Playtest Tournaments

[Rat With Wings]

approved by Birkal

Jumbao's playtest is the largest tournament that has ever in been hosted in CAP. This is excellent...right? Well unfortunately even with ~100 players sign up we only had 6 matches played in the first round with the rest of the matches decided by activity, coin flips and byes. This is not only unfair for the players who have to play and reveal teams but it also looks very bad and is certainly not what I have in mind when I think about celebrating the successful creation of another CAP. Pajantom's playtest was similar albeit not as bad but I'm afraid the trend will continue if nothing is changed.

In this thread I'd like to explore different ideas on how to fix this issue or decide whether this is a problem that even warrants change. In the past, playtest tournaments were completely ladder based and while I think going back to this formula isn't a good idea for many reasons, I think having a mix could work well....

My proposal is simply to change the playtest tournament into a ladder tournament with the play-offs played on the forums.
Example: Playtest ladder that goes for X amount of time and the top X move on and play the rest of the matches in standard forum based matches. With this proposal you get the best of both worlds. You eliminate byes and drastically diminish the amount of activity calls made (also if someone pulls out you can put in the next highest ranked person on the ladder) while still crowning the champion in the fairer standard tournament format. You also allow the meta to grow a lot more by having a ladder with greater incentive to play which makes for a more enjoyable finals series while also making it less about who has the most time to spare to ladder (much less time and effort required to make top 8/16/32 than first, obviously).

I'd love to hear critiques on this proposal and/or completely new proposals for how to tackle this issue.

 

[DetroitLolcat]

I don't believe we should go back to a ladder tournament. We had ladder tournaments in the past, they got pretty low activity and inherently favored people with more time to dedicate to power laddering. Ladder tournaments also put up a barrier to entry for people who want to playtest the CAP, since you have to play dozens of ladder battles in order to have a chance to even make the bracket. In the current format, anyone with interest in testing out the new Pokemon can do so; I'd rather have a tournament with 100 entrants, even if not everyone seriously plays it, than a tournament with a ladder component.

I agree we need to do something to promote activity in the playtest tournament, because the current activity level is abhorrent. I'm always a fan of positive reinforcement, maybe some sort of reward for whoever wins the tournament?

 

[Rat With Wings]

Ladder tournaments also put up a barrier to entry for people who want to playtest the CAP, since you have to play dozens of ladder battles in order to have a chance to even make the bracket.

Don't you still get to playtest the CAP by participating in the playtest ladder? Also I think having a barrier to entry for the main part of the tournament is unfortunately necessary. Without this you get people signing up who clearly don't read OPs then never get seen again... and while I get that CAP's ethos is that there is open participation for everyone, I think having a ladder tournament portion still targets that.

While I like the idea of a reward, and I think it would increase amount of sign-ups, I don't think it fully addresses the main issue of increasing the percentage of matches played in the earlier rounds because of the aforementioned users.

I agree that ladder tournaments inherently favour those with more time but I think it would be mitigated with 32 qualifiers. CAP ladders have never been so active that getting top 32 in a ladder that lasts for 1 or 2 weeks is a feat that requires an inordinate amount of time spent laddering. I can see there being logistical issues with this however. Checking the validity of the top 32 and then invariably having to check the users in the subsequent ranks because some in the top 32 are alts/don't want to play etc. could become an issue and perhaps more hassle than it is worth.

 

[G-Luke]

I think another way to go by this is by having a series of room tournament eliminations, in which based on scheduled times throughout the week, tournaments can be held in the CAP Project showdown room. After the room tours are completed, winners (and runners up, depending on how many people need to be involved) would then battle it out on a more traditional forum based battle. It allows new people who are interested to jump right in and battle, and since at the time alot of veterans would be around, they can help the newbies build on hand. Some challenges would include publicity, regarding how much people would actually come on, but based on how Jumbao's playtest went out, it is obvious that keeping in a playerbase would be an issue regardless. in the end great replays showcasing how good Jumbao is in the metagame would still be present, and at the playoffs we can have more "high quality" play of sorts.

