Discussion The pattern of more rises than drops in usage-based lower tiers

[missangelic]

At the beginning of Generation 8, there was an update to tiering for lower tiers where Smogon adjusted the usage threshold from 3.41% to 4.52%. Reasons cited for the policy included decreasing how disruptive tier shifts were to lower tiers, counteracting ripple effects in lower tiers, and making drops more frequent than rises. I think this was a change in the right direction, but I don't think it was enough of a change: most tiers still end seeing more losses than gains. At the end of Generation 8, there was another tiering policy, and it froze rises altogether after citing instability from the effects of rises. I'm going to explain why tiers were and are still unstable, and then I will propose a change.

Before the January 2025 tier shifts, I made a post in the usage stats discussion thread, and I've condensed all the information from it into this one so it isn't necessary to read the old post. My observation was Pokemon in lower tiers are rising faster than they're replaced by new drops, which continuously and progressively destabilizes lower tiers. This contradicts the intention of the "Tiering for Generation 8" policy of facilitating more rises than drops, as the 4.52% usage threshold still isn't dropping more Pokemon into lower tiers than it's rising up to higher tiers.

As current gen usage-based higher tier metagames develop, they tend to pull up Pokemon from lower tiers. Pokemon may become better as they metagame develops, they secure new niches, or they just gain consistent usage. One example of an impactful rise is Zapdos's movement from UU to OU in the October 2024 tier shift. When a tier loses a staple option, it creates a vacuum, and that vacuum pulls up Pokemon from tiers below. When Pokemon from those lower tier fill a vacuum, they create bigger vacuums in the tiers they leave behind. This effect creates disruptive tier shifts like the following:

July 2024 shifts

Krookodile moved from NU to RU
Magnezone moved from NU to RU
Noivern moved from NU to RU
Reuniclus moved from NU to RU
Slowbro moved from NU to RU
Thundurus moved from NU to RU

October 2024 shifts

Bronzong moved from PU to NU
Gligar moved from PU to NU
Heracross moved from PU to NU
Kilowattrel moved from PU to NU
Milotic moved from PU to NU
Scream Tail moved from PU to NU
Slowbro-Galar moved from PU to NU
Staraptor moved from PU to NU
Tauros-Paldea-Aqua moved from PU to NU
Toxicroak moved from SU to NU

January 2025 shifts

Conkeldurr moved from RU to UU
Revavroom moved from RU to UU
Mew moved from NUBL to UU

Decidueye moved from PU to NU
Gastrodon moved from PU to NU
Meloetta moved from PU to NU
Tornadus moved from ZUBL to NU

Cramorant moved from ZU to PU
Decidueye-Hisui moved from ZU to PU
Frosmoth moved from ZU to PU
Lycanroc moved from ZU to PU
Rhydon moved from NFE to PU

How severe is the tiering vacuum?
I've compiled all of the DLC2 SV tier shifts and tiering actions so no one else has to go through the trouble of checking all of them. However, I've still linked all of them in case anyone wants to check my work or point out something I missed. I'm including Pokemon tiering actions in this because they still remove options from lower tiers.

Tiering vacuum across SV by tier

OU made 4 new Pokemon bans.

OU

December 16th, 2023: Terapagos quickban
February 17th, 2024: Archaludon ban
April 27th, 2024: Volcarona ban
September 7th, 2024: Gouging Fire ban

UU lost a net total of 9 Pokemon to rises and bans. 3 of its gains didn't stick and fell further.

