LogFAQs > #915787854

Finally, it must be noted that this particular result is not an upset. Yes, the prediction percentage is only 23.29% which is too low to even possibly make her the favorite this early in the contest, but that's residual upset effect. KOS-MOS's Round 1 prediction percentage was only 41.48%, which means that 56.16% of brackets that had her in this match had her winning. Not a great percentage, and well within the range where it's possible that brackets that had this particular match-up favored Aqua slightly since she herself fell a little short of 2/3 prediction in Round 1, but it's definitely a decent amount, and while I don't believe that "bracket votes" make a huge difference, it can't be denied that Ellie outperformed Aqua against KOS-MOS and that this effect was even more pronounced among registered voters, which might suggest that the plurality of casuals had Ellie winning this match. (And yes, "voting your bracket" when you've already lost your pick for a match does often mean voting for the one that broke it, especially given those prediction percentages. Pre-match, it would be entirely likely that an Ellie bracket would see that Aqua was a bigger favorite than Ellie in their Round 1 matches and assume that more brackets would be busted by KOS-MOS > Aqua than by Aqua > KOS-MOS--which isn't necessarily false, because even if Ellie > Aqua is the most common pick, if the second-most common is Aqua > Ellie rather than KOS-MOS > Aqua, Aqua could have the second-most brackets anyway.)
---

Also known as Cyberchao X.