2010 Best/Super Moe Tournament

[CrowKenobi]

No comments on random god this time:

10: Mio and Kanu
11: Hanekawa and Zetsubou Sensei's sister
12: Hinata and Kyoko

[Umiko]

10) Touko Aozaki & Mio Akiyama
11) Hatsune Otonashi & Su
12) Masami Iwasawa

[hinakatbklyn]

Got all my main choices in. Mio had more votes than both the 2nd and 3rd place finishers in her group even with the increase in votes. Hanekawa made it close to not getting in though.

I really don't see anyone here making an impact in the main rounds even with some favorites to vote for:

13 - Yuri Nakamura (Angel Beats) / Hiiragi (Hanamaru Youchien)
14 - Illyasviel von Einzbern (Fate/Stay Night Movie) / Konoka Konoe (Negima Another World)
15 - Sohara Mitsuki (Sora no Otoshimono) / Cecily Campbell (Sacred Blacksmith)

[Zeroryoko1974]

13- Yurippe and Hiiragi
14- Konoka, Illiya
15- Cecily Campbell and uhhhh Yuki Morikawa

[wontaek]

At work, so sorry for short post which actually is a request for something I can't do now.

It would be interesting if someone can show who are the top 10 in the approximate male votes and top 10 in approximate female votes in the 2nd round so far. Male/female votes can be approximated by looking at total number of the votes for each characters and then checking the ratio of male to female voters in the View Voting Pattern Analysis Graph . This is rather important to prove or disprove some claims that has agitated many people. Thanks in advance.

[CrowKenobi]

13: Sawako and Hiiragi
14: Lizlet and Lucy
15: Cecily and Alicia

[Kotohono]

Votes this round:
13. Yuri Nakamaru, Kanae Naruko
14. Shizuku Sango, Lizlet L Chelsie
15. Noel Kannagi, Natsuki Mizuhashi

[Team Rocket Elite]

Originally Posted by wontaek

At work, so sorry for short post which actually is a request for something I can't do now.

It would be interesting if someone can show who are the top 10 in the approximate male votes and top 10 in approximate female votes in the 2nd round so far. Male/female votes can be approximated by looking at total number of the votes for each characters and then checking the ratio of male to female voters in the View Voting Pattern Analysis Graph . This is rather important to prove or disprove some claims that has agitated many people. Thanks in advance.

The graphs seem to only give a very rough approximation. The graphs are a bit hard to read so I can't be sure I punched in all of the numbers correct. I'm not even sure I'm reading the graph properly. >_> I assumed that if it was up to level 7 for male and 3 for female, 70% of the votes were male. I'm not sure why some of the graphs go up to level 10 while others go up to level 6. If my assumption on how the graph works is wrong, please tell me so I can fix the results. I'm not sure how much you can get from this due to all the problems but if you're willing to put up with them here you go:

Results up to Group 12. I'm not sure if I'll update this with later matches or not.

Male
1 - Mio Akiyama - 429.38
2 - Holo - 359.38
3 - Yui Hirasawa - 325.54
4 - Mikoto Misaka - 322.67
5 - Yui - 269.82
6 - Tsumugi Kotobuki - 253.09
7 - Kazari Uiharu - 243.09
8 - Ritsu Tainaka - 232
9 - Tsubasa Hanekawa - 231
10 - Konjiki no Yami - 222.6

Female
1 - Anri Sonohara - 322.7
2 - Celty Sturluson - 317.82
3 - Winry Rockbell - 269.82
4 - Sango - 259.33
5 - Mio Akiyama - 257.63
6 - Rima Mashiro - 252.67
7 - Ritsu Tainaka - 232
8 - Shinobu Oshino - 227.77
9 - Beatrice - 222.6
10 - Holo - 215.63

Overall
1 - Mio Akiyama - 687
2 - Holo - 575
3 - Yui Hirasawa - 529
4 - Mikoto Misaka - 484
5 - Ritsu Tainaka - 464
6 - Tsumugi Kotobuki - 464
7 - Anri Sonohara - 461
8 - Celty Sturluson - 437
9 - Winry Rockbell - 424
10 - Yui - 424


