AnimeSuki Forums

Register Forum Rules FAQ Members List Social Groups Search Today's Posts Mark Forums Read

Go Back   AnimeSuki Forum > Anime Related Topics > General Anime

Notices

Closed Thread
 
Thread Tools
Old 2013-10-22, 10:14   Link #1721
Triple_R
Ho Ho Honoka!
*Author
 
 
Join Date: Jan 2008
Location: Newfoundland, Canada
Age: 33
Send a message via AIM to Triple_R
Quote:
Originally Posted by Archaeon View Post
What's SDO?
ISML has different periods (Aquamarine, Amethyst, Ruby, etc...)

At the end of each period, there's a 7-way contest for a secondary award. That secondary award is the necklace corresponding with the period (Aquamarine Necklace, Amethyst Necklace, etc...).
Won-Loss record, and whether or not a character has already won a Necklace in the current calender year, are the main determinants when it comes to who gets into the 7-way Necklace matches, IIRC. This much I really like, since I think it nicely spices up the ISML regular season, while also giving a wider range of girls a chance of winning something at least.

However, every participant going into the Necklace match has a predetermined period-based SDO score. This SDO score is based on each participant's Won-Loss record just during the current period (not the full regular season), and is also based on the strength levels of her opponents during that period (it's a bit complicated). For all three of the last three Necklace matches of 2013, Mikoto Misaka was stuck with a much-lower SDO than her strongest opponent SDO-wise. So in spite of how Mikoto received the most votes in two of those three matches, she never won a Necklace this year.

I think that SDO encourages heavy amounts of strategic voting, which means less heart-based voting (i.e. voting for who you like best in each match, period). I think that SDO also increases the importance of luck while decreasing the importance of character strength. And as we saw with Mikoto this year, it can be very punishing to certain characters.

For these reasons, I wish SDO didn't exist in ISML.

But this is a critic's side of things - To get a full picture of SDO, it might be good if a fan of SDO explains it as well.
__________________

Last edited by Triple_R; 2013-10-22 at 10:26.
Triple_R is offline  
Old 2013-10-22, 12:58   Link #1722
Ithekro
The Comet is Coming
 
 
Join Date: Feb 2008
Location: Republic of California
Age: 37
The finals this year show that a year restriction is not needed in ISML. Time will tell when the older girls are no longer considered moe' enough. The voters choose the girls to begin with so over time the older series fade if they are not considered moe enough any more.

The likes of Nanoha and Fate are gone. Fate was the first champion. The Rozen Maidens are gone, as is Higurashi, Kanon, and Clannad seems like it is on the way out to give room for Angel Beat and Little Busters.

Only the best of the best are suppose to remains, but even at the top, there will be someone on the bottom. If I remember correctly, this was the first year were everyone in stella won at least a single match. No one lost ever single match. Someone in Nova went all zero.
__________________
Dessler Soto, Banzai!
Ithekro is offline  
Old 2013-10-22, 16:00   Link #1723
Triple_R
Ho Ho Honoka!
*Author
 
 
Join Date: Jan 2008
Location: Newfoundland, Canada
Age: 33
Send a message via AIM to Triple_R
Quote:
Originally Posted by Ithekro View Post
The finals this year show that a year restriction is not needed in ISML.
I really don't follow this argument. Kanade and Kuroneko have been two of the strongest ISML girls since 2011. I wonder if the bulk of ISML's voters will ever move on from 2011? Is it good that a girl from an anime that hasn't aired in three years is still dominating regular seasons and going very deep into the Tiara rounds (winning it all in 2012, coming just short of that this year)?

Let's suppose, just for argument's sake, that ISML had the same entry rules as J-SaiMoe. Kuroneko still would have won, since she's from a recently airing anime. But her opponent would have been somebody other than Kanade. There's a good chance her opponent would have been Mikoto. Kuroneko vs. Mikoto probably would have attracted more voters than an all-HanaKana finale, no offense to HanaKana fans.
__________________
Triple_R is offline  
Old 2013-10-22, 16:36   Link #1724
Ithekro
The Comet is Coming
 
 
Join Date: Feb 2008
Location: Republic of California
Age: 37
You are looking too close to the present. ISML has been going since 2008 and none of the finalists from that year made it into the 2013 finals. ISML was set up as a Best of the Best of all Time (Relative to Anime), rather than a Who is the Best Girl from This Year Only, type tournament. The first year took in the winners of prior Saimoe tourmanents as a baseline (discounting 2003 and 2004 due to lack of international voting those years, and the 2002 winner entered by popular vote) and Korean Best Moe and Superbest Moe prior to Winter 2008. Then characters were voted in with the cast of Clannad dominating the votes. Yet it was Fate that won.

