Okay guys, out of curiosity I've analyzed BluePikmin's poll again. Took a while to count it all up, analyze it, and decide to write this long post, so my data is actually a couple hours old, heh. I actually counted both total votes of "5" and the percentage of the character's total votes that were 5 votes. That way I could get a feel for who was leading in votes but also determine trends for those categories that had less total votes anyways. For instance, the part of the poll with characters from already repped franchises had 1608 votes when I analyzed it, whereas the third party category only had 1058 total votes. So although Krystal has more total votes than Banjo, we can see that Banjo is trending better than Krystal even with less total votes due to his percentage lead. Percentage allows you to extrapolate some while totals are just hard data regarding what's been voted. Make sense? Well, anyways, here's my data. Top 30s...
Total "5" Votes
1. King K. Rool --- 678 votes
2. Ice Climbers --- 619 votes
3. Wolf --- 612 votes
4. Inkling --- 607 votes
5. Isaac --- 555 votes
6. Snake --- 505 votes
7. Krystal --- 468 votes
8. Dixie Kong --- 433 votes
9. Banjo & Kazooie --- 431 votes
10. Bandana Dee --- 410 votes
11. (Impa) & (Bomberman) --- 400 votes (tied)
12. Shovel Knight --- 356 votes
13. Tetra --- 346 votes
14. Paper Mario --- 344 votes
15. (New Mii Class) & (Ridley) --- 330 votes (tied)
16. Chibi Robo --- 328 votes
17. Shantae --- 320 votes
18. Phoenix Wright --- 310 votes
19. Sceptile --- 305 votes
20. Waluigi --- 302 votes
21. Squirtle --- 296 votes
22. Professor Layton --- 293 votes
23. Simon Belmont --- 288 votes
24. (Wonder Red) & (Rayman) --- 287 votes (tied)
25. Ivysaur --- 277 votes
26. Black Shadow --- 267 votes
27. Hades --- 261 votes
28. Bayonetta --- 256 votes
29. Captain Toad --- 254 votes
30. (Classic Ganon) & (Takamaru) --- 248 votes (tied)
...
Percentage of "5" Votes
1. Ice Climbers --- 54.5%
2. Wolf --- 53.9%
3. Inkling --- 53.2%
4. Isaac --- 48.6%
5. Snake --- 47.7%
6. King K. Rool --- 42.2%
7. Banjo & Kazooie --- 40.7%
8. Bomberman --- 38.2%
9. Shovel Knight --- 33.6%
10. Shantae --- 30.2%
11. Phoenix Wright --- 29.3%
12. Krystal --- 29.1%
13. Ridley --- 29%
14. New Mii Class --- 28.9%
15. Chibi Robo --- 28.7%
16. Professor Layton --- 27.7%
17. Simon Belmont --- 27.2%
18. Rayman --- 27.1%
19. Dixie Kong --- 26.9%
20. Waluigi --- 26.6%
21. Squirtle --- 26.1%
22. Bandana Dee --- 25.5%
23. Wonder Red --- 25.1%
24. Impa --- 24.9%
25. Ivysaur --- 24.4%
26. Bayonetta --- 24.2%
27. Amaterasu --- 22.9%
28. Takamaru --- 21.8%
29. Tetra --- 21.5%
30. Paper Mario ---21.4%
At the time of my data gathering, the other two major catagories I haven't referenced yet - unrepped Nintendo franchises and "deconfirmed characters" - were both around 1140 total votes, so roughly a 100 more votes than the third party category but still nearly 500 less than the first category. This means that generally, the two categories with similar total votes around 1140, ended up not moving much relative to themselves between the total count and percentage. They did make up some pretty big ground on the first category, though the third party category made up even more signifcant ground from total to percentage. You might also say that the first category lost ground to the others in particular the third party category. For instance, King K. Rool was far and away the leader in 5 votes, but still had a smaller percentage of 5 votes compared to ICs and Wolf in spite of his 5 vote lead.
Now, I think the percentages allow us to extrapolate some, but not really too far. We can't assume that if Snake had 1608 people rate him like K. Rool did, that he'd have more total votes than K. Rool. I also think that the difference in total votes for each category is a result of some people only voting in the categories that mean anything to them. If they support Snake, they'll seek out the category with Snake in it to make sure he gets a five, for instance. Also, the generally high number of votes in the first category may indicate that more people are interested in characters in that category or it may mean that the first category is just the first one they see and some voters get weeded out with subsequent categories.... or both of course, which is the likely scenario.
So if you're wondering, "do the totals mean more or the percentages?", my personal feeling is that the totals are more indicative of total support. I think that both sets of data have something to offer though, only that the totals should be weighted more if we're to combine both sets of data. Not sure how that would happen. I already explained that the percentages allow some extrapolation and allow categories with different vote totals to be compared, but we can only predict so far like I referenced in the Snake comparison in the last paragraph. Also, the point that character supporters would seek out the category with their character in it supports the idea that the percentages may not convert to extrapolated totals. So although the first category has more votes than the third party category it could also point to there being just less fans of third party characters. The total votes have their own issues what with each category having a different number though. There's obviously some balance between the data sets, but I imagine there will always be fair sized margin of error, and that's just in analyzing what we have. Of course, there's a whole ballot of votes out there that this poll is struggling to make sense of. For all we know, there's 1000 more votes for Paper Mario out there, or maybe there's 10. The percentages don't tell us, only the trend so far. It's not my place to say, "This doesn't look right, there's no way Waluigi has more supporters than Rayman" or "There's no way that K. Rool has less support than the ICs even if the percent tells me he might"... so I'm not going to try and balance the data sets until it looks "right".
Oh, and thanks to @
BluePikmin11
for making this poll and Source Gaming for letting him make it more easily available to character supporters everywhere! I know I just analyzed and talked about this poll like I own it, but I certainly do not. I only find it very interesting, even if really when you think about it, it may not be actually telling us much about the ballot. You got to work with what you can get when it comes to these things.
)
