I forgot 1.8 million votes were cast. Makes the 40k on that google doc seem pretty insignificant...
I thought I'd comment on this as it seems to be a high point of contention right now. Basically this is the source of everyone's nervousness, because they see the numbers and it says good things about Isaac, but they aren't sure about the data they're seeing. Just because you have 40k votes, which pales in comparison to the 1.8 million that we got, it doesn't mean we can't make some insights into the data here.
In fact, it's sometimes much worse to look at larger and larger sample sizes because the resources necessary to collect the data overflow your capability to sift through it. The smaller number doesn't actually matter that much
if it's a representative sample. If you look in the tab, you see reporting from multiple polls taken from Smashboards, Reddit, Straw Poll, Tumblr and Poll Maker. Isaac really doesn't fall outside of the top 5. There's only one instance in the Tumblr poll where that's the case. Like I have commented on before, you really don't have to worry too much about unique voting polls for Isaac. These are basically 12 separate polls pooled together (For first party) and averaged out for characters. While Isaac is not #1 in even most of them, he still averaged a 2nd place and above Banjo.
The real question is, who was really voting in the character ballot? It's an interesting question. One can assume that it certainly includes the people that voted in these separate polls; effectively avid smash fans. Were there a lot of people voting that barely have contact with smash? Only told to vote by others? Play smash casually or very rarely? I get a sense that extrapolation would really not change a whole lot. Isaac may not have been #2 or #3, but would it fluctuate so wildly that he would end up out of the top ten? I think that's very unlikely.
Especially when you have the law of averages on your side. Lets say this was a decathlon; you have to do well in ten events to get first place. Paper Mario, man he's so good a throwing a javelin, but he can't run faster than your grandma in the 100meter dash. Magolor, that guy can throw the shot put and run short distance, but he really can't do the long jump. Isaac... well Isaac basically gets 3rd place every time, in every event. At the end of the day that's basically all you need in the decathlon. Isaac, as far as I've followed him in smash speculation, has been steady eddy in polls, popularity and fan desire. Sometimes it cools off, but it's pretty persistent, unlike many other characters who pop in and out of existence and try to make believe that they've always been that way. That's why the exit polling speaks to me, I really don't think it's far from the truth that Sakurai is seeing.
At the end of the day, yeah, some dudebros are gonna vote for Goku and Shrek on their only vote. That really doesn't amount to much, even if you add on top of that 5,000 more dudebros. With characters like K. Rool, Ridley and Isaac, there's a reason their popularity hasn't waned. A lot more people want them in Smash or back in video games in general than you think. Goku be damned to think otherwise.