Squidboards Ranking System Proposal

[Box]

Re: Team Sizes
I'm not convinced that you can really increase the number of matches being played by splitting bigger teams up. I think larger groups on the website will split up naturally into smaller teams unless they just don't have enough active members to form a second team. If you really want to put a hard limit on the number of players on a team, these are two ways to do it.:
1) Estimate the number of players after which a team doesn't benefit from adding another member. Make that maximum that number.
2) Estimate the number of players for which having one fewer member would significantly hinder the team. Make the maximum that twice that number minus one so that if you go over the maximum, you can comfortably split into two teams.
That's at least methodology with some reasoning behind it. Method 1 will probably get you more small teams. Method 2 is more accommodating to the big groups.

Re: Seasons
By splitting up into seasons, you record which teams were the best over time. It also allows for a natural point where you can review how the metagame has evolved and see about changing rules for the next season. In this current system, you can actually hang onto your rating just by playing a match once in a while. There's little need to continually prove yourself against an evolving metagame. Elo doesn't handle inactivity very well which is why I imagine you included an inactivity penalty in the first place.

Re: Tiers
Not telling people what their actual ratings are is really opaque. You also lose information by not including the numbers. It won't help anyone understand or respect your system. Numbers are actually really easy for people to understand. You can look at the difference between two teams' Elo ratings and understand the difference in their estimated skill.

 

[Flying_Tortoise]

Re: Team Sizes
I'm not convinced that you can really increase the number of matches being played by splitting bigger teams up. I think larger groups on the website will split up naturally into smaller teams unless they just don't have enough active members to form a second team. If you really want to put a hard limit on the number of players on a team, these are two ways to do it.:
1) Estimate the number of players after which a team doesn't benefit from adding another member. Make that maximum that number.
2) Estimate the number of players for which having one fewer member would significantly hinder the team. Make the maximum that twice that number minus one so that if you go over the maximum, you can comfortably split into two teams.
That's at least methodology with some reasoning behind it. Method 1 will probably get you more small teams. Method 2 is more accommodating to the big groups.

I see what you are saying, but do you believe a team will benefit from having more than 7 players? If one of your players is inactive you can always get new players when the period for changing players is available. I honestly can't see more than 7 being useful or fun in a match especially if you are a player in a big team that wants to be playing. Also having more than 7 players allows for players to be carried by teammates who are really good and would mean their ranking might not be at their true skill level.
As for the 2 methods we do believe we have tried to keep this in mind, but we will continue to try to do this and we are definitely still open to new perspectives.

Re: Seasons
By splitting up into seasons, you record which teams were the best over time. It also allows for a natural point where you can review how the metagame has evolved and see about changing rules for the next season. In this current system, you can actually hang onto your rating just by playing a match once in a while. There's little need to continually prove yourself against an evolving metagame. Elo doesn't handle inactivity very well which is why I imagine you included an inactivity penalty in the first place.

I do like your point about allowing an opportunity for rules to be changed, I am not sure if it is too much of an issue but I'll definitely bring it up to the group.

Re: Tiers
Not telling people what their actual ratings are is really opaque. You also lose information by not including the numbers. It won't help anyone understand or respect your system. Numbers are actually really easy for people to understand. You can look at the difference between two teams' Elo ratings and understand the difference in their estimated skill.

Yes the rankings are very easy to understand, however that isn't the issue we were trying to avoid. We were trying to avoid encouraging teams from seeing their number and trying to push their ranking that little bit into another subdivision. At first this appears to not be an issue but it is, and has been shown in many communities. Of course it isn't liked by anyone, I myself would like to know where exactly I am at however I definitely see the point to not giving people their numbers. CS:GO was a good example for this, but yes we are aware that other gaming communities do show numbers. However we also looked into smashladder.com as they are a great website that does Elo rankng for Smash games. They initially used to give numbers to their players however they changed it to merely give people their divisions due to problems they were facing because the players knew their numbers.

