Akir
A true villain!
The first DPP PU tournament was done by Honko, and was a fairly sleepy-but-fun tournament held deep in the tournament forums. The second DPP PU tournament was made by me, and I was expecting another sleepy tournament to burn some time before gen7 arrived for PU. I was very wrong, so now I have to make this thread. I think it is important that these tiers are formally addressed: for closure for the people who devote so much time to them, to explain what they are to those who do not know, and to map out their futures.
Currently the classification for DPP PU is: "created tier". This is not an ideal name imo so I will be calling them retro tiers, as they were made retroactively. A retro tier is a tier that was made in a generation after the generation had already come to a close. A regular tier has to fulfill these requirements:
So then there's the question of "well if they aren't made in the usual way, how are you making these tiers?" Well great news! Making lower tiers in ADV and older is incredibly easy, all you need is a council of people from the above tier to give a VR cutoff to make the tier and done! Incredible. This can, in theory, go on and on until the generation runs out of mons...although I personally don't have much interest in going "too low" where there is not an interesting meta to create. The usage retro tiers are created in a different way, a way that I spell out right here in this post. If you do not want to follow that link, or if it gets deleted fsr, I will spell it out again in this hide:
While this process might seem clunky at first glance, the results from it are actually phenomenal. Both ladder usage and tournament usage are flawed in different ways (ladder usage favors samples and the overly-eager singular users, tour usage is a much smaller sample size prone to seasonal trends), but combining the two it is possible to find some common ground and make a more accurate picture. After that, the above tier has to weigh in and knock out the more nuanced borderline mons, to iron out the last string of flaws that the system has. The current version of DPP PU was created with only ladder usage, and has gotten negative feedback from DPP NU players in terms of accuracy. Applied to DPP PU, the system works like this:
This system actually works so well that I plan on redoing DPP PU in order to follow this system and make precedent. Assuming there aren't a ton of objections, of course. I want to make a dialog above basically anything else.
So great, that's the what and the how, and here is the why this thread is necessary: They are fun, they are already made and growing, and I doubt that they will stop being created. I think it is important that we address these tiers and their playerbase at some point and talk about how we are going to both create these and address them in policy, and before another generation launches seems like the appropriate time. Here are the lingering questions that I have about retro tiers going into the future:
I think the usage system is good enough. These tiers are speculative either way, so achieving the closest answer to the "what if" question should be an ideal. I think the tournament question should largely be left to the TDs or the Old Gen councils, even though I fear that the action of doing so is very easily a deathbed...DPP PU has gotten largely positive feedback but it does have the phrase "PU" attached to it and all of the connotation from it too. I think if you make a tier official then you also make previous legitimate generations of that tier official as well, otherwise they just sit in limbo. I also think that retro tiers should push to resolve their metagames and then lock themselves when done, to mimic other lower tiers, as their ultimate goal.
Well i guess that's it, so I would love to hear other opinions
Currently the classification for DPP PU is: "created tier". This is not an ideal name imo so I will be calling them retro tiers, as they were made retroactively. A retro tier is a tier that was made in a generation after the generation had already come to a close. A regular tier has to fulfill these requirements:
- Created during the initial run of the generation in question
- based on time-relevant and time-sensitive data
So then there's the question of "well if they aren't made in the usual way, how are you making these tiers?" Well great news! Making lower tiers in ADV and older is incredibly easy, all you need is a council of people from the above tier to give a VR cutoff to make the tier and done! Incredible. This can, in theory, go on and on until the generation runs out of mons...although I personally don't have much interest in going "too low" where there is not an interesting meta to create. The usage retro tiers are created in a different way, a way that I spell out right here in this post. If you do not want to follow that link, or if it gets deleted fsr, I will spell it out again in this hide:
1. Find usage via tournaments. PL usage is used here, and Classic if applicable.
2. Find usage via ladder. There were RoA Spotlight ladders that occur from time to time, and those are used.
3. Contrast the 2 lists. Both of the ways to find usage have their own unique flaws, but they are used to double check the work. Any mons that fall under the cutoff for BOTH lists is declared to be solidly in the tier below. As for the remaining outliers...
4. Triple check the unique outliers with representatives from the tier above. The outliers need a more nuanced hand to figure out who goes where. The representatives vote on the outliers to determine their fate, and some will rise/fall depending.
2. Find usage via ladder. There were RoA Spotlight ladders that occur from time to time, and those are used.
3. Contrast the 2 lists. Both of the ways to find usage have their own unique flaws, but they are used to double check the work. Any mons that fall under the cutoff for BOTH lists is declared to be solidly in the tier below. As for the remaining outliers...