Procedures have changed over time, especially the automatic inclusion of Saimoe and Best Moe winners once those tournaments started banning international voters. But the basic core challenge remains to show who is the Best of the Best of All Moe Girls in Anime. Regardless of year.

When the voters get tired of a girl, they will stop voting for her and eventually she will disappear into the mists of time like the others that have gone before them.
__________________
Dessler Soto, Banzai!
Ithekro is offline  
Old 2013-10-22, 17:33   Link #1725
Team Rocket Elite
Senior Member
 
 
Join Date: Jul 2008
Quote:
Originally Posted by Triple_R View Post
And like Hachiko, I also prefer how there's more turnover within J-SaiMoe from year-to-year. J-SaiMoe thankfully has nobody like ISML Kanade.
There isn't really one but the Puella Magi sure are trying. 5 characters at the top that are competitive with each other vs 1 character at the top who beats everyone (until Kuroneko came along at least). I think you'll find a lot of people who dislike the former more. Also Angel Beats isn't even a year older than Madoka.

Quote:
Originally Posted by Triple_R View Post
Let's suppose, just for argument's sake, that ISML had the same entry rules as J-SaiMoe. Kuroneko still would have won, since she's from a recently airing anime. But her opponent would have been somebody other than Kanade. There's a good chance her opponent would have been Mikoto. Kuroneko vs. Mikoto probably would have attracted more voters than an all-HanaKana finale, no offense to HanaKana fans.
There's also a good chance it would have been perceived as a a landslide victory for Kuroneko. There's a trade off here as well. Kanade is also a Hanazawa Kana character but she actually had a legitimate chance to beat Kuroneko. Misaka is from a different fan base but now neither Kuroneko nor Misaka fans have any reason to rally because Kuroneko is nearly a lock to win. This isn't to say it's a guarantee. Misaka has on very rare occasions shown to be able to challenge Kanade and Kuroneko but who knows when that power will show up. For that reason, Kuroneko vs Misaka may not draw more voters.
__________________
A miracle that you believe in when you know it won't happen......... is hope.
Team Rocket Elite is offline  
Old 2013-10-22, 18:10   Link #1726
Triple_R
Ho Ho Honoka!
*Author
 
 
Join Date: Jan 2008
Location: Newfoundland, Canada
Age: 33
Send a message via AIM to Triple_R
Quote:
Originally Posted by Ithekro View Post
You are looking too close to the present. ISML has been going since 2008 and none of the finalists from that year made it into the 2013 finals.
Five years is a long time to wait for turnover.


Quote:
When the voters get tired of a girl, they will stop voting for her and eventually she will disappear into the mists of time like the others that have gone before them.
The problem with having the same handful of girls remain strong for three or more years straight is that it makes it harder for new girls to gain a following. Asuna, KYH, Mashiro, and Rikka have all been held back from making a real name for themselves due to the continuing dominance of Kanade, Kuroneko, Eu, and Mikoto. With J-SaiMoe-like entry rules, at least Kanade and Eu wouldn't be there this year, so chances are a Nova girl at least gets into semi-finals, and likely comes close enough to garner a following.

Now, I'll admit that it's not a huge deal yet. But yet another year of the old 2011 standbys dominating the top of ISML would definitely not be good for ISML. I hope that I'm wrong about this, but I very much think that the old 2011 standbys will continue to dominate in 2014. Four straight years of the same girls at the top is a bit much, imo.


Quote:
Originally Posted by Team Rocket Elite View Post
There isn't really one but the Puella Magi sure are trying. 5 characters at the top that are competitive with each other vs 1 character at the top who beats everyone (until Kuroneko came along at least). I think you'll find a lot of people who dislike the former more.
Bad comparison for two important reasons:

1. Madoka Magica took a full year off in 2012. In my view, dominating 2 out of 3 years is significantly different than dominating 3 straight. If you don't like Madoka Magica, at least you get a break from it in 2012. If you don't like Kanade, you're not getting a break at all.

