You simply can't ignore the 125 votes of "leave everything as it is". Subtract 125 from all the buff/nerfs of each option and it's a much more telling picture. Then you have to account for the people who voted to buff/nerf a class simply because it's their class or the counter to their class. Cosmos' graphs are nice, but they don't take either of these two things into account.
You're being silly, for two reasons.
a) 125 votes for no change, yes, but out of 973. I don't know about you, but to me that suggests most people still want to see something changed. Now, we have how many poll options? 19. So what you're trying to do is to count each of those "no change" votes no less than 18 times. The correct number to subtract would be seven votes from each other option. Hardly much of an issue, and honestly a pretty pointless thing to do in the first place.
b) You don't have access to that data, so it's useless to speculate about. Best thing to do is assume it's more or less equal for all cases and look at the balance of nerf/buff votes, like Cosmos did -- at least while we wait for what Vargas has to say.