4. Triple check the unique outliers with representatives from the tier above. The outliers need a more nuanced hand to figure out who goes where. The representatives vote on the outliers to determine their fate, and some will rise/fall depending.
While this process might seem clunky at first glance, the results from it are actually phenomenal. Both ladder usage and tournament usage are flawed in different ways (ladder usage favors samples and the overly-eager singular users, tour usage is a much smaller sample size prone to seasonal trends), but combining the two it is possible to find some common ground and make a more accurate picture. After that, the above tier has to weigh in and knock out the more nuanced borderline mons, to iron out the last string of flaws that the system has. The current version of DPP PU was created with only ladder usage, and has gotten negative feedback from DPP NU players in terms of accuracy. Applied to DPP PU, the system works like this:
After ladder usage and tour usage are found, the mons that are found to be "common ground" are immediately written off as PU due to them having 2 chances to make a usage cutoff and failing both times. The mons that also don't make the cutoff but are unique to one list are as follows:
Tour-Specific: Camerupt, Crawdaunt, Golem, Mantine, Ninetales, Pinsir, Porygon2, Torkoal, Whiscash
Ladder-Specific: Cacturne, Dusclops, Glalie, Lickilicky, Muk, Poliwrath, Solrock, Drifblim (banned so not in the vote)
Some are obvious as to which tiers they go to. Others are...less so. I held a vote to determine where each go and here are the results. After the results, DPP PU is shifted in this manner:
Gains: Mantine, Pinsir, Whiscash, Crawdaunt, Torkoal, Camerupt, Solrock, Muk, Glalie
Losses: Cacturne, Lickilicky, Poliwrath, Dusclops, Drifblim
The result is that many of the obviously wrong mons, like Cacturne, are gone and the other obvious choices, like Mantine, are added. This is a significantly more accurate list. You can see the original list here if you are curious.
Tour-Specific: Camerupt, Crawdaunt, Golem, Mantine, Ninetales, Pinsir, Porygon2, Torkoal, Whiscash
Ladder-Specific: Cacturne, Dusclops, Glalie, Lickilicky, Muk, Poliwrath, Solrock, Drifblim (banned so not in the vote)
Some are obvious as to which tiers they go to. Others are...less so. I held a vote to determine where each go and here are the results. After the results, DPP PU is shifted in this manner:
Gains: Mantine, Pinsir, Whiscash, Crawdaunt, Torkoal, Camerupt, Solrock, Muk, Glalie
Losses: Cacturne, Lickilicky, Poliwrath, Dusclops, Drifblim
The result is that many of the obviously wrong mons, like Cacturne, are gone and the other obvious choices, like Mantine, are added. This is a significantly more accurate list. You can see the original list here if you are curious.
This system actually works so well that I plan on redoing DPP PU in order to follow this system and make precedent. Assuming there aren't a ton of objections, of course. I want to make a dialog above basically anything else.
So great, that's the what and the how, and here is the why this thread is necessary: They are fun, they are already made and growing, and I doubt that they will stop being created. I think it is important that we address these tiers and their playerbase at some point and talk about how we are going to both create these and address them in policy, and before another generation launches seems like the appropriate time. Here are the lingering questions that I have about retro tiers going into the future:
- Is the system for finding usage-based retro tiers "good enough"? They will be speculative either way, but the current system sacrifices full "usage objectivity" for more accuracy.
- Will these tiers ever be added into larger tournaments? This is a long-term question that might change over time, but an expectation would be nice. For the record, I'm expecting a no and I'm fine with that.
- What about tiers that aren't retro tiers but weren't included in the list provided on this post? Do they get grandfathered in when a tier becomes official, or does the starting point for an official tier start when they are declared official? The on-hand example is BW PU, who is not a retro tier but is not recognized as official either.
- How should retro tiers be handled in general in policy, going into the future?
- In what way do retro tiers interact with OU bans? It has been decided that pre-existing lower tiers are locked and aren't necessarily hit by them, but what about retro tiers?
I think the usage system is good enough. These tiers are speculative either way, so achieving the closest answer to the "what if" question should be an ideal. I think the tournament question should largely be left to the TDs or the Old Gen councils, even though I fear that the action of doing so is very easily a deathbed...DPP PU has gotten largely positive feedback but it does have the phrase "PU" attached to it and all of the connotation from it too. I think if you make a tier official then you also make previous legitimate generations of that tier official as well, otherwise they just sit in limbo. I also think that retro tiers should push to resolve their metagames and then lock themselves when done, to mimic other lower tiers, as their ultimate goal.
Well i guess that's it, so I would love to hear other opinions
Last edited: