The Democrats seem bound and determined to nominate a candidate who will guarantee high turnout for the GOP and who has little support with independents. The GOP -- once the Trump bubble bursts, as it will -- will probably nominate someone quite a bit younger and more charismatic than Clinton, and will spend a few years in the wilderness with the White House, the Senate, the House, and the majority of states under Republican control -- and no one to blame for it but themselves.
The Q-Poll findings: "Sanders does just as well [as Clinton against Rubio], or even better, against [the other] top Republicans [Trump, Carson,and Cruz]." Against each of the latter three, Sanders' winning margin exceeds Clinton's by an additional 2%, 3% and 5% respectively, compared to a survey margin of error of +/- 2.6%.
It is by just such narrow margins that modern elections are won or lost. For example Slate opines it to be "a sign of how accustomed we've become to razor-thin margins of victory that Obama's 2.3-percent popular-vote victory [in 2012] seems almost like a rout...[T]hree out of four of our last elections have been decided by a popular-vote margin of less than 3 percent" which, the author observes, "best resembles the Gilded Age" when choice was similarly limited to pluto-Dum and pluto-Dem candidates.
Sanders' additional margin of safety places him beyond the margin of polling error around which Clinton's fluctuating numbers for her Republican match-ups are more commonly found. Sanders' numbers also seem "almost like a rout" compared to a toss-up for Clinton.
Democratic voters therefore have it exactly backwards. It is Sanders that should be attracting their near certain (87%) confidence of victory next November now accorded Clinton, while their doubts about a Sanders toss-up (49%) should properly attach to Clinton. Democrats are not just misinformed, but grossly misinformed, about the key issue of whether Clinton or Sanders will more likely win against Republicans.