UU

January 6th, 2024: Blaziken, Iron Hands, and Latias quickbans
January 29th, 2024: Ceruledge ban
February 2024 Tier Shifts
- Alomomola moved from UU to OU
- Iron Treads moved from UU to OU
- Barraskewda moved from RU to OU
- Iron Crown moved from OU to UU
- Pecharunt moved from OU to UU
- Toxapex moved from OU to UU
February 19th, 2024: Garganacl ban
February 24th, 2024: Moltres-Galar quickban
March 2024 Tier Shifts
- Blissey moved from RU to OU
- Deoxys-Speed moved from OU to UU
- Excadrill moved from OU to UU
- Skeledirge moved from OU to UU
March 25th, 2024: Garchomp ban
April 2024 Tier Shifts
- Deoxys-Speed moved from UU to OU
- Iron Moth moved from UU to OU
- Ribombee moved from UU to OU
- Iron Boulder moved from OU to UU
- Pelipper moved from OU to UU
- Barraskewda moved from OU to RU
April 11th, 2024: Iron Boulder quickban
May 2024 Tier Shifts
- No quickdrops
May 13th, 2024: Pelipper ban
June 2024 Tier Shifts
- No quickdrops
June 12th, 2024: Kommo-o ban
July 2024 Tier Shifts
- Iron Crown moved from UU to OU
- Moltres moved from RU to OU
- Ribombee moved from OU to UU
- Serperior moved from OU to UU
- Torkoal moved from OU to UU (watch this guy take half a year to get to ZU)
July 21st, 2024: Ursaluna ban
August 2024 Tier Shifts
- No quickdrops
September 2024 Tier Shifts
- No quickdrops
October 2024 Tier Shifts
- Zapdos moved from UU to OU
- Blissey moved from OU to UU
- Clodsire moved from OU to UU
- Heatran moved from OU to UU
- Skarmory moved from OU to UU
- Weavile moved from OU to UU
November 2024 Tier Shifts
- No quickdrops
November 11th, 2024: Okidogi ban
December 2024 Tier Shifts
- No quickdrops
December 10th, 2024: Hoopa-Unbound ban
January 2025 Tier Shifts
- Blissey moved from UU to OU
- Pecharunt moved from UU to OU
- Araquanid moved from NU to OU
- Deoxys-Speed moved from OU to UU
February 2025 Tier Shifts
- No quickdrops
March 2025 Tier Shifts
- No quickdrops
March 11th, 2025: Quaquaval ban

Lost: 27
Gained: 18

RU lost a net total of 14 Pokemon to rises and bans. 6 of its gains didn't stick and fell to lower tiers.

RU

February 4th, 2024: Hydreigon and Manaphy quickbans
February 21st, 2024: Iron Jugulis quickban
March 2024 Tier Shifts
- Blissey moved from RU to OU
- Donphan moved from RU to UU
- Mamoswine moved from RU to UU
- Rotom-Wash moved from RU to UU
- Slowking moved from RU to UU
- Tinkaton moved from RU to UU
- Indeedee moved from NU to UU
- Amoonguss moved from UU to RU
- Cinccino moved from UU to RU
- Deoxys-Defense moved from UU to RU
March 18th, 2024: Zarude ban
March 27th, 2024: Hoopa-Unbound quickban
April 2024 Tier Shifts
- Barraskewda moved from OU to RU
- Jirachi moved from UU to RU
May 2024 Tier Shifts
- No quickdrops
May 11th, 2024: Thundurus-Therian ban
May 16th, 2024: Iron Leaves quickban
May 27th, 2024: Enamorus-Therian ban
June 2024 Tier Shifts
- No quickdrops
July 2024 Tier Shifts
- Moltres moved from RU to OU
- Cobalion moved from RU to UU
- Rhyperior moved from RU to UU
- Goodra-Hisui moved from UU to RU
- Weezing-Galar moved from UU to RU
- Indeedee moved from UU to NU
August 2024 Tier Shifts
- No quickdrops
August 17th, 2024: Blastoise ban
September 2024 Tier Shifts
- No quickdrops
October 2024 Tier Shifts
- Okidogi moved from RU to UU
- Ninetales-Alola moved from UU to RU
- Ribombee moved from UU to RU
- Torkoal moved from UU to RU
October 18th, 2024: Thundurus ban
November 2024 Tier Shifts
- No quickdrops
November 23rd, 2024: Yanmega ban
December 2024 Tier Shifts
- Rhyperior moved from UU to RU
January 2025 Tier Shifts
- Araquanid moved from NU to OU
- Conkeldurr moved from RU to UU
- Revavroom moved from RU to UU
- Mew moved from NUBL to UU (had enough usage to make RU the shift Deo-S went UU...)
February 2025 Tier Shifts
- No quickdrops
March 2025 Tier Shifts
- No quickdrops

Lost: 26
Gained: 12

NU lost a net total of 21 Pokemon to rises and bans. 2 of its gains didn't stick and fell to lower tiers.

