Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
Not sure why you quoted my post then made up a quote that I didn’t say? Kind of dishonest tbh but anyways, if you actually read my post you’ll see I acknowledge people may be unhappy with the system, and if they can figure out a way to fix it for this gen then a decision should be made soon. Otherwise, if it’s not made soon, then yes wait for next gen for a more constructive overhaul. That would have been a more accurate quote other than coining the term “true usage based tiering”. Tiering based on drops only isn’t usage based tiering anyways, “true” or not (your quote). You can do it if that’s what people want, it just isn’t usage based. Yes my first post said we shouldn’t consider changes this gen, but after seeing the support for it it seems more plausible even if I think it’s ill advised. At the end of the day we play Pokémon on the internet, we can use whatever method we want that makes the majority of people happy. That’s why we can give our opinions and pool them together to see what everyone wants to do and not be scared to share them.
As I said, the entire premise of usage based tiering is flawed, which just continues to get more and more proven as time goes on and we get later into the gen. Its almost always rough for the lowest tiers, but eventually it reaches a point where even UU starts getting messed with (see: the Ttar and Excadrill situation). Again, if people want to leave the overarching problem for next gen, the least we could do is prevent the current gen's tiers from getting beaten up further. Personally I think changes should've already happened or if they are being worked on the common player should be more in light about it, but if nothing is happening in regards to how we tier, we could at least stop the current system from making itself look even worse.
My opinion on the matter as a whole is that keeping solely usage based tiering was never going to work in the long run, it has already needed a bunch of fixes and the list of problems with it just gets larger and larger. Mons used to get stuck in tiers they were terrible in or at best didnt really fit well (and still do!), but things are full on rising to tiers they dont really belong in based on the whims of singular or a few ladder players. This stems from what others have said where interest in ladder from tournament players has fallen off, especially in lower tiers like NU or PU, so statistics become less and less accurate to what the tier actually is.
I appreciate missangelic for starting this thread as it gave us a place to start talking about how we could alter the tiering system, but I feel there's an issue with people saying suggestions are "non-starters" for pretty.... crummy reasons? Ill just leave this with the people saying the problems with the system are features and not bugs, and counterpoint by saying bad features should not be in a system if they can be worked out. I dont have much else to say on this matter Ill just end this by quoting a quote that missangelic quoted.
From Monky25's Prohibiting Rises in the Final Months of the Generation during Tier Shifts post:
"Every lower tier is set to lose integral Pokemon in the final months of the generation, some suffering more than others... [Stopping rises] gives lower tiers time to take action on any last threats before the gen ends and the majority of the playerbase moves on, it’s hard to take action on old gens with the limited sample size. I also want to say we should keep drops. Why? Like said earlier it’s easier to handle drops than rises. If something mad OP drops we can just ban it, but we can’t control what rises. Drops can also help tiers. For example, NU can get a solid defensive staple in Gastrodon returning, while PU can get Guzzlord, another likely balanced Pokemon. Why gatekeep options to make these tiers better? Health of the tiers comes before some policy and tradition, all the lower tiers seek to only gain from removing rises while keeping drops near the end of the generation and I find it’s very easy to implement. Some might ask "Isn't this just another form of veto?" The difference here is that this is an objective method for stopping unwanted rises, not subjective where you have to determine whether the Pokemon is viable enough, which is the main criticism of the previously proposed veto system. Overall, removing rises from the last 2 shifts of the generation while keeping drops is the best action to take right now. It makes playing and tiering a lower tier far better at the cost of nothing. I’d hope this gets implemented very soon, before the July shifts, so we can continue to run our lower tiers and have them solved by the time the gen ends."
Was largely referring to this when I said "true usage-based tiering."
As I said, the entire premise of usage based tiering is flawed, which just continues to get more and more proven as time goes on and we get later into the gen. Its almost always rough for the lowest tiers, but eventually it reaches a point where even UU starts getting messed with (see: the Ttar and Excadrill situation). Again, if people want to leave the overarching problem for next gen, the least we could do is prevent the current gen's tiers from getting beaten up further. Personally I think changes should've already happened or if they are being worked on the common player should be more in light about it, but if nothing is happening in regards to how we tier, we could at least stop the current system from making itself look even worse.