Thank you for bringing up your points as they are a new perspective :)

 

[StarForce]

I see what you are saying, but do you believe a team will benefit from having more than 7 players? If one of your players is inactive you can always get new players when the period for changing players is available. I honestly can't see more than 7 being useful or fun in a match especially if you are a player in a big team that wants to be playing. Also having more than 7 players allows for players to be carried by teammates who are really good and would mean their ranking might not be at their true skill level.
As for the 2 methods we do believe we have tried to keep this in mind, but we will continue to try to do this and we are definitely still open to new perspectives.

This post seems very organised how it was done with all the correction last time I saw it.

As for the limit I also aprove of 7 as due because I also wanted that when I started my Squad Thread which is a part of the SCO now. While Squads have grown out very big (some of them) I have not objected to it. It was just a table note to expand on it. But even as they form and grow big in the end when participating it's just 4 vs 4 by the game. While teams of 4+ form ultimately it is the game itself who decides it's 4 not 3 or 5 players. This is why I stand on 7 as the best number. My main concern and a point I will always support is that People like to play. There is no other simple way of explaining it. I f I am part of a team I want to play. Being part of a say 10 to 12 member team and the game allows only 4 at a time I will grow hasty and uneasy that I can't play. I want to play all the time so this is the problem of having bigger teams. Of course substitute players are needed but there is a point were you just ad too many to a team. Some people can't always play because nobody plays to lose and have F or 0 points all the time. People play and compete to win. Therefore teams in the game with their leaders will send out the best of players, those who are not very good stay on the bench like in real sports.

I personally want in teams for everybody to be active and have fun so a minor number such as 7 is perfectly fine. The relationship of a team with less members is also more beneficial which is the basis of my Squad; Inkzooka. The less people the more everyone is aquainted with each other. you developt more bonds of trust as friends and playmates. It's easier to spar and your teammates respond better to your playstyle. I don think the US Army, or the Eurpean Naval or maybe The Japanese Militia go to a camp and all sit down asking all members to know each other. A smaller team is also much more manageable for the leader and offers less complications. This is my take on a small team.

As for tiers. I think we have too many but we also need to think that we are starting so going out to the stret (in this case the boards) for feedback is the best. Splatoon is quite a hit and I'm sure many players play it daily to back up the many iers. Some are just waiting on August as it's the best settup for friends. I personally like seeing things acoomodate fairly well in tiers and that's how I understand them of course numbering people or teams from 1st to 500th might be an idea but it is too simple and offers not much diversity. I prefer a more detailed style like tiers. we'll jut let the community decide on the numbers.

On the weekly-bi weekly publish I have to say it seems fine but maybe weekly. Several matches can come up in a single day. It may be a bit of hassle to handle the info daily so weekly is the best option. Bi weekly seems fine to some extent but I'd rather see the results show up faster. The faster the results come the easier it is for teams to challenge others which will help speed up the challenging and matchmaking. Teams are also eager to show up how good they are and it makes a good timelapse for those who want to show how much they progress to tothers while it's a good wtime too for those not wanting to show the loss so eagerly haha. I'm up for weekly, but no more than two weeks.

 

[Kbot]

Well, we're currently working on automating the system so that it requires significantly less work on our end to run. We're getting some outside help with that and hopefully it will work out. And even if that falls through, we've got a dedicated team of people who will run all of the behind the scenes stuff, and we can always get more volunteers from the community. We plan on publishing the ranks regularly (but not real-time). Probably weekly or bi-weekly. As for it being integrated into the website, that's not in our hands. That's up to the admins.


Well there's nothing we can do if it's the second thing, so we're just assuming it's the first. If it does pair you randomly, there would be almost zero competitive community for this game, and that would suck.

Sounds like you guys have got everything under control. Just voicing my concerns. Thanks for taking the time to come up with something like this. Cause I know I sure wouldn't have the time or patience to come up with a system.

At this point, I'd launch it and see where it gets us. There's always room for improvements later, right? Just see what kind of feedback you get from the community and roll with it.

 

[Hinichii.ez.™]

8 people prettyplz

 

[StarForce]

8 works for two teams. Still open I believe.

 

[Flying_Tortoise]

Sounds like you guys have got everything under control. Just voicing my concerns. Thanks for taking the time to come up with something like this. Cause I know I sure wouldn't have the time or patience to come up with a system.

