I'll dispense with the full Bayesian treatment here, but what we have here is a case of the base rate fallacy (although not the form that is usually discussed): failing to account for the prior probability of the evidence. That is, we have some evidence, which is that the anecdotes we see are largely in favor of standing desks being helpful, and we need to account for how likely that evidence is regardless of whether standing desks are helpful.
Unfortunately, as our friends with the magnetic bracelets demonstrate, that probability turns out to be very high. If something works, we see overwhelmingly positive anecdotes. If it doesn't work, we see overwhelmingly positive anecdotes.
Why might this be? Well, for one thing, there is sensitivity to placebo effects, because have purely qualitative phenomena that are being reported only subjectively.
But even stronger is a selection bias that has to do with people's behavior in reporting and reading. Someone is unlikely to write about their experiences publicly who tries a standing desk for a while, finds no significant benefit, and goes back to a sitting desk. She won't find it interesting to talk about, and she may also be self-conscious that she didn't do it right, and would be inviting criticism. After all "I tried X and it didn't work" is, on some level, an admission of personal failure, and people aren't particularly prone to those. And if she does write it up, her write-up isn't likely to get all that much attention because "X didn't work for me" isn't a particularly interesting story.
But one who finds the standing desk to be a miracle cure for all of her problems is relatively likely to write about it, and people will take notice, because "making a simple lifestyle change improved my quality of life" is an attention getter.