My opinion on the matter as a whole is that keeping solely usage based tiering was never going to work in the long run, it has already needed a bunch of fixes and the list of problems with it just gets larger and larger. Mons used to get stuck in tiers they were terrible in or at best didnt really fit well (and still do!), but things are full on rising to tiers they dont really belong in based on the whims of singular or a few ladder players. This stems from what others have said where interest in ladder from tournament players has fallen off, especially in lower tiers like NU or PU, so statistics become less and less accurate to what the tier actually is.
I appreciate missangelic for starting this thread as it gave us a place to start talking about how we could alter the tiering system, but I feel there's an issue with people saying suggestions are "non-starters" for pretty.... crummy reasons? Ill just leave this with the people saying the problems with the system are features and not bugs, and counterpoint by saying bad features should not be in a system if they can be worked out. I dont have much else to say on this matter Ill just end this by quoting a quote that missangelic quoted.
You see to keep misunderstanding what I’m saying for some reason. I agree with you, it’s flawed. It’s an issue with usage based tiering. And if people want to change to something not-usage based tiering, that’s fine. I was just saying that doing what you are explaining here is not usage based tiering. That’s ok though - if that’s what people want. But please stop taking what I say out of context to make it seem like I am largely dissenting to changes at this point because it’s incorrect and no productive, and also, again, a bit dishonest.
You see to keep misunderstanding what I’m saying for some reason. I agree with you, it’s flawed. It’s an issue with usage based tiering. And if people want to change to something not-usage based tiering, that’s fine. I was just saying that doing what you are explaining here is not usage based tiering. That’s ok though - if that’s what people want. But please stop taking what I say out of context to make it seem like I am largely dissenting to changes at this point because it’s incorrect and no productive, and also, again, a bit dishonest.
My issue is that you keep bringing up that it "isn't usage based" when that doesn't really mean much at all given the circumstances. The reason I was even responding is because I was under the impression that you thought it was bad that it didn't follow the specific guidelines of usage-based tiering, and for that, I apologize. The thread was indeed initially made to propose changing the usage thresholds, but I, and many others it seems, feel the problem lies with the system itself. As I said, if we are not addressing it this gen, freezing rises would be the best option even if there's still a year left as the interest in ladders will only continue to dwindle. Again, I apologize for my misunderstanding, but I still do think the current system is flawed beyond proper repair.
I get why there’s a lot of discussion on freezing in this thread after I brought it up in the OP. It’s been a year since my original post on the tiering decay, and enough has happened to justify actually bringing it out as an option now that time has passed.
We’re 4-8 weeks away from the new format that’s being used for VGC for this year. Champions isn’t like BDSP where it’s seeing no serious consideration for competitive play by its developers, and I intuit that some people are not going to want to stick with playing SV when Champions releases. That leaves every SV lower tier with a limited amount of time to get their end-of-gen tierlists in a state that’s as best as possible, or else they’ll end up kicking their problems down the road to a smaller old gen council. I don’t think it’s realistic to pretend SV has over a year and a half left in the tank when ladder activity is reaching new lows for just about every tier, rises continue to disrupt stability, and there’s lack of agreement on internal tier issues following surveys. One of my most recent posts in this thread outlines all the changes that have happened since July 2024 alone, and every SV lower tier community can look forward to wading through all that tiering decay again while activity is at its lowest, which is going to make ladders even less reliable as a snapshot of the way tiers are played competitively.
From a lot of the posts in this thread, it seems like the consensus is that there should be at least some reform of this system. I’m personally not in agreement with abolishing ladder usage-based tiering (or rises) or hybridizing it with viability or tournament usage, but I think getting those on the table and actually under a critical lens is important so that the atmosphere surrounding the tiering of Gen 10 isn’t as confused.