NU

March 8th, 2024: Lilligant-Hisui, Regidrago, and Suicune quickbans
March 15th, 2024: Armarouge, Gyarados, and Necrozma quickbans
April 2024 Tier Shifts
- Chansey moved from NU to RU
- Overqwil moved from NU to RU
- Feraligatr moved from RU to NU
May 2024 Tier Shifts
- No quickdrops
June 2024 Tier Shifts
- No quickdrops
June 14th, 2024: Mew quickban
July 2024 Tier Shifts
- Rhyperior moved from NU to UU
- Krookodile moved from NU to RU
- Magnezone moved from NU to RU
- Noivern moved from NU to RU
- Reuniclus moved from NU to RU
- Slowbro moved from NU to RU
- Thundurus moved from NU to RU
- Indeedee moved from UU to NU
- Araquanid moved from RU to NU
- Breloom moved from RU to NU
- Cinccino moved from RU to NU
- Cresselia moved from RU to NU
- Deoxys-Defense moved from RU to NU
July 17th, 2024: Lycanroc-Dusk quickban
July 25th, 2024: Deoxys-Defense and Feraligatr quickbans
August 2024 Tier Shifts
- No quickdrops
August 5th, 2024: Cloyster and Lucario quickbans
August 23rd, 2024: Gallade ban
September 2024 Tier Shifts
- No quickdrops
October 2024 Tier Shifts
- Quagsire moved from NU to RU
- Talonflame moved from NU to RU
- Umbreom moved from NU to RU
- Overqwil moved from RU to NU
October 13th, 2024: Cresselia and Iron Thorns quickbans
October 27th, 2024: Oricorio-Pom-Pom quickban
November 2024 Tier Shifts
- No quickdrops
December 2024 Tier Shifts
- No quickdrops
January 2025 Tier Shifts
- Araquanid moved from NU to OU
- Amoonguss moved from RU to NU
- Ninetales-Alola moved from RU to NU
- Torkoal moved from RU to NU
January 16th, 2025: Mienshao ban
January 29th, 2025: Oricorio-Sensu quickban
February 2025 Tier Shifts
- No quickdrops
February 4th, 2024: Cetitan quickban
March 2025 Tier Shifts
- No quickdrops

Lost: 32
Gained: 11

PU lost a net total of 18 Pokemon to rises and bans. 3 of its gains didn't stick, falling to lower tiers.

PU

April 7th, 2024: Drednaw quickban
April 14th, 2024: Oricorio-Pom-Pom quickban
April 28th, 2024: Cetitan ban
May 2024 Tier Shifts
- No quickdrops
June 2024 Tier Shifts
- No quickdrops
July 2024 Tier Shifts
- Copperajah moved from NU to PU
- Raikou moved from NU to PU
July 7th, 2024: Raikou quickban
July 21st, 2024: Inteleon ban
August 2024 Tier Shifts
- Ninetales moved from NU to PU
- Venusaur moved from NU to PU
September 2024 Tier Shifts
- Brute Bonnet moved from NU to PU
October 2024 Tier Shifts
- Toxicroak moved from SU to NU
- Bronzong moved from PU to NU
- Gligar moved from PU to NU
- Heracross moved from PU to NU
- Kilowattrel moved from PU to NU
- Milotic moved from PU to NU
- Scream Tail moved from PU to NU
- Slowbro-Galar moved from PU to NU
- Staraptor moved from PU to NU
- Tauros-Paldea-Aqua moved from PU to NU
- Indeedee moved from NU to PU
October 6th, 2024: Indeedee quickban
October 13th, 2024: Duraludon and Flamigo quickbans
November 2024 Tier Shifts
- No quickdrops
November 28th, 2024: Scyther ban
December 2024 Tier Shifts
- No quickdrops
January 2025 Tier Shifts
- Decidueye moved from PU to NU
- Gastrodon moved from PU to NU
- Gastrodon moved from PU to NU
- Meloetta moved from PU to NU
- Tornadus moved from ZUBL to NU
- Torterrra moved from NU to PU
January 13th, 2025: Torterra quickban
January 29th, 2025: Oricorio-Sensu quickban (from NU)
February 2025 Tier Shifts
- Torkoal moved from NU to PU
March 2025 Tier Shifts
- No quickdrops

Lost: 26
Gained: 8

ZU lost a net total of 12 Pokemon to rises and bans. 4 of its gains didn't stick and fell out of the tier.