2. I'm not just talking about Kanade here. Kanade, Kuroneko, Eu, and Mikoto are no less dominant in ISML 2011 through 2013 than the Madoka main cast are in J-SaiMoe this year and in 2011. So it's not 1 vs. 5, it's dominant group vs. dominant group. It's the same thing, except with one group, at least you get an one-year break from them.


Quote:
There's also a good chance it would have been perceived as a a landslide victory for Kuroneko.
I strongly disagree. Kuroneko only faced Mikoto one-on-one once all year, and it was an extremely narrow victory for Kuroneko as she won by fewer than 200 votes. Mikoto also defeated Kanade in a match that had over 16,000 votes (which is more votes than what even the Tiara 2013 Final had). Mikoto won the Tiara in 2011, and she had a currently airing anime on her side. I very much doubt that Kuroneko vs. Mikoto for the Tiara would have been a landslide.

I definitely think you're underestimating Mikoto, and I think that Kuroneko vs. Mikoto would have had more votes than Kuroneko vs. Kanade for the ISML Tiara Final.
__________________
Triple_R is offline  
Old 2013-10-22, 18:23   Link #1727
Ithekro
The Comet is Coming
 
 
Join Date: Feb 2008
Location: Republic of California
Age: 37
How long was Shana dominating ISML? About three years. Only recently has her fall been felt. Or Taiga, the suprise of 2009, who is still hanging on strong.
__________________
Dessler Soto, Banzai!
Ithekro is offline  
Old 2013-10-23, 07:25   Link #1728
wontaek
Senior Member
 
 
Join Date: Oct 2006
Location: Planet Earth
Age: 45
I think we also need to admit that there are some years that doesn't give as much characters to be excited about compared to some other years. It is possible that we had really fruitful year in 2010, thus 2011 newcomers were strong, but rather dry year in 2011.

Here is a good way to check the reality of the situation. For every year in ISML's existence, let's see what is the highest finish among the "new comers" and how many "last year" anime characters made it to top 16. I will do the last 3 years, but I encourage others to check my numbers and also see what happened in the first 3 years as well. For inaugural 2008 season, just use any new anime character from 2007 as a "newcomer" to the event.

2013 : Highest finish - 5th ( Shiina Mashiro ) , Number in top 16 - 5
2012 : Highest finish - 14th ( Hasegawa Kobato ) , Number in top 16 - 2
2011 : Highest finish - 3rd ( Tachibana Kanade ), Number in top 16 - 6

Sadly, the hype surrounding new characters that appeared in 2013 seems even less than the 2011 crop ( those that got into ISML in 2012 ). That could mean that 2014 might finally be a year that a past Tiara winner gets to win the Tiara again.

Let's think one more thing. It is possible that random fluctuation gives us rather poor harvest for moe characters in some years. Why should we deliberately reward the poor crop by letting them avoid being matched against crops from good year?
__________________
wontaek is offline  
Old 2013-10-23, 10:55   Link #1729
hinakatbklyn
Senior Member
 
Join Date: Nov 2009
2008 ISML:
- Exclusively 2007 or later - (3) Lucky Star, (3) Clannad, (1) Hayate no Gotoku = 7 new comers in top 16
- Highest top 16 finish = Hiiragi Kagami (3rd Place)

- Add in Code Geass <C.C.> (1st season was between 2006 and 2007) = 8 new comers in top 16

2009 ISML:
- 10 out of 16 repeat top 16 finishes
- New series in 2008 - (1) Toradora, (1) Spice & Wolf = 2 new comers in top 16
- New series in 2008 not included in 2008 ISML - (1) Toradora = 1 new comer in top 16 (I see Holo/Horo at 19th in 2008 ISML)
- Highest top 16 finish = Aisaka Taiga (7th Place)