Also, I do believe this system can be reformed to be better. The point of having discussions about this side of tiering policy shouldn’t be to shut down entirely and accept a more flawed system just because its proposed replacements would still have flaws. The way tiering is set up is arbitrary in the first place - 4.52% makes no sense out of context or to outsiders and it’s still arbitrary at its core. There’s also always going to be users who have a kneejerk reaction to any kind of proposed changes on the grounds that they’d want to tear apart any change, but they don’t have much to say about the intentional choices we make to support the system we have now.
I think the best thing is to pay attention and keep discussing ways we can improve tiering policy to work better for lower tiers and their stability instead of pretending that the Gen 8 updates got it right the first time, let alone formalized much of a process for how to end a generation. I know other users in this thread have raised issues about parts of the tiering process, but the one that’s still untouched from my OP and follow-up that I really want to dig into is the frequency of changes. I think the frequency of changes is what is disruptive to lower tiers, and having too high of a frequency of changes leads to the system being easier to manipulate over shorter periods of time.
Last part, but I want to echo the call to freeze rises. Only a few weeks or months until Champions is going to be a really quick turnaround, and most tiers would benefit from not losing staples in this upcoming shift.
hi hello, im ming, i like ru, i started in ru, have since branched out a little bit everywhere and as such, have had a lot of first hand experience with rises/shifts and how they impact metas and communities so I want to give a bit of my opinions here. I've already shared most of these in rucord and a lot of similar sentiments have already been expressed in this thread but I think we're starting to lose the plot a little bit so I want to hopefully try and get that back on track.
Firstly I want to just reiterate that the original, primary purpose of this thread was NOT TO UPEND USAGE BASED TIERING!!! Beyond talking over the consequences our current system has had on various tiers over the years, the legitimacy of freezing rises, or trying to put forth new methods of tiering all together, missangelics original proposal was to quite simply bump the threshold for rises. It's something that we did before because of these EXACT same complaints and it's something I think makes sense to do again at this point in time. 4.52% (our current benchmark) gives you a 50% chance to see x mon in 15 games, whereas the proposed 6.78% is that same 50% chance but this time in a 10 game sample size. I will admit that this is a somewhat slippery slope as these are completely arbitrary benchmarks and could be abused if taken to the extreme. With that being said, I do think 50% in 10 games is a healthy number and would lessen how "disruptive" shifts can be but more importantly, and this is my big chagrin with the current system, lessens the impact that individual users can have on shifts.
For this next part I will be referencing two instances specifically: uberfiend and this thread from december of last year. For those of my nu enjoyers, or people who have been around long enough to remember the hitmontop incident, you should already be familiar enough with uberfiend. For those of you who are NOT familiar with uberfiend here's a bit of context. I have been around on this site for a long time. I really started picking up mons around the back end of gen 6/start of gen 7 and have racked up a fair number of games amongst generations since then. Grand total on my most used alt, thesky12321, lands me at 14,446 games played across the entire lifetime of like a 13 almost 14 year old account. uberfiend has played 14,074 games of gen 9 nu. He is one of the most dedicated mons players I have ever seen with an unbelievable commitment to the nu ladder, reflected by that unbelievable quantity of games played. The uberteam has gone through various iterations with a couple of them included below, but the main mon I want to bring up is braviary who uberfiend basically single handedly rose to nu.
(the more iconic versions of these teams had a diancie > dudunsparce and reuni respectively)
While we don't have usage stats for individual accounts and their impact, anyone who has actually meaningfully engaged with the NU ladder this generation can tell you about uberfiend and the very obvious influence he has as far as ladder stats should be concerned. Even for those not plugged into NU, the impact he exerts can be SEEN when braviary jumped from ZU all the way to NU during the january shifts of this year (its second stint in NU mind you, the first of which was also fueled by uberteam inflation). This is where the improved usage cutoff comes into play. Raising the threshold means just naturally theres a higher barrier to entry for mons to jump tiers which makes it less likely for one person to be able to single-handedly raise mons. It is worth mentioning that the proposed 6.78% threshold actually would not have stopped braviary from rising, this is again a testament to the sheer volume of games uberfiend is putting up, but it COULD still be relevant for other mons. Take for example Hariyama who at this point is the only non NU mon included on these teams (btw theres probably an argument for uberfiend being responsible for decidueye and plume being NU as well but thats much less definitive so ehhhh). In the most recent shift stats (february 2026) yama is sitting at just over 4% usage which although still a bit away from the 4.52% cutoff, is not NEARLY as far away as it would be in a more active ladder like OU and is, at least imo, legitimately at risk to rise considering uberfiends past trends.