 

[SHSP]

A discussion yesterday broke out into this topic so I figured I’d bump this a little bit with some of my thoughts.

I’m totally in favor of an OLT style bracket as proposed by Rat, either with one session as in his proposal or multiple sessions a la OLT itself. A ladder tour format fits a whole lot better in my eyes than a standard tour format for a few reasons:

Playtesting: Ladder is inherently a little bit “safer” for innovation and playtesting than a standard tournament. Sure, innovative, underappreciated sets will often catch wins in game ones of tournament bo3s, but the same format harkens itself to “cheese” strategies quite a bit that muddy the waters of, well, playtesting the mon. Ladder gives you room to innovate and test strategies that have more risk to them or are underappreciated: if you lose a game or two, you can get the points you lost back on ladder. If you lose two games in a tour set you are done, which leads into my next, and probably biggest, point.

Growth: The CAP ladder is widely considered a joke. You need to climb to mid-high ladder to realistically get meaningful games, which is often a slog through bad sets of 6-CAP teams and purely unviable sets. Wait times are long especially at rough times in other timezones (late night and early morning EST are notoriously dead from my personal observations of hawking battles to watch). For what is basically the entry point to CAP, alongside the showdown room, it’s not good for keeping users around or letting them develop. Picture this scenario:
A newer, younger user sees playtest and thinks it rocks, signs up, and builds a bad team or uses a sample that he doesn’t really understand. He waits on ladder for a while, runs into other similar teams that he beats, maybe meets one or two of the knowledgeable players who either beat him handily or struggle some being unprepped for, say, physical trace Kril. He makes some improvements to the team... and loses in a 2-0 thrashing in playtest bracket to someone who knows what they’re doing. They’ve learned nothing, have now lost in bracket and don’t really have anywhere to go. Ladder doesn’t let them get any better and if they stick around they struggle in single tours, get glossed over for CAPTT... the list goes on.
Changing the format to a ladder tour, I think, fixes this scenario and a common problem in two ways: For one, ladder, at least for a short time, is much more active with knowledgeable players. Names start to stick when you play people a lot or see people play a lot: When I started out, I think I queued into NitrousAcid four or five times in a row thanks to playing only at 6 AM est and struggled, but learned from that. Higher profile matches, the Reachzero vs Heaven Jays, are common at high ladder and easily viewable to people who have interest in them: if both have smashed their six cap teams before, I know I would want to tune in to see the victor and what they do against someone who’s their equal. Ladder being alive, especially at such a time as a new CAP release that naturally draws a crowd, makes CAP a lot more open to the public so to speak. Ladder wait times arent as long, and you’re not given two or three matches like that scenario; you can keep playing, winning, losing, learning.
For two, the ladder is inherently linked to one of the best places for new CAP players: the showdown room. The room often has a lot of players who know the meta idling, and a lot of auth that are happy to point people in the right direction. By starting on ladder and finding the room, you’re letting people ask about CAP: how we work, and from a metagame standpoint, how to make CAP work. It’s one question away from seeing the metagame subforum, analyses and sample teams, and lets people have back and forths with competent players you don’t see in a tournament thread. People can hear “oh, I think Scarf Kit is coming back” from an experienced CAP player and ask what it is; a little while later they’re getting teambuilding help to fit it on their team and improve the whole, talking about what works and what doesnt. It’s important to consider that we really do attract people in playtesting (see the sheer number of signups for Jumbao even though most never played), and that we do grow from keeping these players (from last Playtest, some good names to know are Againa [who I believe started out in the last playtest with top 4 but i’ll edit if I find out otherwise] who’s undefeated this CAPTT, PrototypeTrainer aka Xythiant who’s been helping me with stats and scouting for the Pajantoms, Deku Prince, who went from being a mainly ladder player to a CAPTT draftee with good success, among other names like Narusin12 and BoltSapphire who participated to good success but didn’t exactly stick around).