2010 ISML:
- 8 out of 16 (2009 challengers) repeat top 16 finishes
- New series in 2009 - (3) K-ON, (2) Monogatari series = 5 new comers in top 16
- Highest top 16 finish = Mio Akiyama (Tiara)
* 3 returning characters with their first appearance in top 16 = Mikoto, Nagi Sanzen'in, Saber
__________________
hinakatbklyn is offline  
Old 2013-10-23, 11:38   Link #1730
Ithekro
The Comet is Coming
 
 
Join Date: Feb 2008
Location: Republic of California
Age: 37
Makoto in her first year did horrible. It took here getting her own series to get her to the top 16.
__________________
Dessler Soto, Banzai!
Ithekro is offline  
Old 2013-10-23, 12:26   Link #1731
hinakatbklyn
Senior Member
 
Join Date: Nov 2009
Seeing that the other tiara winners already had ambitions of winning the tiara before they started their first year, 42nd place for Mikoto (Railgun) would be considered horrible. She still managed 23 wins and there were a total of 64 challengers in 2009.

23 - 40 and 42nd place was still good enough for Mikoto, even in 2009 to be the most popular Index character stat wise. Main character Index was 16 - 47 and 48th place
__________________
hinakatbklyn is offline  
Old 2013-10-29, 23:06   Link #1732
broken270
Defying gravity
 
 
Join Date: Jul 2010
I feel as though we need "something" to gauge character strength, not just by votes for alone. SDO (strength of the opponent) is at least a something: characters should be rewarded by how difficult their fight was with their supposed more difficult opponent. While it can be manipulated, I currently cannot think of an alternative method to create a proper "gauge" on who is worth going to the necklace round. The only thing I can think of at the moment is to diminishing further the power SDO has over matches, which I believe they have done if we compare last year to this year. Maybe, instead of dividing the final SDO by 3, divide it by 4 or 5, although that sounds like a lot, even though it does a job of decreasing SDO power. Or maybe have the maximum value of SDO be 2 rather than 3, so the highest anyone can make is 120 in comparison to 180, although I am sure they used 3 for a reason. Or, we can make SDO so complicated that not only do voters have to care about the strength of their opponents, they have to care about the strength of that person's opponent, which means they have to care about that characters strength for her opponents, etc., unless that is how SDO actually does its magic, since I still don't understand well the concept of all of that is. However, I think that provides rather minor numbers as opposed to the core 7 that manipulators should look at, but they do ultimately add up in the end one way or another. These are some of my thoughts on this topic to be taken with a grain of salt.

Sadly, I believe wontaek is correct that the probability of a former Tiara holder winning next year is very high. Unless Nova gets a surprisingly strong contender that can fend against the likes of Kanade, Ruri, and Mikoto, or someone from Stella not already mentioned get a major popularity boost (i.e. another season, etc.), I don't know who can face them other than themselves. I'm looking at Henneko and Date a Live, but I'm having my doubts. That, or I'm not watching enough "moe" anime by international standards.

Last edited by broken270; 2013-10-29 at 23:22.
broken270 is offline  
Old 2013-10-30, 08:34   Link #1733
iamadooddood
NOM.
 
 
Join Date: May 2013
Location: Singapore
Age: 22
Quote:
Originally Posted by Triple_R View Post
ISML has different periods (Aquamarine, Amethyst, Ruby, etc...)

At the end of each period, there's a 7-way contest for a secondary award. That secondary award is the necklace corresponding with the period (Aquamarine Necklace, Amethyst Necklace, etc...).
Won-Loss record, and whether or not a character has already won a Necklace in the current calender year, are the main determinants when it comes to who gets into the 7-way Necklace matches, IIRC. This much I really like, since I think it nicely spices up the ISML regular season, while also giving a wider range of girls a chance of winning something at least.

However, every participant going into the Necklace match has a predetermined period-based SDO score. This SDO score is based on each participant's Won-Loss record just during the current period (not the full regular season), and is also based on the strength levels of her opponents during that period (it's a bit complicated). For all three of the last three Necklace matches of 2013, Mikoto Misaka was stuck with a much-lower SDO than her strongest opponent SDO-wise. So in spite of how Mikoto received the most votes in two of those three matches, she never won a Necklace this year.

I think that SDO encourages heavy amounts of strategic voting, which means less heart-based voting (i.e. voting for who you like best in each match, period). I think that SDO also increases the importance of luck while decreasing the importance of character strength. And as we saw with Mikoto this year, it can be very punishing to certain characters.

For these reasons, I wish SDO didn't exist in ISML.

But this is a critic's side of things - To get a full picture of SDO, it might be good if a fan of SDO explains it as well.
This SDO seems unfair. (Lower SDO = disadvantaged, I take it?)

So what's SDO stand for, again?

Quote:
Originally Posted by broken270 View Post
I feel as though we need "something" to gauge character strength, not just by votes for alone. SDO (strength of the opponent) is at least a something: characters should be rewarded by how difficult their fight was with their supposed more difficult opponent. While it can be manipulated, I currently cannot think of an alternative method to create a proper "gauge" on who is worth going to the necklace round. The only thing I can think of at the moment is to diminishing further the power SDO has over matches, which I believe they have done if we compare last year to this year. Maybe, instead of dividing the final SDO by 3, divide it by 4 or 5, although that sounds like a lot, even though it does a job of decreasing SDO power. Or maybe have the maximum value of SDO be 2 rather than 3, so the highest anyone can make is 120 in comparison to 180, although I am sure they used 3 for a reason. Or, we can make SDO so complicated that not only do voters have to care about the strength of their opponents, they have to care about the strength of that person's opponent, which means they have to care about that characters strength for her opponents, etc., unless that is how SDO actually does its magic, since I still don't understand well the concept of all of that is. However, I think that provides rather minor numbers as opposed to the core 7 that manipulators should look at, but they do ultimately add up in the end one way or another. These are some of my thoughts on this topic to be taken with a grain of salt.

Sadly, I believe wontaek is correct that the probability of a former Tiara holder winning next year is very high. Unless Nova gets a surprisingly strong contender that can fend against the likes of Kanade, Ruri, and Mikoto, or someone from Stella not already mentioned get a major popularity boost (i.e. another season, etc.), I don't know who can face them other than themselves. I'm looking at Henneko and Date a Live, but I'm having my doubts. That, or I'm not watching enough "moe" anime by international standards.
Simple, absolute number of votes for and votes against. Works in various sports leagues (and tournaments with group rounds) as ranking tiebreakers. Since ISML is pretty much a league, it should work here too?

Take a certain girl A. Her first match is against girl B, where she wins 3456-2341. Second match is against girl C, where she loses 2322-3376. Her score after these 2 matches will be: 1 win, 0 ties, 1 loss, 5778 votes for, 5717 votes against.

In the event of the same number of wins/ties/losses (points) for 2 girls (if more girls are tied, adjust the criteria accordingly), use the following steps to determine a winner. Not all need to be used; it varies from league to league. Of course, these are not the only tiebreaker methods; there are many others, some of which suit the scoring system of a specific league.

In the event of a tie, votes for/against or head-to-head (either one can have higher priority, but the other must come into play if the first one fails to determine a winner) can be used.

If after applying both factors both are still tied (or there is no head-to-head encounter as of the time where a tiebreak is needed), then the girl with more votes (implies having played in tougher matches than the other girl and still came out on top) wins. Some leagues actually put head-to-head encounter at a lower priority than this.

If it's still a tie you can repeat the steps in the above paragraphs, this time taking only the scores of opponents both girls have had; this is more or less indirect head-to-head encounter. Or you can take only the scores of opponents both girls have had the first time round, then take overall scores; it's up to the league really. (Obviously this is not likely to work at certain times, such as when all the league matches are over, but just putting it here for when it works.)

If both are still tied (and they haven't had their head-to-head encounter yet), they could have their match right here as a playoff, where the winner goes through, and when their scheduled match comes up they'll have a break.

If that match is a tie... only then do you use SDO.
__________________
When life gives you lemons... enjoy reading them.

Anime Saimoe Tournament 2013 is screwed up. Everyone's argument is invalid.

Last edited by iamadooddood; 2013-10-30 at 09:22.
iamadooddood is offline  
Old 2013-10-30, 09:05   Link #1734
Triple_R
Ho Ho Honoka!
*Author
 
 
Join Date: Jan 2008
Location: Newfoundland, Canada
Age: 33
Send a message via AIM to Triple_R
Quote:
Originally Posted by iamadooddood View Post
This SDO seems unfair. (Lower SDO = disadvantaged, I take it?)
Yes, substantially so. The idea behind SDO is to make the Necklace matches less predictable, and hence more interesting. I certainly respect that idea, and see value in it, but in actual application, it just doesn't work.

