Staff's Ideas for ISML 2017 Format

[Fuijiwara]

Having seen the limitations of this year's format, the staff have brainstormed of ways to come up with new formats that may work, without taking up too much of the staff's time, which was the main reasoning behind this year's change.

And so, the staff have come up with 4 different new formats that COULD be applied to ISML 2017. Now, we want the community to choose what they like best amongst these ideas that our 'fabulous' staff have come up with.

Note: This is NOT a poll to decide next year's format. Rather, this is a poll to see what the forum community prefers.

Option #1- 128 Contestants Variant (2016's format)

SpoilerShow

(128 > 64 > 32 > 16 > 8 > 4 > 2; necklace period length 6 match days total, 3 match days per character; 53 match days total)
Round of 128: 32 groups of 4 characters, round robin; Aquamarine, Topaz, Amethyst and Sapphire Period

Round of 64: 16 groups of 4 characters, round robin; Emerald and Ruby Period

Round of 32: 8 groups of 4 characters, round robin; Diamond Period

Round of 16: 4 groups of 4 characters, round robin

Followed by single elimination

Option #2- 120 Contestants (Variant A)

SpoilerShow

(120 > 84 > 60 > 42 > 30 > 24 > 18 > 16 > 8 > 4 > 2; necklace period length 5 to 15 match days total, 5 match days per character; at least 66 match days total)

Round of 120: 20 groups of 6 characters, round robin; Aquamarine Period, Each group 5th and 6th will be thrown into an arena where only top 4 advances

Round of 84: 14 groups of 6 characters, round robin; Topaz Period, Each Group 5th and 6th will be thrown into an arena where only top 4 advances

Round of 60: 10 groups of 6 characters, round robin; Amethyst Period, Each Group 5th and 6th will be thrown into an arena where only top 2 advances

Round of 42: 7 groups of 6 characters, round robin; Sapphire Period, Each Group 5th and 6th will be thrown into an arena where only top 2 advances

Round of 30: 5 groups of 6 characters, round robin; Emerald Period, Each Group 5th and 6th will be thrown into an arena where only top 4 advances

Round of 24: 4 groups of 6 characters, round robin; Ruby Period, Each Group 5th and 6th will be thrown into an arena where only top 2 advances

Round of 18: 3 groups of 6 characters, round robin; Diamond Period, Each Group 6th will be thrown into an arena where only top 1 will advance

Followed by single elimination

Option #3- 120 Contestants (Variant B)

SpoilerShow

(120 > 96 > 72 > 60 > 48 > 36 > 24 > 16 > 8 > 4 > 2; necklace period length 5 to 15 match days total, 5 match days per character; at least 71 match days total)

Round of 120: 20 groups of 6 characters, round robin; Aquamarine Period, Each Group's 5th will be thrown into an arena where only top 16 will advance

Round of 96: 16 groups of 6 characters, round robin; Topaz Period, Each Group's 5th will be thrown into an arena where only top 8 will advance

Round of 72: 12 groups of 6 characters, round robin; Amethyst Period, Each Group's 5th and top three of the 6th will be thrown into an arena where only top 12 will advance

Round of 60: 10 groups of 6 characters, round robin; Sapphire Period, Each Group's 5th will be thrown into an arena where only top 8 will advance

Round of 48: 8 groups of 6 characters, round robin; Emerald Period, Each Group's 5th will be thrown into an arena where only top 4 will advance

Round of 36: 6 groups of 6 characters, round robin; Ruby Period

Round of 24: 4 groups of 6 characters, round robin; Diamond Period

Followed by single elimination

Option #4- 100 Contestants

SpoilerShow

(100 > 40 > 16 > 8 > 4 > 2; necklace period length 3 to 10 match days total, 3 match days per character; at least 55 match days total)

Round of 100: 10 groups of 10 characters, round robin; Aquamarine, Topaz and Amethyst Period. Each group 4th to 6th will be thrown into an arena where only top 10 will advance. Top 3 will advance from each group.

Round of 40: 4 groups of 10 characters, round robin; Sapphire, Emerald and Ruby Period. The rest will be thrown into an arena where only top 4 will advance. Top 3 will advance from each group.

Round of 16: 4 groups of 4 characters, round robin; Diamond Period

Followed by single elimination

Option #5- 64 Contestants

SpoilerShow

(64 > 16 > 8 > 4 > 2; necklace period length 3 to 6 match days total, 3 match days per character; at least 45 match days total)

Round of 64: 4 groups of 16 characters, round robin; Aquamarine, Topaz, Amethyst, Sapphire and Emerald Period

Round of 16: Swiss-Tournament style for 6 rounds; Ruby and Diamond Period

Followed by single elimination

Alternatively, if you have a suggestion of your own, please do tell us in detail how that tournament style might work and we could take into consideration of your method of organising a tournament.