Edit: And I guess while I'm here:

Votes:
Irie
Nakamura Yuri

Ryuumonbuchi Touka
Konoe Konoka

Alicia Testarossa
Kannagi Noel

[imbehindyou]

Originally Posted by Team Rocket Elite

The graphs seem to only give a very rough approximation. The graphs are a bit hard to read so I can't be sure I punched in all of the numbers correct. I'm not even sure I'm reading the graph properly. >_> I assumed that if it was up to level 7 for male and 3 for female, 70% of the votes were male. I'm not sure why some of the graphs go up to level 10 while others go up to level 6. If my assumption on how the graph works is wrong, please tell me so I can fix the results. I'm not sure how much you can get from this due to all the problems but if you're willing to put up with them here you go:

It's approximately 5% for each lvl.

I'm going to use Mio for example.
She got lvl 10 for male voters, that means 50% of male voters voted for her (or it means that 50% total male votes went to her; since you can vote 2 characters/group I guess it gets complicated). It's much easier to calculate when it's 1 vs 1 (or 1 vs 1 vs 1)

But it only counts those who answered the survey (~91%) so it's not 100% accurate but it should be fairly close

[wontaek]

Originally Posted by Team Rocket Elite

The graphs seem to only give a very rough approximation. The graphs are a bit hard to read so I can't be sure I punched in all of the numbers correct. I'm not even sure I'm reading the graph properly. >_> I assumed that if it was up to level 7 for male and 3 for female, 70% of the votes were male. I'm not sure why some of the graphs go up to level 10 while others go up to level 6. If my assumption on how the graph works is wrong, please tell me so I can fix the results. I'm not sure how much you can get from this due to all the problems but if you're willing to put up with them here you go:

Results up to Group 12. I'm not sure if I'll update this with later matches or not.

Male
1 - Mio Akiyama - 429.38
2 - Holo - 359.38
3 - Yui Hirasawa - 325.54
4 - Mikoto Misaka - 322.67
5 - Yui - 269.82
6 - Tsumugi Kotobuki - 253.09
7 - Kazari Uiharu - 243.09
8 - Ritsu Tainaka - 232
9 - Tsubasa Hanekawa - 231
10 - Konjiki no Yami - 222.6

Female
1 - Anri Sonohara - 322.7
2 - Celty Sturluson - 317.82
3 - Winry Rockbell - 269.82
4 - Sango - 259.33
5 - Mio Akiyama - 257.63
6 - Rima Mashiro - 252.67
7 - Ritsu Tainaka - 232
8 - Shinobu Oshino - 227.77
9 - Beatrice - 222.6
10 - Holo - 215.63

Overall
1 - Mio Akiyama - 687
2 - Holo - 575
3 - Yui Hirasawa - 529
4 - Mikoto Misaka - 484
5 - Ritsu Tainaka - 464
6 - Tsumugi Kotobuki - 464
7 - Anri Sonohara - 461
8 - Celty Sturluson - 437
9 - Winry Rockbell - 424
10 - Yui - 424


Edit: And I guess while I'm here:

Votes:
Irie
Nakamura Yuri

Ryuumonbuchi Touka
Konoe Konoka

Alicia Testarossa
Kannagi Noel

Thank you very much. You have given me the result I hoped for. I was hoping that correlation between overall ranking and male ranking is significantly higher than correlation between overall ranking and female ranking without doing proc corr with SAS since only a few would know the TRUE implication of that result. Just looking at the top 10, despite that Female voters have consistently outnumbered male voters, it still is male voters' will that dominates the overall picture. I hope someone will post TRE's results in http://anizone.net/?mid=Moet_Inter_Forum because I am unable to do so at work( Browser and other company internet policy issues ), but someone needs to put out some fires quickly.

[Rajura]

WOW!!!!!!!

Maybe... just maybe... (fill in the blank; you where I'm going with this... hahahahaha)!

[NissanNut]

Mio moving on to next round!
Today it's Yurippe, Shizuka and Sohara!

Spoiler for Prelims 2 13-15:

13 - Nakamura Yuri, Kuronuma Sawako
14 - Shizuka Sango, Kokutou Azaka
15 - Kannagi Noel, Mitsuki Sohara

[Team Rocket Elite]

Based on what imbehindyou said, the results I posted aren't correct so they shouldn't be used.

I'm trying to redo that stats but I need a bit more information. When the site lists the total votes as "Total : 2341(1243)". Is the "1243" the number of unique voters in that match and "2341" the total number of votes(since you can vote twice)?

[imbehindyou]

Originally Posted by Team Rocket Elite

Based on what imbehindyou said, the results I posted aren't correct so they shouldn't be used.

I'm trying to redo that stats but I need a bit more information. When the site lists the total votes as "Total : 2341(1243)". Is the "1243" the number of unique voters in that match and "2341" the total number of votes(since you can vote twice)?

yup.

anyway votes
13: Kuronuma Sawako, Nakamura Yuri
14: Ryuumonbuchi Touka, Konoe Konoka
15: Morikawa Yuki, Amano Tooko

[Team Rocket Elite]

Using Group 10 as an example.

Mio got 687 votes and there are 1243 unique voters.
91% answered the Survey.
44% Male
56% Female
Mio has Lv10 Male. (I'll round this up to 55%)
Mio has Lv6 Female. (I'll round this up to 35%)

Male vote total = (0.91)(0.44)(0.55)(1243) = 273.7 -> 274
Female vote total = (0.91)(0.56)(0.35)(1243) = 221.7 -> 222

Let's say everyone who didn't answer the survey voted Mio.
No gender given = (0.09)(1243) = 111.9 -> 112

274 + 222 + 112 = 608

Even with optimistic assumtions, there are still a lot of unaccounted for Mio votes. Did I make a mistake somewhere?

[Ithekro]

Could it be that the 91% counts even if one did not answer the gender question? A partial answer only, thus the gender numbers don't fully reflect the 91% that answered the survey.

[Team Rocket Elite]

It's possible but it you still require pretty extreme results to make it work. Even if only 79% of voters answer the gender question, it would still require, nearly 100% of the voters who didn't answer the gender question to vote Mio. You have to go down to 56% of voters answering the question before you only need 70% of the non-gender question voter to vote Mio.

[Ash Falls Town]

Well that goes Koihime Musou's hope
(6-2)

13. Hiiragi, Abstain
14. Lucy, Yamamoto
Damn this group is hard. I could have voted for 5 others.
15. Kaori, Noel
I'm sure it's a good tank.

[wontaek]

Originally Posted by Team Rocket Elite

It's possible but it you still require pretty extreme results to make it work. Even if only 79% of voters answer the gender question, it would still require, nearly 100% of the voters who didn't answer the gender question to vote Mio. You have to go down to 56% of voters answering the question before you only need 70% of the non-gender question voter to vote Mio.

TRE, your 1st numbers are just fine.

Spoiler for If you really, really want to be precise:

If you really want to be precise, the 1st estimate TRE gave us would be the initial guess and you can calculate the variance by the binomial distribution. http://en.wikipedia.org/wiki/Binomial_distribution . Since we have 91% of the voters answering, the variance would be small because the n in this case would actually correspond to the 9% of the total votes which would be about 59 for Akiyama Mio. So the 99% confidence interval for Akiyama Mio's male vote would be ( 687 * (10 / 10 + 6 ) = 429.375 ) +- 3 * sqrt ( 687 * 0.09 * 0.625 * 0.375 = 14.4914 ) = 429.375 +- 3 * 3.8 = 429.375 +- 11.4 => [410 , 441] in nearest integer.

[Umiko]

13) Yuri Nakamura & Sawako Kuronuma
14) Illyasviel von Einzbern & Azaka Kokutou
15) Sohara Mitsuki & Noël Kannagi

Had to forgo Hiiragi in Group 13. >.<