Kanade and Kuroneko winning the first two Necklaces this year was horribly predictable anyway. SDO just made it even more predictable.

Asuna, Rikka, and Mashiro all enjoyed massive SDO advantages in Ruby, Emerald, and Topaz respectively, making their eventual victories painfully obvious. In fairness, Mikoto made things surprisingly close in Emerald, but the fact that such an impressively strong performance still wasn't enough to overcome Mikoto's SDO disadvantage makes it even more clear how obvious Rikka's eventual victory was.

SDO does not accomplish what it is supposed to do - Which is make the Necklace matches more interesting/less predictable. In some cases, it actually makes the Necklace matches more predictable than what they otherwise would be. And SDO ends up unfairly screwing some girls, as it did Mikoto, in my opinion. There's no question that Mikoto wins a Necklace this year if not for SDO.
__________________
Triple_R is offline  
Old 2013-10-30, 09:40   Link #1735
iamadooddood
NOM.
 
 
Join Date: May 2013
Location: Singapore
Age: 22
Quote:
Originally Posted by Triple_R View Post
Yes, substantially so. The idea behind SDO is to make the Necklace matches less predictable, and hence more interesting. I certainly respect that idea, and see value in it, but in actual application, it just doesn't work.

Kanade and Kuroneko winning the first two Necklaces this year was horribly predictable anyway. SDO just made it even more predictable.

Asuna, Rikka, and Mashiro all enjoyed massive SDO advantages in Ruby, Emerald, and Topaz respectively, making their eventual victories painfully obvious. In fairness, Mikoto made things surprisingly close in Emerald, but the fact that such an impressively strong performance still wasn't enough to overcome Mikoto's SDO disadvantage makes it even more clear how obvious Rikka's eventual victory was.

SDO does not accomplish what it is supposed to do - Which is make the Necklace matches more interesting/less predictable. In some cases, it actually makes the Necklace matches more predictable than what they otherwise would be. And SDO ends up unfairly screwing some girls, as it did Mikoto, in my opinion. There's no question that Mikoto wins a Necklace this year if not for SDO.
Looks like the only way for SDO to make things more unpredictable was to give those with a lower SDO score the advantage.

I just looked at the rules and saw the scoring format:

Quote:
1. Greater number of wins
2. Winner of the tied Contestants’ head-to-head match, if applicable
a) This is calculated only at the end of each Period and at the end of the Regular Season
3. Greater Strength of Schedule of Defeated Opponents (SDO)
4. Greater Strength of Schedule of All Opponents (SAO)
5. Greater Vote Differential (VD)
6. Greater number of Votes For (VF)
7. Selection by Staff
Looks like they did follow typical tiebreaking rules (except for taking the scores of/against common opponents only; has it been done in ISML before?). Just a few things though.

Methinks something like this kind of scoring coould be used in between SDO and SAO. Also, VD and VF should be higher on the list, but that's just me.

In addition, ties in leagues with this kind of scoring system are just nasty to the girls involved, with this sort of rules. Win-tie-loss should be 2-1-0, not 3-1-0, and "greater number of wins" should be much lower in the priority bracket. I mean, the girl with 1 win and 3 losses is stronger than the girl with 3 ties and 1 loss. This results in stuff like 5 wins and 6 losses being stronger than 3 wins, 6 ties and 2 losses. There's no justice in that. At the very least, "greater win/loss ratio" would have been better than "greater number of wins".

"Selection by Staff"... I really don't like the fact that ties have to come down to this. And yet it's sometimes necessary. Grrrr.
__________________
When life gives you lemons... enjoy reading them.

Anime Saimoe Tournament 2013 is screwed up. Everyone's argument is invalid.

Last edited by iamadooddood; 2013-10-30 at 09:56.
iamadooddood is offline  
Old 2013-10-30, 22:59   Link #1736
RegalStar
Mishaguji-sama
 
 
Join Date: Oct 2010
Regular season standings had never come anywhere near needing staff selection before.

Quote:
Originally Posted by Triple_R View Post
Yes, substantially so. The idea behind SDO is to make the Necklace matches less predictable, and hence more interesting.
In two words: It's not.
RegalStar is offline  
Old 2013-10-30, 23:56   Link #1737
Triple_R
Ho Ho Honoka!
*Author
 
 
Join Date: Jan 2008
Location: Newfoundland, Canada
Age: 33
Send a message via AIM to Triple_R
Quote:
Originally Posted by RegalStar View Post
In two words: It's not.
That's not what I was told before.

I debated this with Eater of All way back in 2011. Here was Eater of All's comprehensive defense of SDO.

Key Excerpt from that linked post: Which forms the basis for why the current staff views necklace as nothing more than spice (important spice, but spice nonetheless) for voters to get through the regular season. It all comes down to this: making necklace matches be sole judge of necklaces (or any method that employs pure strength) would make things predictable, and therefore less fun.

If the fundamental issue is that people find pure strength = more fun than having luck be involved, then yeah, it's something we've got to agree to disagree on.

-------------------------------------------------------

Eater of All (who appears to be speaking on behalf of the ISML Staff) is clearly implying here that SDO is designed to make things less predictable, and hence "more fun". But as 2013 has made clear, it doesn't really work that way in application.

Anyway, for those interested, here was my counterargument to Eater of All back in 2011.


Finally, if "less predictability" isn't the point of SDO, then what is the point of it?
__________________
Triple_R is offline  
Old 2013-10-31, 11:19   Link #1738
iamadooddood
NOM.
 
 
Join Date: May 2013
Location: Singapore
Age: 22
Quote:
Originally Posted by RegalStar View Post
Regular season standings had never come anywhere near needing staff selection before.
I hope it stays that way.

If only sudden death was possible...

Quote:
Originally Posted by RegalStar View Post
In two words: It's not.
It's not unfair, or it's not less predictable/more interesting? These are two completely different things.


EDIT:

@Triple_R: I just finished reading the posts you directed me to, and it gave me some insight.

If ISML's tiebreaker rules were entirely up to me, then I'd do it like this (at least tentatively, in case any better suggestions come along):
  1. Scores against common opponents + head-to-head only
    1. Points total
    2. Vote differential
    3. Votes for
    4. Head-to-head (if applicable)
  2. Total score
    1. Vote differential
    2. Votes for
    3. SDO + STO/2 (where STO = Strength of Schedule of Tied Opponents)
    4. SAO
  3. Playoff (only if head-to-head has not been carried out yet)
  4. Selection by staff
Comments are welcome (especially from actual ISML admins). Of course, read the notes below first.

First and foremost, I'd completely scrap the "greater number of wins" part, because ties are typically extremely rare in this kind of competition (think ties in sports like cricket, and you'll see what I mean), and one vote can determine if it's a tie or a win/loss. It's as good as leaving it up to luck, and even worse than what SDO's doing. Of course, I'd use the 2-1-0 points system for a win/tie/loss as well.

I just noticed bonus fractional points being awarded to losers of a match, depending on the % of total votes obtained in the match. (How did I miss this out the first time. Now that's just unfair to those who got ties.) I'd like to extend this system to either:
  • give this bonus to both sides of the match, regardless of results (so each match gives exactly 3 points; if it were up to me I'd be extremely strict with the fractions as well), or
  • give 1.25 points to each side in the event of a tie, instead of 1, or
  • give 1.125 points to each side in the event of a tie.
Or we could keep the current system (with 2 points awarded for a win instead of 3).

On my stance on luck, it should play as little a part on unpredictability as possible. Because that kind of unpredictability is the "undesired" kind. Unpredictability due to effort and strength are the "desired" kinds of unpredictability, and so should be encouraged as much as possible.

I deemed SDO and STO (SAO would be the same for obvious reasons) unnecessary in step 1 because, I mean, come on, it's against common opponents. It's as fair as can be. No need to add in S-whatever-O and risk screwing things up. Unlikely, but possible. For the same reason, I placed head-to-head only after everything else in step 1.

Step 3 is a completely optional choice; I don't mind if it's removed actually, but it does seem like quite a good idea to me. If implemented, this playoff will be treated as their actual match; when their scheduled match comes up they'll have a break instead.