 

[Drapionswing]

This has my support, I really don't see anything wrong with at least trying this out. With the level of interests grabbed by the process, and redirecting it straight to ladder I think we have a shot of really improving a lot of things here. I think building a Pokemon, and not playing with it defeats the purpose of the CAP Process so to then encourage people to actively play it strenghtens that, while also building competitive knowledge for the future.

I do think Ladder Tours favour people with time, however currently I'd rather favour people with more time than have a repeat of the endless amounts of inactivity calls. Of course if you were to invest time into a Forum tour, it is less straining than a Ladder Tournament, however people who want to actually playtest the mon this is the best time to test because we hypothetically and hopefully have an active ladder.

 

[Birkal]

We really need to get something set here before CAP25 is completed. We are going to be staggering the releases of all three CAP25s, so that should also be taken into consideration. It seems like the popular opinion is to go with an OLT-esque format. We've also discussing on Discord a tournament that you can enter if you have X ladder reqs (GXE, Elo, or whatever we choose), you can sign up for the playtest tournament.

I think we can all agree that we need to change this, so let's pick a format and give it a whirl. Thoughts, Drapionswing, SHSP, Rat With Wings, G-Luke, DetroitLolcat, and others?

 

[Drapionswing]

We really need to get something set here before CAP25 is completed. We are going to be staggering the releases of all three CAP25s, so that should also be taken into consideration. It seems like the popular opinion is to go with an OLT-esque format. We've also discussing on Discord a tournament that you can enter if you have X ladder reqs (GXE, Elo, or whatever we choose), you can sign up for the playtest tournament.

I think we can all agree that we need to change this, so let's pick a format and give it a whirl. Thoughts, Drapionswing, SHSP, Rat With Wings, G-Luke, DetroitLolcat, and others?

I am in full support of an OLT style tournament. Week 1 we release a CAP25, then we ladder until we get a top x amount of players and so on until we have enough for a tournament.

e: The only thing I'm sceptical about is if the turnout will be as great? However I think it will solve the problem of activity calls.

 

[G-Luke]

I am in full support of an OLT style tournament. Week 1 we release a CAP25, then we ladder until we get a top x amount of players and so on until we have enough for a tournament.

e: The only thing I'm sceptical about is if the turnout will be as great? However I think it will solve the problem of activity calls.

A CAP 25? So all three will not be released at once for the playtest?

 

[Drapionswing]

A CAP 25? So all three will not be released at once for the playtest?

Yeah release them in dex order, Grass > Fire > Water which was the staggered release Birkal was referring to.

 

[G-Luke]

Yeah release them in dex order, Grass > Fire > Water which was the staggered release Birkal was referring to.

I assume that each addition would not removed in favour of the next one - in other words, when Fire CAP 25 gets added, Grass CAP 25 will still be allowed, because highlighting the three Pokemon's with each other and on the metagame on a whole should still be a vital aspect to explore.

 

[Birkal]

Drive by post — this is not a place to discuss and determine how CAP25 will be released. Let me make clear that this is still in flux, and reachzero will likely have the final say on how exactly we release them. Dex order release is also not going to be considered, as the 3D modeling team is much more important that following GameFreak tradition. It is hard to give anything solid, but talk with any modeler and they will agree that Fire is the most difficult model. They’ve got some great work already, but it will take time. Please watch the CAP main forum over the next week or two for an exact release schedule (and events!).

Back on topic, could someone write up an exact proposal for what this OLT would look like? Feel free to use official wordage.

 

[SHSP]

Back on topic, could someone write up an exact proposal for what this OLT would look like? Feel free to use official wordage.

on it, boss

Edited OLT Format Template-Ladder Cycles

1. The host of this tournament is (host) We will be using the regular Pokemon Showdown CAP ladder for this tournament, not a new one. This will ensure maximum activity and competition.

2. The qualification period for the playtest will run for a total of (number of weekly cycles) cycles with roughly (days per cycle) days per cycles. After each (insert) day cycle, the top (number of qualifiers) accounts that appear on the leaderboard (using ELO) at the end of the cycle with the sign-up criteria explained in #3 below will qualify for the playoffs of the playtest. Playoffs will be the format explained (link to the playoff format) The ladder is not reset after each cycle, only alts are (meaning you create a new alt after each cycle).

