2012-03-01, 05:02 | Link #761 | |||
Senior Member
Join Date: Jul 2008
|
Quote:
Quote:
Quote:
One example of two characters with very different personalities from the same series splitting votes despite voters getting two votes is the Stella 12 match from Phase 1 Day 1. Ayanami Rei and Soryu Asuka Langley split votes and ended up below Isayama Yomi and only a bit higher than Kagura. Rei and Asuka aren't exactly powerhouses but this was still a fair sized drop from their normal level of strength and their performances in the later matches.
__________________
|
|||
2012-03-01, 05:23 | Link #763 | |
Senior Member
Join Date: Oct 2006
Location: Planet Earth
Age: 54
|
Quote:
1. Since number of allowed choices were 2 ( this is very important ), going from group size of 7 to 9 doesn't make too much of difference, yet. If the group size went to 10, then yes, that would be too much. 2. Bake/Nise characters who narrowly missed the cut shouldn't worry too much. Although number of things could have made them participate right now, I am sure they will have more, and perhaps better chances as more of the sequels and prequels gets animated. We will analyze the numbers and who narrowly missed and make any changes needed in the future. 3. Although Nova 0's large group size made the results very foggy, it isn't totally meaningless. I think at worst, characters are off by about 4 ranks from where they should be. For Homura, that would mean she could end up anywhere from 1st to 7th in the Nova. While Homura might not finish in top 3, I sincerely doubt she will miss top 8 in Nova.
__________________
Last edited by wontaek; 2012-03-01 at 08:53. |
|
2012-03-01, 08:43 | Link #764 | ||||||
Senior Member
Author
|
Quote:
In other words, I'm largely with Last Sinner here. I don't necessarily think that Yuri and Yui are really that weak, but I do think that if you throw them in with a few other strong Nova girls, maybe they are that weak in that sort of contest format (similar to how Haruhi Suzumiya did much better in one-on-one matches than in several-way arena matches back in ISML 2011). In other words, maybe Yuri and Yui are a lot of people's 3rd, 4th, etc... choice (hence benefiting in one-on-one matches) but just not many people's 1st choice and hence do poorly in formats like this one. As for Kanade, I really doubt there's a lot of AB faction voters out there that vote first and foremost for AB characters as a rule, and on top of that would pick both Yuri and Yui over Kanade. This is really not even worth factoring into Kanade's vote totals, imo. Quote:
Quote:
Quote:
Quote:
Quote:
__________________
Last edited by Triple_R; 2012-03-01 at 09:05. |
||||||
2012-03-01, 10:39 | Link #765 | |
Senior Member
Join Date: Nov 2009
|
Quote:
Going by her overall personality (even without the toothbrush), I might have voted for her.
__________________
|
|
2012-03-01, 17:51 | Link #766 |
North American Haruhiist
Join Date: Oct 2010
Age: 43
|
I find myself wondering which Winter 2012 girls will show up in exhibition matches once the regular season starts.
HanaKana certainly voices a lot of characters for the Winter season. I imagine at least one of them has the potential to do well in future ISML matches. And of course, I am really hoping for a Madoka from Rinne no Lagrange vs Hibiki from Symphogear exhibition match some time this year. |
2012-03-01, 18:14 | Link #767 | |
┌(^o^)┘モリ!┌(^o^)┘モリ!
Join Date: Sep 2010
|
Quote:
|
|
2012-03-01, 21:11 | Link #768 |
NOM
Join Date: Jun 2008
Location: Outside the Asylum
|
It makes sense that having two votes makes vote-splitting a meaningless concept. If a voter is splitting his vote, then he's probably not really that loyal a series voter anyways. In fact, it also kinda makes sense that it would attract block-voting; that's how it works in JSaimoe, in any case.
