Not sure I've ever followed the polling on losing presidential candidates two years after the election. I mean, what is the median here? Do we have comparative figures for all the other losing candidates for the same time period after their loss?
If it is indeed the lowest (which the poster doesn't support with any stats), I can think of a several reasons:
1. Polling trends in the past decades have shown more Americans disenchanted with politics generally and more convinced that politicians are corrupt, a sentiment that extends across party lines. Any hint of corruption, proven or unproven, makes a pile-on easy.
2. Many voters simply vote their pocketbook, so if they like (with their usual myopia) the current tax break that they believe they're getting, then that will tend to translate into a lower opinion of the past rival candidate.
3. We don't know how the question of approval was framed exactly? Does the wording naturally tend to favor the winner?
4. The winner of this most recent election has been notable for enlisting his supporters in a continuing denigration of Hillary, as part of an intensely partisan platform that is constructed on what people dislike (Muslims, immigrants, LIBERALS, LGBTQ, etc.) more than what they hope for. So she remains a pinata and a boogeyman, even though she is currently not politically active. Compare this to Obama, or any other president in the past. How often did Obama rant about locking up JOhn McCain or Mitt Romney or otherwise use him as a whipping boy?
5. A large contingent of Millenials remain who dislike Clinton for what they see as shabby dealing toward Bernie, their chosen candidate. She remains as a symbol for them of something they are mad at, just as she does for the Right.