We are currently in Cycle (number), which means that after (insert) days [ending on (end time)], the top (number) accounts that signed up for this tournament following the proper sign-up criteria explained in #3 below will be qualified for the eventual playoffs of the Playtest.

3. In order to sign-up for the playtest, FOLLOW THE SIGN-UP FORMAT BELOW AND REGISTER YOUR FRESHLY MADE ALT ON POKEMON SHOWDOWN USING THE FORMAT BELOW:

(Code to identify participant, like "PT451") ForumNameHere - "ForumNameHere" obviously denotes your forum name. You are only allowed one alt per cycle, which must follow the aforementioned format.

YOU MUST SIGN UP WITH YOUR FRESHLY MADE ALT BEFORE STARTING BATTLES WITH IT. You will not be allowed to edit this alt after you include it in your initial sign-up for any given cycle. By the way, make sure to register your playtest alt on Pokemon Showdown!. We've had to deal with people who got their account registered by someone else because they were unprotected. It is advised that you once you sign up here, you immediately register the accounts on PS afterwards.

4. Once you make the top (number) and qualify (this does not mean making the top (number) at any given point during the cycle, it means being in the top (number) at the end of the cycle), you’re qualified for playoffs. You are free to ladder any other week you wish. However, making the top (number) more than once won’t give a player anything other than personal satisfaction and bragging rights. If a person qualifies again during a week after their initial qualification, the newly qualified account will be skipped and the next unqualified account will be selected from the leaderboard to participate in the tournament.

That was the easier part of the rules, cause its minor changes, effectively. Lots of blanks to fill in, but the structure remains the same. Minor note: OLT started using a system where instead of registering multiple alts with "OLT2C1" or whatever the code was, they give ONE alt with a specific code like the "451" I spitballed. This prevents people from sneaking by with pre-registering accounts knowing the code beforehand. Now, for the difficult part: Playoffs. I find it unlikely we'll have 8 qualifiers per Cycle, with a smaller community, so I think the need for the most recent OLT change of a Swiss play-in to a single elimination bracket is less effective and necessary. The larger the number of players, the faster the tournament runs with Swiss compared to Double Elim, so it's an idea to keep in mind I could work out (and not to neccisarily rule out Single Elim either, for what it's worth). For now, a Double Elim description:

Double Elim OP

Welcome to the Playoff component of the CAP 25 Playtest! Here, the top (number of) players that qualified through the ladder component of the tour will battle in a double elimination bracket to determine our playtest champion! Here's how it'll work:

In round one, the qualifiers will be placed in a randomized set of pairings and play a Best of Three CAP set. From there, winners will go on to play other winners in the Winners Bracket, and losers will go play other losers of round one in a Losers Bracket. If you lose in the Winners Bracket, you fall to the Losers Bracket. If you lose in the Losers Bracket, you are eliminated from the tour. Here is an example of a 16 man Double Elim tournament, showing how the Losers Bracket works: if you lose in a Winners round, you are dropped to that Losers round, to play someone who climbed to that round in the Losers Bracket.

Remember, all matches are to be played in a BEST OF THREE format, and all standard CAP rules apply.

if need be, we can fit in rules from a Seasonals OP or past Playtests about the tour as a whole, like how "if something's banned midway through a round it's legal during that round" and the like, but I feel this is the most important part to nail down, as there's a possibility for people to not know how the tour phase works (and clarity is always nice).

 

[SHSP]

Bumping this: cap 25s are freshly out and last we talked was this proposal above for an OLT-like format. Whatever we decide should probably be decided rather quickly so we don't stagnate this step of the release. My opinions still haven't changed, still very supportive of a ladder tour, hell if we need a host I'm absolutely willing to step up and host.

edit: a note, as brought up by Frostbyt in discord a bit earlier- throw in a rule against privatizing games (like /ionext), and note that if people try to game the system it's not going to end well for them (somehow this slipped my mind earlier, thanks again frost)