ZU

May 5th, 2024: Articuno-Galar, Hitmonlee, Sceptile, and Tornadus quickbans
May 30th, 2024: Emboar ban
June 2024 Tier Shifts
- No quickdrops
June 2nd, 2024: Vulpix quickban
June 15th, 2024: Uxie quickban and Vulpix quick-unban
June 27th, 2024: Kingdra ban
July 2024 Tier Shifts
- Bombirdier moved from ZU to PU
- Hoopa moved from ZU to PU
- Grafaiai moved from PU to ZU
- Whimsicott moved from PU to ZU
July 7th, 2024: Alcremie and Thwackey quickbans, Hitmonlee and Sceptile quick-unbans
July 28th, 2024: Thwackey quick-unban
August 2024 Tier Shifts
- No quickdrops
August 16th, 2024: Bruxish ban
August 25th, 2024: Electrode-Hisui ban
September 2024 Tier Shifts
- Ninetales moved from PU to ZU
October 2024 Tier Shifts
- Toxicroak moved from SU to NU
- Brute Bonnet moved from PU to ZU
- Delphox moved from PU to ZU
- Froslass moved from PU to ZU
- Hariyama moved from PU to ZU
- Lycanroc moved from PU to ZU
October 5th, 2024: Delphox and Hariyama quickbans
November 2024 Tier Shifts
- No quickdrops
November 30th, 2024: Dudunsparce ban
December 2024 Tier Shifts
- No quickdrops
January 2025 Tier Shifts
- Cramorant moved from ZU to PU
- Decidueye-Hisui moved from ZU to PU
- Frosmoth moved from ZU to PU
- Lycanroc moved from ZU to PU
- Rhydon moved from NFE to PU
- Hitmontop moved from PU to ZU
- Rotom-Mow moved from PU to ZU
January 28th, 2025: Oricorio-Sensu quickban
February 2025 Tier Shifts
- No quickdrops
February 24th, 2025: Venomoth quickban
March 2025 Tier Shifts
- Torkoal moved from PU to ZU

Lost: 27
Gained: 15

The Pokemon that fail to stick around make the tiering vacuum more powerful than I'm describing with the net gains/losses. Torkoal having no role in NU means that it's basically an empty drop that doesn't do anything about the tiering vacuum.

Lower tiers lose more Pokemon than they gain, at an average net loss of about 15 Pokemon at this point. Tiers are approximately 37 Pokemon in total, so nearly half of each tier is gone without replacement. This caused the severe tier shifts in July, October, and January when some tiers suffered major losses. Pokemon aren't just rising to higher tiers because they're emerging into new roles, they're having to replace what's been lost. I can pretty securely predict that the upcoming April tier shifts are going to be pretty destabilizing too, which will ripple into later tier shifts. Lower tiers are cannibalizing each other just by developing. By the end of SS, most lower tiers saw a net loss of 23 Pokemon even without rises for the last full tier shifts. SV is continuing on for what looks like an extra year compared to SS, so the tiering vacuum in SV is going to be even more severe in SV than SS.

Lower tiers losing pieces drives tiering action, which makes the vacuum stronger. Previously borderline or not-broken elements tend to become banworthy when the things holding them back are sucked up into higher tiers. Bans also create vacuums, which accelerates the process of pulling up more Pokemon from lower tiers. The lowest usage-based tiers, NU, PU, and ZU, are most strongly affected by the progressive tiering vacuum. This is why I've included tiering actions in this discussion, as they contribute to the effect we saw in SS and continue to see in SV.

Lower tier instability creates stress on lower tier communities. Lower tier councils have to be even more proactive and engaged, metagame resource contributors have to work harder, and tournaments become more volatile. Overall, this situation creates a lot of burnout across the types users affected by it.

My proposal: raise the threshold for rises and keep the same threshold for drops, making ~6.78% the new threshold for a lower tier Pokemon to move to a higher tier. 4.52% usage reflects a 50% chance that a player will see a Pokemon in a series of 15 games on the competitive section of the ladder. 6.78% reflects roughly that same chance in a series of 10 games, just as 3.41% usage reflected a 50% chance a player would see a Pokemon in a series of 20 games.