Just a gentle reminder that you can offer criticism, but don''t do it baselessly. Also, this poll will last for 1 week.

Also, you can ask maglor or Shmion84 if you have any burning questions to be answered.

[Chibasa]

Before posting the option I gave on my survey, I want to comeback on some points for other options :3

Option #3 seems way better than Option #2 to me. Having a round of 18 leading to only 2 characters eliminated is kinda weird, and even before from 30 to 24 and 24 to 18, with group of 6, it would be hard to seed every time avoiding that some matches happen 4 times in the same year.

Option #5 is a maybe close to past editions, having 4 leagues instead of 2 + Postseason. So if the goal is to comeback to something closer to past editions (with technical possibilities/issues we have nowadays) it may be the best one. It's also the option with less characters passing nominations but since there's less match days in this one, Prelims comeback could be a possibility?

Overall the "Arena" idea can be good since it's avoid some characters die early cause they were seeded in a "death-group".

[shadowhunter]

I like an option #4.

Though I still dream about Double Elimination....

[Reverend]

I think, for next year at least, Option #5 would work best.
The amount of new characters from 15-16 generations with strong enough support to actually compete in Saimoe tournament are lesser than 11-12 gen. There's no need to confer that much spot to newbies.

Option #4 is also nice enough.

I personally dont like the rest. Too many group stages, it left some people confused this year. They would rather see a tree... or few big blocks.

[renfire]

Personally, I like 2015 Nova/Stella format. I can count my-version-statistics of which characters I vote/like the most among 36 characters of nova/stella. This how I determine my ISML favorite characters of the year. I haven't read other format yet, might vote/edit my post later.

[maglor]

Personally, I like 2015 Nova/Stella format. I can count my-version-statistics of which characters I vote/like the most among 36 characters of nova/stella. This how I determine my ISML favorite characters of the year. I haven't read other format yet, might vote/edit my post later.

The reason I feel we are not yet ready for 2015 Nova/Stella format are the following reasons

1. Need to have 112 or more characters in each voting date . This sheer size has potential to cause problems

2. Need to have complex and lengthy prelim stage . This would require frequent web page changes and different vote checking method for each date

3. Need to refine Nova/Stella definition.

Due to these areas of potential trouble, I prefer waiting at least two more years before considering 2015 format again. For next year, I am hoping for something bit shorter and likely of less work.

[renfire]

I see, Nova/Stella format has too many matches per day. My choice is either #2 or #3 then. I want each characters to have more matches. I hope the staffs can manage 66-71+ days. I feel that this-year-53-match-days is not enough.

If we are only doing 45-55+ match days. I want #1 because it has more contestants (I don't want #5).

I don't quite understand option #4. 10 groups of 10 characters, round robin. Each characters should have 9 matches, not 3. If it means that we are only doing 3 matches per 10-contestant group and whoever at top 3 will advances, I don't want this option.

[maglor]

I see, Nova/Stella format has too many matches per day. My choice is either #2 or #3 then. I want each characters to have more matches. I hope the staffs can manage 66-71+ days. I feel that this-year-53-match-days is not enough.

If we are only doing 45-55+ match days. I want #1 because it has more contestants (I don't want #5).

I don't quite understand option #4. 10 groups of 10 characters, round robin. Each characters should have 9 matches, not 3. If it means that we are only doing 3 matches per 10-contestant group and whoever at top 3 will advances, I don't want this option.

In option 4, during the round of 100 and Round of 40, all character will have 9 matches spread over 3 necklace period. Therefore, when everyone had 3 matches out of 9 in round of 100, the Aquamarine period will be over and we will be having Aquamarine Necklace match. Then will be Topaz Period when everyone had their 4th, 5th, and 6th match. Next will be Amethyst during which the last 3 matches of round of 100 will be performed. Round of 40 undergoes similar format.

[renfire]

In option 4, during the round of 100 and Round of 40, all character will have 9 matches spread over 3 necklace period. Therefore, when everyone had 3 matches out of 9 in round of 100, the Aquamarine period will be over and we will be having Aquamarine Necklace match. Then will be Topaz Period when everyone had their 4th, 5th, and 6th match. Next will be Amethyst during which the last 3 matches of round of 100 will be performed. Round of 40 undergoes similar format.