At this point, I'd launch it and see where it gets us. There's always room for improvements later, right? Just see what kind of feedback you get from the community and roll with it.

No problem at all, it definitely would be a tall order if we didn't automate it. Because we are still in that process, it will take a little bit of time until we release it to the community. Perhaps Mid July (it should be done a lot sooner, but just in case) if everything goes according to plan as we are talking to the person who is helping us atm. Thanks for your post.

8 people prettyplz

Interesting :). May I ask why? I am assuming it is because you are worried of inactive members and that if you split you risk losing both teams?
Well first off, I'm pretty sure we could allow on a case by case basis some exceptions to rules, such as team count. I would have to ask the group first, but if your reasoning is sound then I don't think there should be a problem. May I also say that should inactive members be a problem you could always find new members through the Competitive Squad Searching Thread in the Competitive Section, so I'm sure you could find more than eight members and then form 2 teams.
But again I'd love to hear your input as I was just assuming your reasoning.

 

[grandpappy]

I'll tell you why I'd personally like to see 8 members on a team. Team practice matches. With 8 members you can organize team practice times and you won't have to pull somebody from somewhere else just to make it happen. I don't know if that matters to anybody else, but I think it will be important to have these team practice matches to have everyone on the same page and to get used to playing with each other. And if teams have to pull somebody else from one of the other teams on their squad or from elsewhere, I just think it will be inconvenient (and potentially not that beneficial to the one random person). I'll agree that having any more than that will probably not be necessary.

Other than that, I think everything sounds pretty good to me. I can see why some people might have issues with the one team limitation, but I think it's fine and it will work especially with the two week team roster change thing you guys have proposed.

Thanks to you guys for the hard work, and for sharing this with us. Looking forward to the future.

 

[Box]

I see what you are saying, but do you believe a team will benefit from having more than 7 players? If one of your players is inactive you can always get new players when the period for changing players is available. I honestly can't see more than 7 being useful or fun in a match especially if you are a player in a big team that wants to be playing. Also having more than 7 players allows for players to be carried by teammates who are really good and would mean their ranking might not be at their true skill level.
As for the 2 methods we do believe we have tried to keep this in mind, but we will continue to try to do this and we are definitely still open to new perspectives

I really don't know if a team will benefit from having more than 7 players. 7 might even be too much if all of your points apply to teams of 6 as well. On the flipside 7 might not be enough because if you want to add one more, you have to split into teams of two teams of 4 which means you go from having 3 subs to 0 subs which might discourage the split in the first place.

I do like your point about allowing an opportunity for rules to be changed, I am not sure if it is too much of an issue but I'll definitely bring it up to the group.

And here I thought I might actually be talking to "the group".

Yes the rankings are very easy to understand, however that isn't the issue we were trying to avoid. We were trying to avoid encouraging teams from seeing their number and trying to push their ranking that little bit into another subdivision. At first this appears to not be an issue but it is, and has been shown in many communities. Of course it isn't liked by anyone, I myself would like to know where exactly I am at however I definitely see the point to not giving people their numbers. CS:GO was a good example for this, but yes we are aware that other gaming communities do show numbers. However we also looked into smashladder.com as they are a great website that does Elo rankng for Smash games. They initially used to give numbers to their players however they changed it to merely give people their divisions due to problems they were facing because the players knew their numbers.

I can't really tell what you think the problems were with those other communities, but they sound like problems with the fact that Elo doesn't really encourage ladder play. It's just an estimation tool. You're not asking teams to demonstrate skill by winning sets. You're just using a formalized way to guess how the teams' skill rankings match up against each other. Hiding the numbers doesn't really change that. I think it's better to either accept the limitations of Elo or try to find something more appropriate for a ladder.

 

[D3RK]

Is it possible that each squad could have 2 teams? Lets call them A and B to keep it simple. As other have said it would give you 8 players that you could practice with freely while 7 would give you issues in practicing with all your members unless you have a friend or someone outside of the squad to fill in. Having everyone play together will build teamwork regardless of which team you are on since you will see how everyone in your squad prefers to play and can offer feedback. If all 8 members are free on a given day then you could enter 2 teams and if you are down a few players then Team A and B could send the top 4 that are available and make 1 team from those players. This also prevents benched players that are just sitting around because they are simply "outclassed" by the better players in the squad.

