On a better team than Barkov and a -11 (2nd last on team). Offensively much worse than him( Yes, offense has always been a factor in Selke too). Backlund also has 74PIM - this alone should drop him from the Selke conversation. A guy who puts his team in the box that much has no reason to be there. Ooh and he is a career <50% at FO. This year at 48,79%. That is nowhere near selke-caliber.
Thank you for a list of statistics which are in some cases moderately appropriate and in others pretty irrelevant to the conversation.
Now let's list some Backlund numbers that make his case. He is second in ice time among Calgary forwards (behind Gaudreau) despite lacking strong offensive value. His value comes from defence. He has the best "raw Corsi" among all Calgary forwards. 56.27% corsi for while playing with 44.23% defensive zone starts. He plays defensive situations and nevertheless the puck is going in his direction. That is statistically about as well as we can show defence. He plays more penalty killing time than any of the other candidates listed in this thread. Then of course there is the eyeball test.
You can stick to +/- ratings and the statistical uncertainty that comes from the limited sample size if you wish. With the larger sample size Corsi provides, the story opposes the +/- conclusions. What else do you quote as stats? Penalty minutes? That's a stretch. Faceoffs. Everybody in the NHL is 50 +/- epsilon if they take a bunch of faceoffs and as we see from the Corsi numbers this doesn't prevent the puck from going in Calgary's direction when Backlund is on the ice. Faceoffs are largely irrelevant to the Selke (or should be) as some past winners rarely took them. Then of course you ask for offence. Offence is clearly not defence. Largely, your problem seems to be that just because a number can be measured you want to use it in your analysis regardless of its relevance.
That isn't to say that I expect voters to pick Backlund for the award. I don't. But I see him as the front runner.