My first choice is option 4 then. It has less rounds, simpler. I think we should choose a format that is easy to be understood by everyone. I still like option #3 and #2, but it has too many rounds. I prefer #3 over #2 for the same reason as Chibasa.

[Chibasa]

I think it can be useful, so I put this here.

For wildcard round in option #4, in the final 10 from round of 100 to advance to round of 40, I am heavily leaning towards simply letting the voters choose up to 10 characters to let the top 10 character advance. I thought about using Single Transferrable Vote scheme, but that will require most of the voters to rank all 30 which probably would be tougher task than simply choosing the 10 they like. Similarly for final 4 from round of 40 to advance to round of 16, I would like the voters to pick up to 4 characters so we can simply go with top 4 in number of votes received.

[Chibasa]

My idea is basically a variant from Option #1. With weird stuff.

Hax option:
(128 > 64 > 32 > 16> 8 >4 > 2); necklace period length 6 match days total, 3 match days per character; 68 or 69 or 70 match days total)
Round of 128: 32 groups of 4 characters, round robin; Aquamarine, Topaz, Amethyst and Sapphire Period

Round of 64: 16 groups of 4 characters, round robin; Emerald and Ruby Period

Round of 32: 4 groups of 8 characters, round robin; Diamond Period

Round of 16 until the end: Single Elimination

Why?Show

I think making larger group in the end is worth, cause it helps getting a way better seeding, and it makes more characters from equivalent strength face each other. (With seeding, this year, we had still a lot of difference between character fighting each other in many matches in Diamond). And it also allow more "comeback" possibilities for character (Not just: "Oh I have Mashiro in my group it's finish for me").

+ A change to necklace:
Two rounds necklace.
1st round: Take 8 group winner + 4 best CVP second, for a total of 12 characters, seed by CVP in 3 group of 4 characters (2 qualify in each group).
2nd Round: 1 group of 6 characters

Note: Can be only one round in Diamond

Why?Show

I prefer 6 characters in necklace round (unlike many people answers in survey) because most of time there's many characters you know they have like ZERO chances to win (We had a lot of characters with 5/6% votes in 8-way Arena this year, this is not even half average vote (12,5%)). But the 2 round-necklace also add a bit spice, allowing new faces to believe in their dreams. Plus, since necklace rounds have always more voters than regular round, it may affect in a good way amount of voters ? And finally, to win necklace you need to be able to win 1v1 (to qualify), little group matches and Bloodbath Arena, which is moe.

+ Adding ToC during Necklaces matches:
With 8 characters in TOC, we can make a league which need 7 matches between each other.
We have 7 periods. Each period we could make a ToC day.
(With 2 nacklaces rounds, that mean 2 ToC match/round, but it also works with other options actually with 4 match/necklace)
In the end it gives a full ranking, and #1 + #2 could face each other for title during PS maybe :3

ConsShow

It does not work for Males since they are not 8 yet.

Moe CalendarShow

Here is Calendar with 70 match days. (There can be one less round in both R16 and Diamond Necklace technically).

Feel free to judge this one as a whole or point by point (since each point can technically makes sense in some other options). =3

[Reverend]

My idea is basically a variant from Option #1. With weird stuff.

Hax option:
(128 > 64 > 32 > 16> 8 >4 > 2); necklace period length 6 match days total, 3 match days per character; 68 or 69 or 70 match days total)
Round of 128: 32 groups of 4 characters, round robin; Aquamarine, Topaz, Amethyst and Sapphire Period

Round of 64: 16 groups of 4 characters, round robin; Emerald and Ruby Period

Round of 32: 4 groups of 8 characters, round robin; Diamond Period

Round of 16 until the end: Single Elimination

hmm...
why you have 8 characters in 1 group in a late stage of the tournament (when there are only a handful, chosen ones, remaining) when on the earlier stages you only have 4 per group (and the girls that got eliminated in those rounds are the ones that could use a second chance) ?

[Chibasa]

My idea is basically a variant from Option #1. With weird stuff.

Hax option:
(128 > 64 > 32 > 16> 8 >4 > 2); necklace period length 6 match days total, 3 match days per character; 68 or 69 or 70 match days total)
Round of 128: 32 groups of 4 characters, round robin; Aquamarine, Topaz, Amethyst and Sapphire Period

Round of 64: 16 groups of 4 characters, round robin; Emerald and Ruby Period

Round of 32: 4 groups of 8 characters, round robin; Diamond Period

Round of 16 until the end: Single Elimination

hmm...
why you have 8 characters in 1 group in a late stage of the tournament (when there are only a handful, chosen ones, remaining) when on the earlier stages you only have 4 per group (and the girls that got eliminated in those rounds are the ones that could use a second chance) ?

