Proposal RE: OU Retest Timelines/Implementation

[Finchinator]

When we retest a Pokemon in OU like we just did with Palafin, the metagame shifts dramatically. People's first ideas for checks and counters are spammed, initial usage of the retest recipient seldom resembles what it settles into, and any prior balance is thrown into disarray in favor of a singularly focused metagame state. The deeper we get into retests, the closer we get to normalcy and perhaps balance depending on the retest itself. This makes it so that people who get reqs or ladder during the first few days have an uneven experience, especially relative to those who get voting reqs later on in the process.

134 people currently have Palafin reqs in OU, but I guarantee you Star's experience finishing reqs today varied from Lily's experience getting reqs nearly two weeks ago. I am not trying to discredit Lily's reqs or the experience of anyone who got them early either -- Lily has been on of our most vocal and passionate council members to the point that I sometimes trust her more than I do myself. It is just a matter of vastly different circumstances and a forcibly volatile position playing out over abbreviated samples.

This is not the fault of either of them as we encourage everyone to participate in suspects and we only have so much time after all, but it does leave us with different understandings and potential for skewed results. We do not have retests frequently enough to have these discussions without tangible inspiration like this retest I admit. I wish that I could do some deeper research like I have with regular suspect tests as we have seemlessly adapted them in various fashions this generation, but this simply is not possible. After witnessing the wide spectrum of responses Palafin has received and receiving feedback from metagame participants, I have drawn the conclusion that we need to reevaluate how we conduct retests in SV OU.

My proposal is as follows:

• Make retests closer to 3 weeks than 2 weeks, which is the current standard
• Prohibit people from getting voting reqs during the first 3-5 days of the retest while still allowing the retested Pokemon on the ladder
• Formally "open" the suspect to getting reqs for the last 2 or 2.5 weeks of the retest
  • Have discussion threads mirror this timeline as kneejerk posts have similar issues

I can technically do this without a PR thread or any approvals as we can set suspect timelines as we see best fit as councils, but this is a unique circumstance and I have limited experience in this area as a leader, so I would like community feedback on this proposal. While the pros of this are outlined above in my post, I will admit one main con is actually being able to fit three full weeks (plus voting time) into a messy tiering calendar that could have other suspects, game releases, and awkward tournament overlap factoting into our internal calculus. I do not know if this is just a downside of retests altogether or enough of a problem specifically with the extra time that it should prohibit it, but I am all ears.

Please provide any feedback you have on this proposal or how we conduct retests altogether. Thank you

 

[Tio Chico]

First, thank you for your transparency and for your constant attempts to optimize our experience with the game. Unfortunately, there was no success, but there was certainly learning. You did not hesitate to admit mistakes, things like QB Volcarona DLC1 being premature, and you are always open to listening to the community and creating several PR debates like this one.

On the topic, for Retest, I broadly agree with a longer period of three weeks. But I raise the question of whether it is really necessary to block the requirements in the first few days. I believe that players who get the requirement early, for the most part, continue to engage with the metagame from other points of view besides obtaining the requirements. So depriving someone who at that time has more free time to spend on the game may be unnecessary. I have read some reports of players saying that two weeks was not enough to dedicate themselves to the game due to some X factor IRL and I've read about players changing their minds during the suspect/after the reqs.

I believe that most players, when typing between Keep Ubers or Unban, will think about the entire trajectory of the suspect, regardless of when they obtained their requirements.

We can debate a middle ground with something that already exists:
The discussion thread for a new suspect is closed for 24 hours and only then can players post their initial impressions. Blocking reqs for 24 hours could be a good first step.

 

[Magcargo]

Agreed, suspect period should be 3 weeks minimum, if not a month, for retests. My personal issue with these 2 week suspects are that a) most of the teams on ladder do not use the suspected Pokémon in question early on (see Roaring Moon or Deo-S retest in ND) so it makes difficult to gauge if the suspected Pokémon is even having an effect on the metagame & b) 2 weeks are not enough for new metagame adaptations to take place, which can both make it difficult to tell if the suspected Pokémon is broken (as comps are not optimized with it in mind) or allow sufficient counterplay to develop to the Pokémon during the retest.

I also see merit to preventing players from getting reqs early for retests.An issue for retest suspects is that most players are not using the suspected Pokémon, especially early on, so having a grace period for the Pokémon effects to settle on would be a better way to helping suspect voters see the effects of the Pokémon. In a tier like ND, I did not engage with the tier after I achieved my reqs for the roaring Moon suspect, so I didn't know what effect it was happening on the metagame after I achieved my reqs (where i barely faced it). I think a grace period will make it more likely that voters engage with the suspected Pokémon in the tier.

 

[Ehmcee]

Honestly, feels like the simplest proposition for this could be:

- Have a 3 week retest period
- The first week has the retested mon unbanned
- The 2 week suspect test begins when the first week is over

This gives some time for the meta to shift around the pokemon, I don't believe we need to shift the amount of time suspects take, but for a general retest I would suggest the additional week at the beginning.

 

[Alternator]

Three week suspect period: one week for people to spam their initial ideas and massively overprepare for the tested element, by the end of the first week teams have begun properly adapting, start the two week period to earn voting requirements then.

It keeps the timeline extremely simple and clear, which will minimize the inevitable "Why didn't my alt qualify?" "You started them four days in and voting reqs opened five days in," bickering.