Raising the rise threshold above the drop threshold would reduce the number and frequency of rises from lower tiers.
There are Pokemon that hover around the 4.52% threshold, such as Blissey and Meowscarada in OU. Blissey still moved back to OU instead moving from UU to RU in January 2025's tier shifts, and now Blissey is on the verge of dropping to UU again. Blissey even made its way down to NU earlier in SV, but it keeps rubber-banding to OU just because of the small fluctuations around the 4.52% threshold used for drops and rises. A policy of detaching the rise threshold from the drop threshold would let lower tiers hold onto these types of Pokemon instead of dealing with constant shifts.

As proof, here's what would have changed for SV with this decoupling in place from July 2024, the first 3-month average tier shifts, onwards.

July 2024 shifts

Iron Crown would not have moved from UU to OU
Moltres would not have moved from RU to OU

Comfey would not have moved from RUBL to UU
Rhyperior would not have moved from NU to UU (or have gained enough traction in RU to later move from UU to RU instead of NU

Magnezone would not have moved from NU to RU
Noivern would not have moved from NU to RU
Reuniclus would not have moved from NU to RU
Thundurus would not have moved from NU to RU

Bombirdier would not have moved from ZU to PU
Hoopa would not have moved from ZU to PU

October 2025 shifts

Iron Crown would have moved from UU to OU (delayed from July 2024)
Moltres would have moved from UU to OU (delayed from July 2024)

Zapdos would not have moved from UU to OU

Comfey would have moved from RUBL to UU (delayed from July 2024)
Rhyperior would have moved from NU to UU (delayed from July 2024)

Quagsire would not have moved from NU to RU
Talonflame would not have moved from NU to RU
Umbreon would not have moved from NU to RU

Gligar would not have moved from PU to NU
Heracross would not have moved from PU to NU
Scream Tail would not have moved from PU to NU
Staraptor would not have moved from PU to NU

January 2025 shifts

Blissey would not have moved from UU to OU (and would have moved to RU instead)
Pecharunt would not have moved from UU to OU (it would still be guaranteed to rise this coming April 2025 shift)
Araquanid would not have moved from NU to OU

Hawlucha would not have moved from RUBL to UU
Zarude would not have moved from RUBL to UU
Conkeldurr would not have moved from RU to UU (would be guaranteed to rise this coming April 2025 shift)
Revavroom would not have moved from RU to UU
Mew would not have moved from NUBL to UU

Suicune would not have moved from NUBL to RU

Raikou would not have moved from PUBL to NU
Scyther would not have moved from PUBL to NU (would be guaranteed to rise this coming April 2025 shift)
Decidueye would not have moved from PU to NU (would be guaranteed to rise this coming April 2025 shift)
Gastrodon would not have moved from PU to NU (would be guaranteed to rise this coming April 2025 shift)
Meloetta would not have moved from PU to NU
Tornadus would not have moved from ZUBL to NU (and would have moved to PU instead)

Articuno-Galar would not have moved from ZUBL to PU (would be guaranteed to rise this coming April 2025 shift)
Delphox would not have moved from ZUBL to PU (would be guaranteed to rise this coming April 2025 shift)
Cramorant would not have moved from ZU to PU (would be guaranteed to rise this coming April 2025 shift)
Frosmoth would not have moved from ZU to PU (would be guaranteed to rise this coming April 2025 shift)
Lycanroc would not have moved from ZU to PU

This wouldn't stop some shifts, just delay them. That said, it can be hard to say which changes might be "legitimate": a Pokemon appearing in the teambuilder biases players to use it more, even on the competitive end of the ladder. Since ladder is not completely detached from metagame trends, we'd probably still see changes like Iron Crown moving to OU and Decidueye moving to NU. Delayed and reduced rises would make it easier for lower tiers to anticipate how dramatically things would change in an upcoming tier shift.

Unlike the rise-freezing policy of SS, this policy wouldn't stop rises altogether. Tier shifts like RU-UU Cobalion, NU-RU Slowbro, PU-NU Bronzong, and NFE-PU Rhydon still would have happened when they did. I've put together a compilation of rises that still would have happened "on-time" like with how they did in the current system.

Further proof

Here are all the rises that still would have happened "on-time" with how they did in the current system.

July 2024 shifts

Cobalion still would have moved from RU to UU

Krookodile still would have moved from RU to UU
Slowbro still would have moved from RU to UU

October 2024 shifts

Hoopa-Unbound still would have moved from RUBL to UU
Thundurus-Therian still would have moved from RUBL to UU
Okidogi still would have moved from RU to UU

Inteleon still would have moved from PUBL to NU
Bronzong still would have moved from PU to NU
Kilowattrel still would have moved from PU to NU
Milotic still would have moved from PU to NU
Slowbro-Galar still would have moved from PU to NU
Tauros-Paldea-Aqua still would have moved from PU to NU
Toxicroak still would have moved from SU to NU

January 2025 shifts

Polteageist still would have moved from RUBL to UU

Uxie still would have moved from ZUBL to PU
Decidueye-Hisui still would have moved from ZU to PU
Rhydon still would have moved from NFE to PU

Even though I only applied this to the 3-month average DLC2 era of SV, it would also stabilize the pre-DLC2 period. Lower tier councils would have more breathing room during the most volatile period of the generation, and playerbases would not have to deal with as much chaos.

My goal with this discussion is to increase awareness that rises are still more frequent than drops in the current tiering system. My perspective is we should do something about this effect driving the instability that usage-based SV lower tiers have been dealing with. I think SV would be better off than SS by getting earlier jump on this rather than having to potentially freeze rises and conduct several suspect tests.

Tl;dr:
1. Rises, tiering action, and useless drops still outnumber useful drops for usage-based lower tiers.
2. Lower tiers are experiencing instability, which is bad, and the lowest ones have it the worst.
3. Making rises and drops separate could help lower tiers.

I'd really appreciate hearing others' thoughts on this pattern, my proposal, and experiences with the different sides of the issue, especially from those who drive lower tier development.

 

[BigFatMantis]

This may sound a bit cynical but it is a genuine question I have - is usage based tiering supposed to care about “destabilizing” lower tiers? Or is it a very black and white concept in tiering where the usage dictates what it is? Personally I think the more we doctor up usage based tiering, the less authentic it becomes. Having different cutoffs seems weird to me if it’s supposed to be pure usage based tiering where “above x%” is supposed to mean it belongs in the higher tier. That being said this proposal at least makes more sense than the freeze in rises if we stick to it.

 

[missangelic]

This may sound a bit cynical but it is a genuine question I have - is usage based tiering supposed to care about “destabilizing” lower tiers? Or is it a very black and white concept in tiering where the usage dictates what it is? Personally I think the more we doctor up usage based tiering, the less authentic it becomes. Having different cutoffs seems weird to me if it’s supposed to be pure usage based tiering where “above x%” is supposed to mean it belongs in the higher tier. That being said this proposal at least makes more sense than the freeze in rises if we stick to it.

From my perspective, I do think the goal is to have generally stable lower tiers during current gen. Policies like having full tier shifts every three months instead of every month, abolishing quickrises, and making the usage threshold more elite reflect this.

 

[Arcticblast]

I don't have the mental energy right now to really dig into this post, but I do have one immediate concern - if a UU Pokemon is able to hold, say, 6.5% usage in OU for multiple tier shifts in a row, wouldn't this proposal just keep an unquestionably OU Pokemon from actually rising to OU?

 

[shooting star]

Volatility is a feature, not a bug. Lower tiers by nature will always have to contend with metagame trends in higher tiers and adapt accordingly, they are never meant to be “stable”. As acknowledged, the proposal merely serves to delay the inevitable, which I struggle to understand how lower tiers will benefit from this. If Iron Crown and Moltres are always meant to rise to OU, that essentially means 3 more months of wasted development in the lower tiers since under the proposal, they will only rise in October 2024 instead of July 2024.

Moving on to the technicalities, most of the shifts are generally “legitimate” because higher-level players are more adept at identifying lower tier mons that are able to function well in their tier, after all these players are the ones responsible for influencing the tier shifts. Lower ladder takes a longer time to identify which mons are good, for example if they see in the builder that Magnezone is NU, they will assume that it is not good in RU. In the screenshot below, which shows the 1500 usage stats (i.e. mid-low ladder) from April to September 2024, Magnezone, Noivern, and Reuniclus immediately saw a substantial increase in usage after July shifts simply because they are classified as RU instead of NU in the builder.

Lastly, I think it is time we view rises as something to be embraced and not feared or nerfed. All this talk about destabilization and so-called tier cannibalism are characteristic of lower tiers, and I think it is a fool’s errand to engage in convoluted tiering that ironically may not even be in the interest of lower tiers. Furthermore, rises are not necessarily a bad thing. In fact, after July 2024 when NU got “cannibalized”, prominent NU players reacted positively (see below) that the new meta is “much better” than the previous meta where Krookodile and Slowbro dominate. And even currently in PU, despite being “cannibalized” twice, the general playerbase adapted accordingly and likewise also found the current meta to be enjoyable, as observed in the PU Discord.

Positive reactions

The only flaw of usage-based tiering is the idea of one person single-handedly make an unviable mon rise in a ladder that sees general inactivity (and thereby making it more susceptible to usage fluctuations) such as SS NU Hitmontop, and flavour of the month teams such as ORAS OU Azelf. But other than that, usage-based tiering is working just as intended and therefore I do not support any policy aimed at nerfing rises.

 

[missangelic]

I don't have the mental energy right now to really dig into this post, but I do have one immediate concern - if a UU Pokemon is able to hold, say, 6.5% usage in OU for multiple tier shifts in a row, wouldn't this proposal just keep an unquestionably OU Pokemon from actually rising to OU?

With the new way tiers would be defined under the policy, then definitionally no, because lower tier Pokemon would need to meet the more elite threshold for a staple Pokemon to make OU status rather than being a Pokemon that hangs out in the buffer zone near the drop threshold. The “cutoff” for OU would be understood as “Pokemon that don’t start OU or stay OU and keep consistent usage or reach a staple level of usage”.

The current system already has some of these quirks, such as Pokemon exceeding 4.52% usage in one or two months, even the most recent months, but failing to rise. Pokemon can even hit as low as 2.3% usage and still not quickdrop to a lower tier after two months.

I get what you mean though, as it would be a little awkward to say “this Pokemon is above the drop threshold, and it’s used more than some other OU Pokemon, but it’s not OU because it didn’t start or stay OU”. Still, I think that allows for more clarity when discussing how much something is actually used. We have Pokemon in this current system, such as Blissey, that are often on the cusp, so right now we currently acknowledge Pokemon that are questionably OU, which causes some of the instability.

Part goal of this kind of policy would be to get ahead of needing a bandaid solution like freezing rises altogether during the last two tier shifts. That kind of rise-freezing policy is a much bigger contradiction to the goal of lower tiers reflecting the usage trends of higher tiers.

 

[missangelic]

Volatility is a feature, not a bug. Lower tiers by nature will always have to contend with metagame trends in higher tiers and adapt accordingly, they are never meant to be “stable”. As acknowledged, the proposal merely serves to delay the inevitable, which I struggle to understand how lower tiers will benefit from this. If Iron Crown and Moltres are always meant to rise to OU, that essentially means 3 more months of wasted development in the lower tiers since under the proposal, they will only rise in October 2024 instead of July 2024.

Moving on to the technicalities, most of the shifts are generally “legitimate” because higher-level players are more adept at identifying lower tier mons that are able to function well in their tier, after all these players are the ones responsible for influencing the tier shifts. Lower ladder takes a longer time to identify which mons are good, for example if they see in the builder that Magnezone is NU, they will assume that it is not good in RU. In the screenshot below, which shows the 1500 usage stats (i.e. mid-low ladder) from April to September 2024, Magnezone, Noivern, and Reuniclus immediately saw a substantial increase in usage after July shifts simply because they are classified as RU instead of NU in the builder.

View attachment 726596
View attachment 726597

Lastly, I think it is time we view rises as something to be embraced and not feared or nerfed. All this talk about destabilization and so-called tier cannibalism are characteristic of lower tiers, and I think it is a fool’s errand to engage in convoluted tiering that ironically may not even be in the interest of lower tiers. Furthermore, rises are not necessarily a bad thing. In fact, after July 2024 when NU got “cannibalized”, prominent NU players reacted positively (see below) that the new meta is “much better” than the previous meta where Krookodile and Slowbro dominate. And even currently in PU, despite being “cannibalized” twice, the general playerbase adapted accordingly and likewise also found the current meta to be enjoyable, as observed in the PU Discord.