Because it means there will be a lot of matches between contestants from equivalent strength, with large fanbase too. It helps to get a way better seeding for PostSeason. It allows comeback on a stage with a lot of suspense, it creates an event bigger just before the end of the tournament? While putting it at the start would make a decrease of tension during tournament I guess.

[Euphony]

Having seen the limitations of this year's format, the staff have brainstormed of ways to come up with new formats that may work, without taking up too much of the staff's time, which was the main reasoning behind this year's change.

And so, the staff have come up with 4 different new formats that COULD be applied to ISML 2017. Now, we want the community to choose what they like best amongst these ideas that our 'fabulous' staff have come up with.

Note: This is NOT a poll to decide next year's format. Rather, this is a poll to see what the forum community prefers.

I made some calculations to help compare all 5 initial options and also Chibasa's Hax option. An image of the spreadsheet is spoilered below.

Calculation of total number of 1v1 matches in the tournament for…Show

Each row is for a stage of the tournament. Note that "1v1" means a match with one contestant facing one other contestant, "one versus one". I have been using some combination of "arena", "match", and "1v1" interchangeably, as well as "round" with "stage". If any of this usage is incorrect, redundant, or confusing, please let me know what would be better. If there are any calculation errors, please point them out.

Details of data labels:

• Contestants: The number of individual contestants remaining in the stage of the tournament.
• Groups: Number of round-robin groups in the stage.
• Contestants per Group: Number of contestants in each round-robin group.
• 1v1s per Group: Total number of matches among all contestants in a single group. Note that A's match against B and B's match against A are counted as one match, not two.
• Total 1v1 Arenas: Total number of matches among all contestants among all groups in the round.
• 1v1s per Contestant: Number of matches each contestant has throughout the round.
• Cumulative 1v1s per Contestant: Number of matches each contestant will have had from the start of the tournament through the given round.
• Grand total: Number of matches in the tournament among all contestants, groups, and rounds.
  • Wild card matches are counted separately from the Total.
• Ratio of Total 1v1 Arenas to Initial Contestants: Ratio to give a sense of the size of the tournament relative to the size of the contestant pool. Calculated as the Grand Total divided by the maximum Contestants.

I invite anyone with insights into what I've posted above to please share them. I'll add my own thoughts once I've figured them out.

Alternatively, if you have a suggestion of your own, please do tell us in detail how that tournament style might work and we could take into consideration of your method of organising a tournament.

Just a gentle reminder that you can offer criticism, but don''t do it baselessly. Also, this poll will last for 1 week.

I will post my detailed tournament suggestions in a later post. For now I'll just say that, contrary to the apparent consensus of the staff, I would prefer if ISML avoided round-robin tournaments altogether. I have yet to see data that suggests in general that ISML voters either prefer or expect a round-robin format for any aspect of the ISML tournament, nor have I seen any data that suggests that voters would dislike or reject major changes to the tournament's voting system. If anyone here has such data, please present it for discussion. Otherwise, perhaps the staff might consider asking about voting systems on the next ISML voter survey.

Or better yet, just do something completely different, regardless of the voter expectations and preferences, so long as it's convenient for the staff. As a voter, I don't want to demand so much time from the staff, and I accept that the staff may do whatever they please, regardless of voter preferences such as my own. But if I'm understanding the objective of ISML correctly, I hope the staff looks more skeptically at their present options and implements a voting system that is both convenient and satisfying for all involved with ISML, regardless of convention or precedent. More commentary to follow.

[IGNITE]

I will participate next year regardless of what sort of format is used. However the one limitation that I see from this year's change is the lack of proper seeding. Unlike the rest of the options, Option 1 has a lot of groups and low number of characters in the group, but the seeding are relaying on result of one match (nomination). There is no quick way to seed them all precisely as their 1 vs 1 performance but using only nomination is pretty insufficient when the characters are scattered to many groups. So maybe a simple seeding/prelim before start main tournament, something like how Stella Nomination worked in 2013 but not relaying on one match only.

ExampleShow

160 characters go into prelim.
Prelim Day 1
Each voter have 16 votes, match last for 24 hours.
Prelim Day 2
Repeat Day 1
Prelim Day 3
Repeat

All votes from 3 matches added up and the top 128 characters go into Main tournament, seeded into their groupings based on their results.