 

[N_Mareanie]

Make it a month/4 weeks imo. Realistically speaking you need like two months or more to really determine brokeness or effect on a meta, esp with a mon on the level like Palafin, but that would probably be a bit too long of a period, but 3 weeks doesn’t seem that much more effective than 2 weeks, although a good improvement nonetheless. whatever gets implemented i support a longer period of retesting pokemon

 

[FayaWizard]

Make it a month/4 weeks imo. Realistically speaking you need like two months or more to really determine brokeness or effect on a meta, esp with a mon on the level like Palafin, but that would probably be a bit too long of a period, but 3 weeks doesn’t seem that much more effective than 2 weeks, although a good improvement nonetheless. whatever gets implemented i support a longer period of retesting pokemon

I think 4 weeks is pushing it too far: one week was enough for the metagame to start settling down, and while it’s true that we can’t fully explore a Pokémon’s effect on the metagame for a suspect, the two week period of this suspect was enough for its preliminary effects to become clear, and I don’t think we need to double suspect lengths for this. I support the one week grace period, then two week suspect test idea suggested by Ehmcee.

 

[Pais]

My proposal is as follows:

• Make retests closer to 3 weeks than 2 weeks, which is the current standard
• Prohibit people from getting voting reqs during the first 3-5 days of the retest while still allowing the retested Pokemon on the ladder
• Formally "open" the suspect to getting reqs for the last 2 or 2.5 weeks of the retest
  • Have discussion threads mirror this timeline as kneejerk posts have similar issues

I understand the appeal of this idea, but I believe it's more effective in theory than in practice. I also see some potential drawbacks:

• SV OU, in particular, is a metagame that shifts rapidly, almost daily. Players who grind both high ladder and tournaments can attest to this. Therefore, I don't think a larger suspect window would significantly improve the situation.
• The ladder plays a crucial role in testing teams for tournaments. I'm not entirely comfortable with the idea of playing on a ladder that doesn't accurately reflect the current metagame, especially since you can play with these retested Pokémon in case of a retest, particularly for a current generation. Sometimes having a larger suspect window can conflict with important tournaments. For example, right now we have both SPL and OST. Imagine being a tournament player and testing your teams in a ladder where a Pokémon is undergoing a retest for a long amount of time, it's not optimal.
• I don't believe prohibiting suspects in the initial days is the most effective approach. Not everyone would be motivated to play the meta without the possibility of quickly obtaining reqs. Players could easily circumvent this by starting to ladder later in the week.

Instead, I propose the following solution:

• Maintain the current two-week suspect window.
• If a retested Pokémon gets unbanned, implement a system similar to older generations (e.g., Latias in DPP). This would involve initiating a new suspect test for that Pokémon after a specific period of time, to confirm if the pokemon which got unbanned can stay in OU or not.

On a side note, I don't believe missing the possibility to get a Pokémon unbanned from Ubers to OU is as significant a loss compared to the possibility of getting a Pokémon banned from OU to Ubers in a suspect test and not everyone realistically would overthink so much on a suspect and statistically it's not likely people change easily their opinions in two/three weeks. Therefore, a two-week window for suspects is not a major concern.
Even assuming a potentially balanced Pokémon gets unbanned, there are still drawbacks to consider, such as the need for players to re-adapt to the new formed metagame.

 

[N_Mareanie]

while it’s true that we can’t fully explore a Pokémon’s effect on the metagame for a suspect, the two week period of this suspect was enough for its preliminary effects to become clear

preliminary effects are not a 100% clear case of what should be voted and what the true effects are - once a mon drops, ofc the meta is gonna be reacting. it may take a bit before a meta evolves to where you can really determine brokeness, see the start of a generation (except mons like flutter mane or the like). there have often been times where a mon seems broken/unhealthy because of initial effects on the meta, but the meta comes to adapt to it and the later it’s not a problem anymore. see gholdengo always being slated but never getting suspected.

i dont think it’s entirely fair to judge a pokemon by preliminary effects when they may very well change, for the positive, or a negative, i dont know, depends on the pokemon. but agree that at one points it starts to become “too long”, and obs u cant roam something like dialga around for 4 weeks or more when it’s clear that it’s broken (tho, let’s be fair, palafin was not clear broken.) maybe there’s an inherent flaw with suspecting it on ladder. i wouldn’t know an alternative though. but 4 weeks does not seem that out of line in a case of retesting

 

[Mada]

How doable would be trying or implementing new methods to suspect Uber mons into OU? I think the concerns in the OP are very valid, but as mentioned there and by other users there are drawbacks to most solutions. I am not familiar with how or when suspect process was created and started using, but maybe we could implement something like doing a No John Tour with the banned Pokemon, starting it before the actual reqs are ottainable? Or using a new suspect ladder to test this Pokemon, so that having an extended 3-4 weeks test would be less of a problem? Although I remember there being some policy problems about "suspect ladders" so I am not sure if something like that would work, let alone that activity could be a problem. Overall I think the suggested solutions would be a good improving of the current system, being able to get the reqs after 5ish days of the mon being freed and expanding the suspect by a week. I think that discussion threads shouldn't be opened later, as I think that brainstorming and discussing with other people helps you getting a better idea of the Pokemon in the metagame.