8 members would allow you to test a lot of things freely such as "What happens when we run into a team with "X" setup? Will we be able to deal with it or do we need to change "Y" in our setup?". Team A could be the team learning the match up and Team B could switch to the load outs that Team A are trying to beat. This would be a good learning experience for both teams. If Team B does end up beating Team A then Team B can go over what was giving them the advantage and what Team A could do differently to deal with the setup. Win or lose you are helping the squad in some way. Another unique thing with 2 teams is that you can trade players around for weapon variety.

I think that 1 extra person is a small change that would make a huge difference for the better. One issue I could see though is ranking both teams could be a little difficult.

 

[Aweshucks]

And here I thought I might actually be talking to "the group".

Well you are, kind of. @Flying_Tortoise and I are speaking here on behalf of the rest of the group. There are currently about 10 to 15 people who are working on this project, but we asked them to refrain from posting in this thread. We felt that it would be easier and simpler for everyone involved if it were only the two of us responding here. Everybody else is still watching the thread and if they want anything said they can ask tortoise or I to post it on their behalf. When tortoise said that he would bring it up with the rest of the group, I believe he meant that he would bring it up at our next meeting so that we could seriously consider adopting it. We will be doing that for everything that someone brings up in this thread, no matter how inconsequential we think it may be.

Is it possible that each squad could have 2 teams? Lets call them A and B to keep it simple. As other have said it would give you 8 players that you could practice with freely while 7 would give you issues in practicing with all your members unless you have a friend or someone outside of the squad to fill in. Having everyone play together will build teamwork regardless of which team you are on since you will see how everyone in your squad prefers to play and can offer feedback. If all 8 members are free on a given day then you could enter 2 teams and if you are down a few players then Team A and B could send the top 4 that are available and make 1 team from those players. This also prevents benched players that are just sitting around because they are simply "outclassed" by the better players in the squad.

8 members would allow you to test a lot of things freely such as "What happens when we run into a team with "X" setup? Will we be able to deal with it or do we need to change "Y" in our setup?". Team A could be the team learning the match up and Team B could switch to the load outs that Team A are trying to beat. This would be a good learning experience for both teams. If Team B does end up beating Team A then Team B can go over what was giving them the advantage and what Team A could do differently to deal with the setup. Win or lose you are helping the squad in some way. Another unique thing with 2 teams is that you can trade players around for weapon variety.

I think that 1 extra person is a small change that would make a huge difference for the better. One issue I could see though is ranking both teams could be a little difficult.

Yeah, my problem with that is that is messes with the system, in which teams are clearly defined, and having two teams merging sometimes would be difficult to implement. That said, there have been plenty of great arguments for changing the max number of players on a team, which will definitely help us as we reevaluate that rule.

 

[teamdelibird]

I think the number should definitely be 8. If nothing else it allows for a "turf wars" team and a "splat zones" team.

 

[Aweshucks]

I think the number should definitely be 8. If nothing else it allows for a "turf wars" team and a "splat zones" team.

If we are going to increase the number, it will almost certainly not be because of that. A team should be adept at any gamemode they would have to play, and one of the reasons that the limit exists in the first place is to prevent multiple teams from registering as one team. And we know of at least two more gamemodes that are coming out anyway. Also, I've seen many people saying that Turf War isn't suited for competitive play anyway, and I'm inclined to agree (Though this is something we will discuss later when we are deciding on a ruleset).

 

[teamdelibird]

If we are going to increase the number, it will almost certainly not be because of that. A team should be adept at any gamemode they would have to play, and one of the reasons that the limit exists in the first place is to prevent multiple teams from registering as one team. And we know of at least two more gamemodes that are coming out anyway. Also, I've seen many people saying that Turf War isn't suited for competitive play anyway, and I'm inclined to agree (Though this is something we will discuss later when we are deciding on a ruleset).

Bullshit. Turf wars is just as competitive as any other mode, the only thing that separates it in competitiveness from splat zones is that splat zones has built-in ranking.. And anyway you're right, it is a bad argument, but it seems like everyone wants 8, it just seems like it's a number that makes more sense. Why not just do it? I mean, if two teams are gonna register as one, it's not like having 7 is gonna stop them from having two teams that just happen to share one player.

