I don't understand why we're still having polls when:
a) The ballot is over
b) Most people voted 5-6 months ago
We know who the popular characters are. Occasionally a fad character shows up, but generally speaking, the results always stay the same. Even my fellow K. Rool fans are guilty of this. They fixate on these polls constantly. All they do is give people a false sense of security.
.
The ballot doesn't have to be running for these polls to be useful. In fact, regarding the question of identifying "fad" characters (i.e. those with fleeting popularity), the more polls the better. With these polls it is possible, for example, to point to K. Rool as a character with enduring popularity and Shovel Knight as a character whose popularity is comparatively fleeting. (Note that "fad" characters might still acquire a great number of ballot votes, even if their support is more time-dependent).
Does time introduce some bias into these polls? Probably. Consider, for example, some new revelation by Sakurai that he does not like portly characters or characters from games released this year. This might make people who voted for characters like King K. Rool and Inkling wary of voting for these characters in fan polls; they might choose another character they voted for (or supported) instead. In their minds, Sakurai's comments might "disqualify" votes for Inkling and K. Rool. As time passes, new information might arise that might affect how people express their choices. While biases like this one and others exist, I think these polls (and this poll, in particular) is very useful, even if just for its novelty among other fan polls.
K Rool seems to have both kinds of fans, very vocal groups that want him in and very vocal groups who don't.
That is a very facile way of describing the data from the poll. K. Rool might have a higher percentage of people who disapprove of him (1s and 2s) than Inkling and a few others, but those in his favor (4s and 5s) are in the majority (approximately 60%). This is better than the vast majority of characters and not too far away from Inkling's and Wolf's own approval percentage (approximately 69% and 70%, respectively). Your wording implies more equality between those who actively approve of K. Rool and those who actively disapprove of the character. That does not seem to be the case at all.