This is all to say, that there is still merit to implementing a new threshold as a RIGHT NOW type of solution. At this point I'm more or less finished with my thoughts on the initial purpose of this thread. The TL;DR is I think the higher rise/drop threshold is a good idea and there is ample opportunity and reason to implement it during this current generation.
The next point I want to bring up is like a somewhat already resolved point of contention which leads into a different but also important talking point. During the backend of last year, OU experienced an issue with bots spamming the ladder with some nonsense triple weather team that put guys like torkoal and excadrill at risk of rising to OU and opened up a conversation about bots and their place on ladder in general. At the end of the day those guys did NOT end up rising (ik uu enjoyers were shivering in their timbers when they saw how close drill was) and we were promised that a "technical fix" was in the works by the tiering leads. The issue here is a similar one to the uberfiend situation in the previous section where one person, or in this case a group of bots, with an inflated games played total can artificially move shifts in a way that is counterproductive to the spirit of usage based tiering, which I interpret as giving the community as a whole the opportunity to affect Smogon tiers. However in this case, I'm more or less cool with how things resolved. An issue was presented, issue turned out to be a nothing burger hurrah (unless you were in the group which thought the blim bots inflated dozo's usage to keep him in ou levels idk what to tell yall), and a solution was proposed. The point that I'd like to move onto now is the lack of transparency regarding tiering.
During the close of this thread, which I mentioned earlier, shiloh referenced that there was some technical solution that was kept in the wings until necessary. In this specific instance, most people (myself included) will probably be unable to understand the exact specifics behind the technical solution that the programmer friends behind ladder and usage stats cooked up. But even if such a solution is in an early developmental state or still needs testing, I see no benefit to keeping those mechanisms hidden. Senior Staff have always acted off a "need-to-know" basis which makes a lot of sense as far as protecting confidentiality regarding user safety and such, but, when it comes to tiering which is by its nature a community based effort, this is a different matter. Ishtar also touches a bit upon the topic of transparency among tiering staff in this post above:
I'm chiming in here after months of posting in this thread and seeing discussions brewing to ask tiering leaders about the plan to change the tiering system in WW. I understand that this is very preemptive still with SV still existing for quite a while, but me and other leaders put in a lot of work (before I stepped down from PUTL) to improve on the system. This improved system was never discussed publicly and after I stepped down the discussion was taken over by another TL who ended up getting perma'd. I'd hope that meaningful progress made in good faith by many of us didn't get totally lost in these changes.
Months after this discussion, the conversation of people wanting changes would randomly pop up, and I always had to give the answer: "stuff was being worked on behind the scenes", which was true while I was TL and allegedly after too. I feel like the common user has been left in the dark about this and seeing so many posts still looking for solutions while there was already a framework in place in regards to these changes seems bad. For that reason I'm asking whoever runs this nowadays if we can please get an update on my thread and the changes coming to tiering come WW, and to also consider a lot of the points here in regards to stopping individual account's influence to the process.
Even just knowing that these changes are actually still coming, even if too vague for my liking, would serve of great help to anyone wanting change in this thread I'm sure!
As someone who has spent a lot of time with the PU community and has discussed the issues of usage based tiering with ishtar at length in the past, being told to essentially trust the process where "stuff was being worked on behind the scenes" was something that I heard firsthand and just took at face value for a while. But even since the start of this thread, its been almost a year with no news from the team about potential amendments to the system beyond Marty popping in out of left field with a "freezing rises is now on the table" message. More recently I've talked to etern about these transparency issues and he can share his opinions as well if so inclined, but basically I got the impression that things ARE still moving behind the scenes but were severely limited by the banning of the former SS member in charge of the project. While I don't expect like a weekly report on this type of thing, at least some sort of place where we as community members can look to see our leadership working towards a solution, I think, would give a lot of faith back to what has become a largely disillusioned population when it comes to how usage based tiering fixes have been handled behind closed doors. TL;DR there should be more avenues of communication between tiering leads and the communities impacted by their decision making.