 

[Aweshucks]

Bullshit. Turf wars is just as competitive as any other mode, the only thing that separates it in competitiveness from splat zones is that splat zones has built-in ranking.. And anyway you're right, it is a bad argument, but it seems like everyone wants 8, it just seems like it's a number that makes more sense. Why not just do it? I mean, if two teams are gonna register as one, it's not like having 7 is gonna stop them from having two teams that just happen to share one player.

Well, I think that turf wars is too swingy. Winning is predicated on what the field looks like when time is out, and doesn't take into account anything that happened before that. It's a fun mode and I play it a ton, but I don't know if it has a place in a competitive format. But that's something that can be debated at a later time and I'd like for the discussion here to stick to the actual post right now.
And yeah, everyone here has definitely convinced me that we need to take a much closer look at the number.

 

[Egregore]

I like the general idea. I do hope the squad member limit can be increased to 8, though. I feel that this will allow for more of a versatile roster, and also would make sense for practice sessions.

If you need some more input on the matter, I'm more than happy to chime in. I'm a long time FGC and Smash community member and TO.

 

[Box]

Well you are, kind of. @Flying_Tortoise and I are speaking here on behalf of the rest of the group. There are currently about 10 to 15 people who are working on this project, but we asked them to refrain from posting in this thread. We felt that it would be easier and simpler for everyone involved if it were only the two of us responding here. Everybody else is still watching the thread and if they want anything said they can ask tortoise or I to post it on their behalf. When tortoise said that he would bring it up with the rest of the group, I believe he meant that he would bring it up at our next meeting so that we could seriously consider adopting it. We will be doing that for everything that someone brings up in this thread, no matter how inconsequential we think it may be.

That makes it kind of awkward for me. I'm trying to have a dialogue about some of this stuff and I can't even directly talk to people. It doesn't really do anything for me to know that it'll be brought up in your committee meeting. And yes, I know you offered to include me in the private conversation, but I'm thinking about everyone here. Moving the 'real talk' to a private area doesn't benefit the people who were in my situation at the start of the thread. People should have the opportunity to jump in and offer their thoughts as the conversation develops.

How seriously are you going consider adopting suggestions that you consider "inconsequential"? That's not a reasonable expectation to have of yourself. I suspect that "the group" means well and wants to make everyone feel like they've been considered. But by putting a liaison between yourselves and the conversation, you're giving yourselves a lot of leverage to ignore people. The right thing to do here if you want to get people involved is engage in the conversation and not hold a private meeting to debate the merits of people's suggestions.

 

[Aweshucks]

You are talking directly to people. Tortoise and I are perfectly qualified to speak on behalf of everyone else on the team. And as I said, others on our team are watching this thread as well and can ask me to post whatever. This is just more streamlined and simpler.

and by "inconsequential," I meant small changes, not ones that we don't care about. We will be going through this thread and looking at any changes suggested. This thread is the conversation you're referring to in your last sentence. We want community feedback and are asking for it. We want to do something that the meets the community's expectations. But, like I said before, simply going on the forums and saying "hey guys, let's make a ranking system for squidboards" might come up with a few interesting ideas, but ultimately nothing would come of it without structure. We've taken it upon ourselves to get this up and running. So while we do want to hear what the community wants and will certainly do our best to meet their expectations, we will have final say because we will be the ones running it.

This thread is about the system that we plan to have here on squidboards come August. I'm glad that you're critical of the plans, but I'm going to ask that you stick to criticizing the plan rather than our methods of interacting with the community

 

[StarForce]

It's also to notice that we are here to receive feedback because we want to improve. That said there is some etiquette to follow so accepting or simply implementing all the idea everyone will have is not what feedback is. So we can't either accept or implement every single whim everyone desires. Chances and feedback are tied to mass audience. If 60 want X and 40 want Y, X will win because it's what the most people desire. This conversation is the best way to hear the community. But I do get your point. It is valid in the long run.

 

[DaisyFan]

Very looking to see a league or a ranked system to be honest, this game can be a lot competitive.