[maglor]

I will post my detailed tournament suggestions in a later post. For now I'll just say that, contrary to the apparent consensus of the staff, I would prefer if ISML avoided round-robin tournaments altogether. I have yet to see data that suggests in general that ISML voters either prefer or expect a round-robin format for any aspect of the ISML tournament, nor have I seen any data that suggests that voters would dislike or reject major changes to the tournament's voting system. If anyone here has such data, please present it for discussion. Otherwise, perhaps the staff might consider asking about voting systems on the next ISML voter survey.

Or better yet, just do something completely different, regardless of the voter expectations and preferences, so long as it's convenient for the staff. As a voter, I don't want to demand so much time from the staff, and I accept that the staff may do whatever they please, regardless of voter preferences such as my own. But if I'm understanding the objective of ISML correctly, I hope the staff looks more skeptically at their present options and implements a voting system that is both convenient and satisfying for all involved with ISML, regardless of convention or precedent. More commentary to follow.

Interesting thing is, the round-robin may not be the most popular method among the staff. The trouble is, our voters probably likes the round-robin format better than the staff. According to comments we received in the survey, while the majority was silent about whether or not we should make a change, the vocal minority that wanted return to 2015, 2011, or even 2009 format far outnumbered those who wanted reduction or elimination of round-robin. Unless someone does a massive petition drive, I think it might be best to try a format that is somewhere between 2016 and 2015 format for the next two years while the the staff can make further improvements behind the scene and gain more experience with diverse situation. At the end of 2018 season, may be we should have another survey with questions more directly about the format. In order for ISML to abandon round-robin format, very thorough discussions about the alternative methods need to be done among the non-staff voters. This should help build up support for a viable alternative or two. If the support for the alternative becomes large enough to show significant change in majority's expectations in the 2018 survey, the staff would have the mandate needed to try out new methods that doesn't use round-robin.

[Toady]

I like round-robin tournaments, because it is very interesting to see each character fight against each other. The truth is I liked the 2015 and previous editions much more than the 2016, that I found quite boring until the top 16. Also, it was very complicated to follow.

The 2016 still featured round-robin format, but in the most boring way. Round-robin become interesting once you have a good number of contestants in the group. I'd say at least 10. Then you have the thrill of a hockey tournament (where each match is played a lot of times, which wouldn't make sense here ofc), not the boring start of the FIFA World Cup (or worse: the last version of UEFA EURO with 24 contestants). The possibilities are too few with only 4 players.

I don't necessary ask for the return to 2015 format. Option 5 looks nice to me. Or a version with more characters and at least 8 characters per round-robin group.

[maglor]

I like round-robin tournaments, because it is very interesting to see each character fight against each other. The truth is I liked the 2015 and previous editions much more than the 2016, that I found quite boring until the top 16. Also, it was very complicated to follow.

The 2016 still featured round-robin format, but in the most boring way. Round-robin become interesting once you have a good number of contestants in the group. I'd say at least 10. Then you have the thrill of a hockey tournament (where each match is played a lot of times, which wouldn't make sense here ofc), not the boring start of the FIFA World Cup (or worse: the last version of UEFA EURO with 24 contestants). The possibilities are too few with only 4 players.

I don't necessary ask for the return to 2015 format. Option 5 looks nice to me. Or a version with more characters and at least 8 characters per round-robin group.

Thank you for your input. Although it falls short of your 8 character minimum per group criteria, you may be pleasantly surprised at the dynamics that will likely happen in option 2 and 3 as well.

[kanohistorm]

#5, but you guys knew from day 1 i was against having an influx of characters

[Veon]

Option 1 was tried in this year. Wasn't too great.
Option 2 and 3 have many rounds that just feel pointless with the fact how many characters advanced to next round. Respectively 70% and 80%.

In this case option 4 and 5 are most interesting for me.
But certain thing about option 4 bothers me:

Round of 100: 10 groups of 10 characters, round robin; Aquamarine, Topaz and Amethyst Period. Each group 4th to 6th will be thrown into an arena where only top 10 will advance.Top 3 will advance from each group.

One arena with 30 characters? Too many... Easy to miss some characters.
Would prefer something like 5 arenas of 6 characters where top 2 from each arena advance.

Round of 40: 4 groups of 10 characters, round robin; Sapphire, Emerald and Ruby Period. The rest will be thrown into an arena where only top 4 will advance.Top 3 will advance from each group.

The same case. One arena with 30 characters... But in this case limiting to 4th to 6th places in groups would solve the problem.