EDIT 2: Here's how I'd do the necklace system if it were up to me (tentatively, that is):

5 or 7 necklaces a year is up to the admins. Heck, I wouldn't even mind if they alternated between 5 and 7 every other year.

Method to determine 4th-ranked contestant (if not by playoff)
  1. Total score
    1. Vote differential
    2. Votes for
    3. SDO + STO/2 (where STO = Strength of Schedule of Tied Opponents)
    4. SAO
  2. Playoff (optional)
  3. Selection by staff
In addition, unless anyone can convince me otherwise, I see no reason to add SDO to the Necklace score because it looks completely unfair.

Ties can be resolved as above.


Finally, if the postseason was up to me:

Phase 1 (for each division)

Round 1 (seeding unchanged from ISML 2013's format)
Match 1: 1 vs. 16
Match 2: 8 vs. 9
Match 3: 5 vs. 12
Match 4: 4 vs. 13
Match 5: 6 vs. 11
Match 6: 3 vs. 14
Match 7: 7 vs. 10
Match 8: 2 vs. 15

Round 2
Match 1: R1M1 winner vs. R1M2 winner
Match 2: R1M3 winner vs. R1M4 winner
Match 3: R1M5 winner vs. R1M6 winner
Match 4: R1M7 winner vs. R1M8 winner

Round 3
Match 1: R2M1 winner vs. R2M2 winner
Match 2: R2M3 winner vs. R2M4 winner
Match 3: R2M1 loser vs. R2M2 loser
Match 4: R2M3 loser vs. R2M4 loser

Round 4
Match 1: R3M1 winner vs. R3M2 winner
Match 2: R3M1 loser vs. R3M2 loser
Match 3: R3M3 winner vs. R3M4 winner
Match 4: R3M3 loser vs. R3M4 loser

Results (also seeding for Phase 2)
1: R4M1 winner
2: R4M1 loser
3: R4M2 winner
4: R4M2 loser
5: R4M3 winner
6: R4M3 loser
7: R4M4 winner
8: R4M4 loser
9-16: Determined with ISML 2013's method

Phase 2 will be done the same way as in ISML 2013.

Additional match: The Phase 1 winner of each division to face off against each other for a second tiara. This second tiara winner to then face off against the Phase 2 winner for the Champion of ISML for the year (unless the same girl won both tiaras, in which case she's the undisputed Champion). (Or you can call the prizes different stuff if you want.)

Again, comments welcome.
__________________
When life gives you lemons... enjoy reading them.

Anime Saimoe Tournament 2013 is screwed up. Everyone's argument is invalid.

Last edited by iamadooddood; 2013-10-31 at 14:14.
iamadooddood is offline  
Old 2013-10-31, 20:55   Link #1739
RegalStar
Mishaguji-sama
 
 
Join Date: Oct 2010
What is true is that the necklace matches are spices to get the regular season moving, and are not important to the overall picture. However, that's not what SDO is for. SDO is meant to be something that indicates the performance of characters over the period in question (and the period alone), and is always supposed to be the main basis for necklaces. The necklace match, to quote wontaek in an answer he gave to an interview (although the actual interview is translated to Chinese before publicized): "Still people may feel certain character’s SDO to be undeserved for whatever the reason. This is why we must have necklace group round. In a way, we are asking people which character deserves higher necklace score given the relative hardship they overcame."
RegalStar is offline  
Old 2013-10-31, 23:51   Link #1740
Triple_R
Ho Ho Honoka!
*Author
 
 
Join Date: Jan 2008
Location: Newfoundland, Canada
Age: 33
Send a message via AIM to Triple_R
So why should some complex period-based calculation scheme matter more than how competitors actually do head-to-head against each other in an exciting 7-way match?

Valuing SDO more than the Necklace match itself really strikes me as the wrong approach to take. Using SDO to determine who gets into the Necklace match is a fine idea, very reflective of pro sports leagues using regular season performance to determine who gets into the playoffs. But then let pure vote count determine who wins the Necklace. In the unlikely even of a vote count tie, use SDO as a tie-breaker.

That's my position.
__________________
Triple_R is offline  
Closed Thread

Tags
isml, saimoe

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 05:28.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2014, vBulletin Solutions, Inc.
We use Silk.