That is all, thank you to those who took the time to read, shoutouts missangelic for opening this thread in the first place and keepin it bumpin, ishtar for her continued efforts to improve tiering, and Shengineerloocas13 and bricknermon for helping keep the uberfiend timeline straight, I look forward to what the future holds not only for gen 10 but hopefully the remainder of what we have in gen 9
Champions now has a confirmed release date that's less than two weeks away, so there's no better time than now to revisit this with more discussion of some points I haven't seen or made yet.
On declining ladder interest: SS & BDSP vs SV & Champions:
During 2021, the first full year of DLC2 SS, the OU ladder declined from ~1.8 million games to ~1 million games by the time BDSP OU released in November. BDSP, which wasn't even a VGC format, still attracted attention that ended up being divested from OU. Still, BDSP's lack of interest waned, and it went from ~520k games on release of its OU ladder to less than 100k games per month half a year later. The SS ladder would recover to about 1.6 million games by the end of the gen. Still, the SS OU ladder saw about a 10% decline to the end of the gen in late 2022.
For comparison, SV has had over 2 full years of DLC2, and its OU ladder declined from ~1.9 million games in January 2024 to just ~1.1 million games in January 2026, which means SV OU's ladder has declined by over 40% in activity. The last time the OU ladder saw greater than 1.6 million games in a month was January 2024. The ladder didn't even reach 1 million games in November and December 2025. Without any other core series turn-based game coming out, SV has settled at levels similar to when SS was being impacted by BDSP.
This should go without saying, but all of the SV lower tiers have seen similar, if not more larger, declines. The UU ladder hasn't broken 100k games since May 2025, RU's been stuck below 50k for nearly as long, NU and PU are hitting new lows of below 20k, and ZU is struggling to break 10k. Less traffic leads to fewer competitive games and a less accurate snapshot of what competitive players use, intensifying a lot of the recurring clusters of issues that people have brought forward in this thread and previous tiering policy actions.
Champions being the VGC format likely means there's going to be a hit to the SV ladder that's even harder than the one SS's took when BDSP released, and certainly not one that it will recover from either. Champions is only going to draw more people to the format when it releases on mobile later this year, and it's clear Champions is going to be the VGC format for 2027 too.
It's also pretty obvious that on top of Champions being the VGC 2026 format, more people are going to be drawn to play a new format. This isn't to say SV has aged badly or even gotten stale, but it's the longest-running CG format since DPP with the fewest changes in its DLC2 era, so it's gotten to a point where a lot of people are going to jump at the chance to play a fresher format with more depth, and it doesn't hurt that it has fan-favorites like Mega Evolutions.
With all this in mind, I don't think there's much time to reform tiering in SV other than by freezing rises or locking tiers as they are now before the April tier shifts. By freezing tiers (or at least rises), lower tiers can take advantage of what's probably going to be a quickly shrinking amount of attention to finish out their tiering action, tournaments, and resources without having to account for tier shifts and the metagame development that happens afterward.
On tiering reformation:
Tournament stats are up there with VR-based tiering as one of the worse approaches to tiering, especially for promoting stability. Aside from the logistical nightmare of trying to coordinate and verify these across tiers and tournaments, tournaments don't provide good snapshots of metagames that make for stable tiers.
Usage-based tiering from the ladder gets to observe a large sample size across a three month period, while tournament usage-based tiering suffers from small sample sizes that are hard to easily weight against each other. The weighting of ladder stats also favors players who are climbing the ladder, not necessarily defending the top positions, so it's also providing a better representation of general competitive players instead of the handful of elite players who were active in a period of time.
Tournament play also evolves much faster and picks up trends that might not stick as long compared to ladder, so they're less reliable as accurate or holistic portraits of a metagame over time. This is a pretty big contradiction with the goals of usage-based tiering, which is to reflect the tier across a period of time. A massively underrated change that was made in the jump to tiering for Gen 8 was the shift away from weighting the most recent month of stats in usage for 3-month shifts.