I thought like you in the beginning. But I've been proved wrong time and time again by ignoring its effects in fantasy predictions. It was, in fact, the first and foremost glaring mistake I made. Sorry Triple R, but what makes sense is not always what is reality, what should probably happen is not always what happens. The stats speak louder than anything. Your only significant argument against this is that they've gotten weaker, but come now, it's only been several days; unless you can argue with a straight face that Kanade has the potential to be beaten by an Idolm@ster girl (sorry if I'm stepping on anyone's shoes here ) by the time SE rolls around, you're grossly overestimating the change in strength over time. Nor has fluctuations in the day of the match ever been shown to impact results in any visible way.
__________________
|
2012-03-01, 21:28 | Link #769 | |||
Senior Member
Author
|
Quote:
All I'm seeing so far from you and Team Rocket Elite is very much disputable conjecture and interpretation. If you have stats proving your case, then put it out there, and let us all evaluate it. Quote:
Quote:
It only makes sense that if you have highly unusual 12-way matches between many popular girls that this will eat into the support of all the participants, and likely bring about some unusual results simply due to vote splitting in general. Not everybody who regularly votes for Kanade has her as his/her top girl, or even 2nd favorite girl. So when Kanade goes up against a lot of other popular girls, she suffers from it just like everybody else does, as a lot of her support is whittled down closer to her core voters who have her as the top girl. Series faction voting may well have nothing to do with it at all, and if your contention is that it does have something to do with it, then please put forward some specific case studies along with clear stats proving your case. I'm open to reconsidering my stance, but not based on conjecture and interpretation alone.
__________________
|
|||
2012-03-01, 22:19 | Link #770 | ||||
Senior Member
Join Date: Jul 2008
|
Quote:
Quote:
Quote:
Quote:
__________________
|
||||
2012-03-01, 22:52 | Link #771 | ||
Senior Member
Author
|
Quote:
I mean, I'm sure there's some arenas were a lot of voters only know 2 to 4 of the girls in it, and a girl having the name recognition that comes with being a character in a popular show can be helpful. Now, if you get a three-way split going on here (such as Kyouko, Mami, and Sayaka all being in one arena) then yeah, I can see one or more (primarily the weakest) girl getting hurt here. And sure enough, Sayaka unfortunately got killed here, doing significantly worse than she usually does. But once name recognition isn't an issue any more - Once you have many of the strongest girls sharing an Arena together - I don't think that sharing an arena with other girls from your anime show will have significant impact if you have multiple votes to go around. That's my position right now, as it makes sense to me, and fits with what I've seen and can recall of ISML 2012 so far. As for how I would define faction voting - To me, it simply means wanting to see a particular anime show do well in these competitions through having multiple representatives of it do well (so it's similar if not identical to your own definition). If faction voting had played a role in Nova 0 then there shouldn't be such a large difference between Kanade's vote total and the vote totals of Yuri and Yui combined. Quote:
__________________
|
||
2012-03-01, 23:49 | Link #773 |
Senior Member
Author
|
Based on what Team Rocket Elite and Eater of All have argued, I looked over the recent ISML 2012 results, and paid special attention to matches involving two or more girls from the same anime series.
