republicans to the Senate in the past decade.

ND was way more partisan than NM in 2012 and both states had "strong challengers." Why did Democrats win ND and Republicans lose NM, then?

[MT Treasurer]

2012 is basically another era. Even if I agree with you that republicans are unfortunately more likely to fall for the ''moderate democrat'' trap you have to agree that the number of red state democrats in the Senate has dropped a lot over the past eight years.

2012 was another era? Let’s see...

2014: KS-SEN competitive, GA-SEN competitive, KY-SEN somewhat competitive, Republicans barely win Alaska, South Dakota isn’t Safe R from day one (mostly because of Pressler)  
2016: Republicans barely hold Missouri (this one could have gone either way, which is embarrassing), IN-SEN competitive (it’s easy to forget that Bayh was ahead by double digits when he entered the race)
2018: TN-SEN competitive (Bredesen was ahead when he entered the race); Democrats win a Senate race in a Trump +42 state which should have been an easy pick-up for Republicans; Democrats comfortably hold a Senate seat in a Trump +20 state (guess Rosendale having a Baltimore accent was just too much for the experts at the NRSC); Heitkamp/Donnelly/McCaskill all lose but still do way better than their state's partisan lean would suggest in spite of their unpopularity (Heitkamp did 21 points better than Obama in 2012, she did 25 points better than Clinton in 2018!)

But sure, keep assuming that every Trump state is guaranteed to vote for a Republican Senator in 2020. There’s really no point in arguing with you because we’re never going to agree anyway, so let’s just leave it at that.