Last, tournament usage-based tiering is going to be really hard, it not impossible, to reconcile with the goal of usage-based tiering as being democratic. Tournaments present a huge opportunity for data from competitive players, but team tournaments aren't systematically organized to select for the players who are objectively the more competitive players at the format they play, so this is inevitably going to take away from Smogon's tiering being accessible to newer players and constrict tiers to being decided by a shrinking pool players, and working with a smaller set of data is only going to create more skew and destabilize tiers further.
I also don't think exploring VR-based tiering is the future. Just to get it out of the way first, they're pretty undemocratic by restricting the pool of people who can influence them.
VRs are one snapshot of one point in a metagame rather than a reflection of tier across a period of time, so they're still prone to issues of instability. Even if I do think this can be reformed with caveats, like preventing Pokemon from rebounding back up one shift after they fall, they still run into the issue of only analyzing a narrow moment of a metagame.
We should continue to push the tiering system towards greater stability and keep reforming it to work for new cases like SV's longer lifespan. Hopefully we can get some kind of freeze to give SV lower tiers a chance to wrap as Champions releases, and more transparency on this process is always welcome. I think the best routes to look into for reforming usage-based tiering would be preventing rebound rises, decoupling the thresholds for rising and dropping, changing the frequency of shifts, or raising the general threshold again, but I do think there's still avenues for changing usage-based tiering that haven't been explored or proposed yet that would make tiering better for Champions and WW down the line.
I don’t have what all the numbers would be, but what if for future tiering in Gen 10 and beyond, you keep usage based tiering with drops every 3 mos at the same % (or whatever % is deemed best), and also kept the same % for rises but only had rises every 6 mos instead of 3? That would help maintain the same definition of what is “UU” by % in usage based tiering, and it would mean only the truly deserving mons end up rising. Usage based tiering definitely doesn’t work without rises, but a very subtle shift like this could help maintain stability better while still appeasing those who like usage based tiering to be mostly uniform in the basis for movement. I don’t think we are going to go towards VR tiering anytime soon, and the logistical challenges of incorporating tours to usage data that were laid out in the post above are valid.
With all this in mind, I don't think there's much time to reform tiering in SV other than by freezing rises or locking tiers as they are now before the April tier shifts.
I have an issue with this specifically because we currently don't even have confirmation if Champions will allow 6v6 Singles in private rooms or whatever, let alone at level 100 due to everything seemingly being locked to 50. We shouldn't be using it as the basis of any major decisions until we have more concrete info when we might not even be changing into a new standard format.
Plus trying to implement a freeze when the next tier shift occurs in like a week and before Champions even drops feels. Wrong? Wasteful? idk but not good.
I have an issue with this specifically because we currently don't even have confirmation if Champions will allow 6v6 Singles in private rooms or whatever, let alone at level 100 due to everything seemingly being locked to 50. We shouldn't be using it as the basis of any major decisions until we have more concrete info when we might not even be changing into a new standard format.
Plus trying to implement a freeze when the next tier shift occurs in like a week and before Champions even drops feels. Wrong? Wasteful? idk but not good.
There’s still lots unknown about Champions, but that’s been the case about every main series Pokemon game before release. I already pointed out in my last post how BDSP, a less competitive game, took activity from the SS ladder. We shouldn’t be sticking our heads in the sand and acting like Champions is going to have no impact on SV or that we can’t do anything about tiering.
We’ve been trying to discuss freezing for over a year in SV because the Gen has gone on for as long as it has. Gen 8’s freeze happened with 2 tier shifts left to go before SV, and it was announced in the shift before the ones that were affected. There could have been some similar formal policy on how usage-based tiering should wrap for SV, but there was almost no transparency for over a year before Marty asked whether or not we’re having a freeze. Now we have a little over 10 days for lower tiers to get their affairs in order before attention economy of their ladders completely crashes. If lower tiers get destabilizing shifts right before a new game release, then the lower tier communities have to deal with their ladders being dead and stifling suspect tests, and that’s just the tip of the iceberg. Let’s not have that happen.