In doing so, what struck me is how much/often they weren't hurt by it. I have to be frank - I'm honestly wondering where Team Rocket Elite and Eater of All are seeing a pattern at all. In Stella 1 for Phase 1, Day 2 Results, we see Mikoto totally dominate the arena and Index finished a very close third behind Hecate. Doesn't seem like in-series vote splitting played much of a role here. In Stella 3 of Phase 1, Day 2 Results, we see Azusa easily win the arena with Mugi coming in a solid third behind Shinobu. There's reason to believe that Shinobu is simply getting a Nise bounce, and hence may be legitimately stronger than Mugi now. In Nova 1 of Phase 1, Day 3 Results, we see Kanade totally dominate the arena and Asami finish a close third behind Shiori. Maybe Asami was hurt slightly by in-series vote splits, or maybe Shiori has decent strength. Kanade certainly wasn't hurt by in-series vote splits here. In Stella 8 of Phase 1, Day 3 Results, we see Tomoyo and Kyo finish 2nd and 3rd behind Haurhi, and comfortably ahead of everybody else... entirely in line with the relative strength of these girls. Doesn't seem like in-series vote splitting played a role here. In Stella 11 of Phase 1, Day 3 Results, we see what to me seems like a strong counterargument to the in-series vote-splitting argument as Saber and Rin comfortably finish 1st and 2nd. In Stella 12 of Phase 1, Day 3 Results we had Last Order pull off a victory that stunned many of us and she had another Index girl in that match with her. Kotomi and Fuko both do well, finishing 3rd and 4th, very much in line with their perceived strength level (if not slightly above it). This arena really seems to fly in the face of the In-Series Vote-Splitting argument. In Stella 2 of Phase 1, Day 4 Results, we have Yuki Nagato pull off a decisive victory while Mikuru finished a very respectable third behind Nagi. Heck, if anything, this arena makes me wonder if Mikuru benefited from a SOS Brigade bounce, lol. In Stella 5 of Phase 1, Day 4 Results, Taiga pulls off a decisive victory while Minori finishes middle of the pack. Given who finished ahead of her (Kagami and Mugi), Minori finishing fourth is the best she could have reasonably hoped for here anyway. And Taiga certainly didn't seem to be held back by Minori's involvement. In Stella 6 of Phase 1, Day 4 Results, we see two Index girls finish fourth and fifth. Hina is unquestionably the strongest girl here, and Holo is arguably 2nd strongest. Ushio finishing third is a little bit surprising so maybe in-series vote-splitting had a slight impact here. It's one of the better cases for it, I'll admit. In Nova 2 of Phase 1, Day 5 Results, we see Kuroneko finish a decisive first and Ayase finish a close 3rd behind Shiori. Either Shiori is the master of taking advantage of in-series vote splits, or she really is decently strong, going by how she did in Nova 1 of Phase 1, Day 3 Results. In Nova 5 of Phase 1, Day 5 Results, we see Kyouko and Madoka finish a solid 2nd and 3rd behind Yuri. Kyouko and Madoka's vote totals are fairly in-line with what I noticed them get in other matches. In Stella 7 of Phase 1, Day 5 Results, we see Mio decisively win the arena while Mugi finishes a solid third behind Holo. In-series vote-splitting didn't seem to play a role here. In Stella 10 of Phase 1, Day 5 Results, we see Kyo and Nagisa finish second and third behind Nadeko. They also both finished ahead of Fate, it should be noted. In-series vote-splitting didn't seem to play a role here. In Stella 12 of Phase 1, Day 5 Results, we see Last Order finish a very respectable 2nd to Rin, while Kaori was also in the match. Kaori is fairly weak, and her performance reflected that. I don't see much evidence of in-series vote-splitting here. In Stella 3 of Phase 1, Day 6 Results, we see Azusa totally dominate her arena, and Ui finished 3rd behind Nagi and ahead of Fuko. Nothing seems out of the ordinary to me here. And... I'm getting exhausted at this point, so I'll leave it at that. No doubt I missed some "multiple girls from same series in one match" matches (and I skipped over a few since I'm not very familiar with the strengths of the girls in them), but from what I focused on, I certainly don't see a pattern of in-series vote-splitting causing significant vote shifts.
__________________
Last edited by Triple_R; 2012-03-02 at 00:06. |
2012-03-02, 00:16 | Link #774 |
Mishaguji-sama
Join Date: Oct 2010
|
In general, same series splitting affects weaker girls (who depend more on name recognition) much more than stronger ones. Basically, the stronger a character is, the less they get affected by it. That's why you don't see strong girls losing much power from same series splitting. And by "strong", I mean characters who are strong relative to other characters in their series, since that means they're more likely to be voted because they are them, rather than because they're representing their series. Also, very weak girls (again, relative to their series) also don't get affected much, since phase I involved large arenas and thus name recognition doesn't get them as far, really.
Day 2: Remember that Hecate was someone who did not even make it into regular season last year, while Index did, and while she didn't do too well she did won some matches. If anything, losing to Hecate here is surprising. (And indeed, she won against Hecate on day 6, when only Uiharu who is almost likely in a different series is within). Also, Mugi lost to Shinobu by like 900 when she was getting close to Kagami on day 4 without anyone from K-on in the same arena. Day 3: Iwasawa is not really that good of a comparison, as she relies much on name recognition to begin with and thus is almost like getting split in large size arena even without anyone actually splitting. Tomoyo and Kyou have no one close to compare to begin with (Maria and Kanzaki Kaori aren't going to be putting up much fights no matter what happens). Ikaros got rather uncomfortable close to Rin, who is someone strong (and maybe a bit lucky) enough to actually advance in phase I. As for Last Order... index and railgun are almost different series, since Uiharu barely showed up in Index II and Last Order is completely absent in Railgun. (And my theory is that Last Order gets vote because people who don't speak English see her image and think she's Mikoto, but that's off topic.) Day 4: Nothing I really can say about Nagi vs Mikuru, though Nagi isn't looking all that strong this year, and they were about equal strength in last year's regular season anyways. Minori depends a lot on name series for what power she has to begin with. Day 5: Ayase lost to Shiori here despite winning in Day 1. Kyouko and Madoka's opponents (other than Yuri) were very weak to begin with. Mugi actually did OK despite the split so I can say nothing. Fate is pretty weak during phase I (please don't kill me Demi), so if anything Clannad girls should've won by more by themselves. Last Order is quite strong (again, thanks in no small part to her Mikoto-like avatar most likely) to begin with, and Kaori... well, like Iwasawa she depends much on name recognition to begin with. Day 6: Ui winning by 30 votes against Fuko isn't really a comfortable margin, considering that Fuko also mostly gets votes because she's from clannad than anything. |
2012-03-02, 00:31 | Link #775 |
Senior Member
Join Date: Jul 2008
|
Looking over your comments, I disagree with you a lot regarding the strengths of characters. Just a few things near the beginning I saw were you don't see Iwasawa losing to Shiori as an underperformance despite Iwasawa beating Haqua the day before. Tomoyo with Kyou getting killed by Haruhi looks fine to you even though Tomoyo then does a lot better on Hitagi two days later. You don't think Index losing to Hecate is odd when Index does well on Konata in the next match. Tsumugi barely beats Ushio with Azusa around then has no problems with Yuuki Mikan. Of course, the problem with these examples are you probably disagree with the strength placement of the characters I'm comparing against which is the trouble with coming up with specific examples.
__________________
|
2012-03-02, 00:41 | Link #776 | ||||||
Senior Member
Author
|
Quote:
Quite the contrary, I think that name recognition is very important. For moe competitions, a girl is often better off being the 3rd, 4th, or even 5th most popular female character in a very popular series than she is being the most popular female character in a much more obscure anime. Quote:
Yuri is such an important part of Angel Beats! in her own right that I'm not sure if it's fair to say the same about Yuri, though. It's almost like saying "Haruhi Suzumiya is helped by being in The Melancholy of Haruhi Suzumiya". Quote:
Quote:
Quote:
Quote:
Does it really matter? A vote shift that doesn't result in a change of the order that the girls come in is not even significant, imo.
__________________
Last edited by Triple_R; 2012-03-02 at 00:53. |
||||||
2012-03-02, 00:46 | Link #777 |
You're Hot, Cupcake
Join Date: Aug 2008
Age: 43
|
That's certainly an exhausting way to look at it all.
But frankly, it still all suggests the final 4 will be Kanade, Shana, Kuroneko and Azusa (or possibly Taiga). I don't see 2011 characters making it that far.
__________________
|
2012-03-02, 00:50 | Link #778 | |
Senior Member
Author
|
Quote:
It's important to keep in mind that the strength of a lot of these girls could very well be in flux (as Mato recently demonstrated very clearly). Who knows exactly how strong Hecate or Shiori or Shinobu are? Reading over your response, I get the impression that the potential for strength fluctuations might be getting underestimated here, especially for girls in ongoing anime shows. And beyond that, we have a bunch of brand new girls here. Assuming perfectly static strength levels across the board seems strange to me.
__________________
|
|
2012-03-02, 01:03 | Link #779 | ||
NOM
Join Date: Jun 2008
Location: Outside the Asylum
|
By the way, in regards to the Kanade thing, let's just say it's impossible to gauge how much Yui and Yuri's presence had an effect on her performance. Kanade gives every girl such beatdown, in places as unstable as group arenas with 2 votes per voter, that anything short of a several-hundred margin close match will be inconclusive.
I'd venture a guess that the split-vote exists, simply because it exists in previous matches. Without any statistical suggestion otherwise, I wouldn't suddenly dismiss a trend. But I do concede that the effect on Kanade would be so miniscule, due to the fact that she greatly overshadows her two kin, that it would barely hold her back. Now to talk about split-voting in general.... Quote:
Quote:
- In Day 1 Stella 11, Ui beat Nagi by 46 votes. In Day 6 Stella 3 (where Azusa was present), Ui lost to Nagi by 577 votes. - In Day 2 Nova 10, Ayase beat Shiori by 83 votes. In Day 5 Nova 2 (where Kuroneko was present), Ayase lost to Shiori by 159 votes. - In Day 5 Nova 1, Houki beat Wendy by 141 votes. In Day 1 Nova 6 (where Charlotte was present), Houki lost to Wendy by 83 votes. - In Day 5 Stella 8, Kallen beat Shinku by 152 votes. In Day 1 Stella 9 (where C.C. was present), Kallen lost to Shinku by 187 votes. I'll be the first to admit these aren't the most profound of evidences, nor are there much of these simple direct rematches (looking for even one-removed relationships where a > b > c would take hours... ). Regal is right that the stronger of the two same-series character gets affected exponentially less by the split-vote, since in one instance, Index was able to win a super-close repeat match with Hecate despite being held back by Uiharu. Still, these same-series split-votes do have an effect on placement, which is enough to make it significant to recognize.
__________________
|
||
2012-03-02, 01:09 | Link #780 | |
Ass connoisseur
Join Date: Feb 2007
Location: Florida
Age: 37
|
Quote:
During the match where Tsumugi and Azusa were together, I did not notice a drop for either of them...Same goes for Ritsu and Mio. Yet when it was Azusa and Ui...Ui took a noticable dive in strength. In the case where Kyou and Tomoyo were paired with Haruhi...I honestly just think Haruhi is noticably stronger than Hitagi. It may not have seemed that way after their seeding match, but I personally think that kind of format means diddly squat as far as strengths goes. If anything, it shows fan base loyalty. Which is why I think it was both an over performance by Hitagi, and an under performance by Haruhi. Haruhi clearly looks stronger to me during any of the preliminary matches, so it's not odd to think Haruhi would garner more votes when against Tomoyo then Hitagi would. For instance, Hitagi defeated Nagisa by 900 votes and Tomoyo by 500 votes. Neither suffered from split votes, and Tomoyo is easily 400 votes stronger than Nagisa. Meanwhile, Haruhi defeated Nagisa by 1600 votes, and Tomoyo by 1200 votes. A rather huge difference compared to Hitagis performance against them. Tomoyo may have suffered from a marginal amount of split votes, but it is not too far off from what her average performance would be against Haruhi. On a side note, I also think it's the only reason Homura performed so well. She has a loyal fan base, but her fan numbers are limited, which explains why she takes a dive while in weaker groups. And probably when it comes time for 1v1 match play. But again, there are several other matches where split voting was noticeable. It's evident with Kuroko and Saten, Yami and Yui, as well as some others...But there is an equal number of them where I just don't see an underperformance despite the supposed split voting. One this is for sure though, I never see a character over perform when split voting is present.
__________________
|
|
Tags |
isml, saimoe, tournament |
|
|