Group Finals

Sums of votes and percentages of votes received by surviving characters in the first three rounds:
The third round largely went as expected, but there were a few interesting developments. The most significant is Misaka's enormous vote total in her victory over Victorique. That's a match where nobody had any reason to tactically vote for Misaka, and she still pulled over 600. As the chart shows, Misaka is completely dominating votes received by an absurd margin. Quite simply, if you aren't a Biribiri fan, you have to consider her public enemy #1. (Update: After the cheating scandal came to light, revealing many of her votes to be illegitimate, she really did turn out to be public enemy #1.)

Predictions for the group finals are pretty straightforward. In each case the predicted winner has a clearly better record than her opponent, except Kuroneko vs Houki in total votes received, where Kuroneko's power is being hidden by her easy matches in rounds 2 and 3. (Update: Well, they looked straightforward at the time. Saimoe never fails to surprise.)


Group A
Prediction: Misaka got over 600 against Victorique, and there's no reason she can't do it again. Consider her to be an unstoppable juggernaut until vote totals start inflating in the last few matches of the tournament.

Update: Astarotte pulled close in the early stages of the match, but that's likely to be the result of fake votes. Biribiri has started escaping and has been leading throughout.
Results: This one turned out to be a shocker. In an electrifying turn of events, Misaka lost 446 to 327. There have been reports that large-scale cheating was responsible for her earlier 600+ vote totals. The tournament organizers tightened restrictions just before this match, and lo and behold, Misaka's power suffered a blackout, and was reduced to the low 300's, about the same as what Index has been receiving. Presumably, some of Misaka's less diehard supporters bailed on her after hearing of the scandal, but the change is so drastic (300 votes) that you honestly can't call Misaka's 100 vote victory over Victorique legitimate. Her 111 vote victory over Homura... 100 votes again? Let's look back at those charts:
Here's the data mining breakdowns for each day:
http://tcode.sakura.ne.jp/anime11/mining0816.html
http://tcode.sakura.ne.jp/anime11/mining0922.html
http://tcode.sakura.ne.jp/anime11/mining1009.html
http://tcode.sakura.ne.jp/anime11/mining1018.html
Among the codes missing from the 10/18 list are U and v, which favored Misaka by 42-9, 27-9, 28-3, and 24-7.

It looks like most of the crazy power difference between Misaka and the other major Index faction characters was, in fact, due to cheating. Victorique and Homura fans were beaten unfairly, although it's somewhat poetic that Homura sacrificed herself to expose this danger before it could threaten Madoka. Is this how anime characters feel when the script writer makes things turn out in a certain way?

Prediction Record: 0/1, 0/1 R4, 111/145 Total

Group B
Prediction: Tenshi should finish off another Strike Witch. SW has been performing very well this tournament, but they don't have the manpower to take on the elite and failed to make much progress against Tenshi with Gertrud. Even Erica's Biribiri-match-inflated 498 is less than Tenshi's round 1 score of 527.

Update: Look at Tenshi's curve flatlining. Just look at that. What's going on around here? This could turn into an upset in a hurry, unless technical difficulties are blocking voting and Erica's progress is fake. If it turns out that the cheater was behind Tenshi too, all of us who were calling this a boring, predictable round will look sillier than girls flying combat missions in their panties.

Results: Erica wins 377 to 323. Tenshi was caught using Guard Skill: Cheating, and the new bans slapper her back down into purgatory. Erica wins a glorious victory for the Strike Witches.

There is a report that the code generator was broken for PCs for an extended period of time, and only phones could get codes. There was some debate about whether this requires the match to be replayed under the 11 hour rule, but the admins let it stand. These technical difficulties explain the oddities in the voting curves.

Prediction Record: 0/1, 0/2 R4, 111/146 Total

Group C
Prediction: Madoka hasn't been tested yet. We may see more than a hint of her power if Elsie is backed for an assassination attempt, but this TWGOK character with a high score of 315, middle of the pack percentages, and a medium sized faction is simply at too much of a disadvantage against the girl I'm predicting to win it all.
Results: Madoka wins with 64.14% of the vote. That's despite the fact that Elsie is a high-quality character who belongs in the top 10. The match was pretty quiet, without heavy faking or shenanigans. Prediction Record: 1/1, 1/3 R4, 112/147 Total

Group D
Prediction: Ika Musume's invasion is going just as planned. Being non-human, she does not possess a nutbladder, and is therefore immune to Yuno's special ability. Although Yuno is one of the best performing veterans in Saimoe, isn't Ika the strongest?

Results: Ika Musume's tentacles reach wider as well as deeper. However, her vote total was on the low side. She did have a safe victory, so there was no need for any more, but she'll need to be much more high-powered in the future. Prediction Record: 1/1, 2/4 R4, 113/148 Total.

Group E
Prediction: Sayaka outclasses Haqua. Haqua, whose popularity is driven mainly by the manga, is weaker than Elsie, and can't be expected to put up more of a fight than Charlotte did.

Results: Sayaka cut Haqua down with over 60% of the vote. Fake votes clouded the match and made it look like Haqua had the lead with a few hours left, but Sayaka went berserk and Haqua's fake votes vanished like the barrier of a defeated witch. Graphs currently unavailable.

Prediction Record: 1/1, 3/5 R4, 114/149 Total.



Group F
Prediction: Mami's vote totals are as big as her mammies, and Shirayuki may be weakened by Aria's spanking. Shirayuki is biting off more than she can chew, so expect Mami to head to the finals.
Results: Mami is at the head of the pack after another victory. She won with just over 70% of the vote, which is absurd in a group final. This happened in a match with almost no fakes, so Mami seems to be flaunting her Tiro Finale of a last minute rush to impress the other magical girls again. Prediction Record: 1/1, 4/6 R4, 115/150 Total.
Group G
Prediction: Kyouko has been ferocious, and Index has been winning with low numbers. Although Index did take down the dreaded Kirino, that had more to do with Kirino's pathetic, humiliating 284 votes than Index's 317. Since Index comes from the same show as the disgraced Misaka, she won't be an effective assassination tool. Kyouko's story took a more moe direction in the end, so she'll win the vote-eating contest between these two gluttons.

Results: Kyouko wins 428 to 250, as expected. The Madoka Magica winning streak extends to a record 16 matches. Prediction Record: 1/1, 5/7 R4, 116/151 Total.

Group H
Prediction: Kuroneko. Houki is at a disadvantage, and will need outside help to get past one of the toughest opponents in the tournament. There won't be much pressure to assassinate Kuroneko with the four Madoka girls and Biribiri drawing all the attention.

Update: Houki's curve is getting pretty strange. Analysis suggests that Houki has far more fake votes than Kuroneko. At this rate, I expect to see Kuroneko in the quarterfinals.

Results: Kuroneko wins. Houki was indeed receiving heavy faking throughout the match. Prediction Record: 1/1, 6/8 R4, 117/152 Total.

No comments:

Post a Comment