Positive reactions

View attachment 726599View attachment 726600

The only flaw of usage-based tiering is the idea of one person single-handedly make an unviable mon rise in a ladder that sees general inactivity (and thereby making it more susceptible to usage fluctuations) such as SS NU Hitmontop, and flavour of the month teams such as ORAS OU Azelf. But other than that, usage-based tiering is working just as intended and therefore I do not support any policy aimed at nerfing rises.

I agree (and just said in my last post lol) that changes are a feature and not a bug of lower tier tiering. However, that’s never been a hard rule against tweaking this system so it works better for lower tiers.

My goal isn’t to completely upend this system but instead to make it less disruptive and more predictable. This is why I took the time to indicate where tier shifts would have happened anyway. We can already observe in the current system that there's tier shifts that are inevitable based on one or two months of usage. However, the threshold for something to become an inevitable rise would actually be much higher in the system I’m suggesting, and slowing it down and not having it show up in the builder would also influence rises that went on to become “legitimate”. I don’t want lower tiers to prevent staple Pokemon from rising, but I do want lower tiers to be less prone to losing their staples when tiering is going on and Pokemon are fluctuating between 4% and 5% usage on a monthly basis.

A tier becoming better after rises take away oppressive presences is actually a sign that tiering action was probably warranted. Slowbro had a suspect test and didn’t meet the threshold for a ban, which contradicts this idea that there was actually majority support for it leaving NU. If a Pokemon is banworthy, then tiering councils should take appropriate action on it. Having fewer rises actually gives these councils more control and reduces “lame duck” periods where there’s not enough time to conduct a suspect, leading people to just wait out a metagame until shifts come. I’ve actually received and seen that the PU metagame was pretty impacted by shifts, and my experience playing ZU has been that the community hasn’t enjoyed adapting to the loss of its staples.

This kind of decoupling of rises and drops would also help with the “Hitmontop effect” you’re describing. It would take an increased amount of dedication and play over this period of time to rise Pokemon that aren’t staples into a higher tier, which I think is a good outcome. However, it’s hard to really claim that this system is working as intended when it’s in contradiction with what was outlined in the policy that kicked off the changes for Generation 8, had to be corrected toward the end, and then continued into Generation 9 unedited.

 

[Heatranator]

I don't usually post in Policy Review simply because I don't know much about the topics being discussed here, but this is a topic I do know about.

Rises have screwed over a bunch of tiers, the prime example being PU earlier this DLC, with it losing basically all of its defensive pieces, forcing an incredibly unstable meta to emerge that while it is now good, is still liable to being screwed over if anything is taken (for example, if registeel is taken from NU, then copperajah does have a legitimate chance of rising from PU, which would cause a major shift in the tier because copperajah is one of only two good steel types in the tier). This change wouldn't make it impossible for a pokemon to rise to a tier, but rather make it so that tiers are more adjusted for when rises occur, meaning that there is a lot less instability in these tiers. I don't mind Cobalion rising from RU to UU, that is a legitimately good pokemon in UU that wasn't a meta stable. But the bunch of PU pokemon that rose should indicate how utterly ruinious rises can be. This proposal isn't about stopping rises all together, but making them more managable for lower tiers so they can stabalise a lot easier.

A side benefit to this proposal would be that it is much harder for one individual or a few select individuals to forcibly rise a pokemon to a tier that it does not deserve. We saw it this gen in UU in January 2024, where one showdown account (who I have to assume is a bot due to the sheer amount of games) forced Indeedee, Gallade, Maushold, Armarouge and Donphan to rise to UU, some like Gallade which absolutely did not deserve to rise. UU did eventually give everything back, but Indeedee in particular, while not a horrible mon in UU, did take a while to drop. Only drop to PUBL where it has stayed. I don't think I need to recount the time in SS where Hitmontop rose to NU despite the fact that it was utter, and pardon my language, dogshit. Stopping this would be a good thing.

Edit: Nevermind, Indeedee did drop next shift with all the others. That's my b.

 

[ishtar]

As a PU leader, and someone who has been very actively involved in these discussions as recently as a few weeks ago when I posted a thread about usage tiering in the leaders subforum, I can safely say that the subject of usage is one that most lower tiers TLs have expressed a great amount of issues with in this current gen. This post I wrote was suggested to be posted in leaders, which is why it is not currently viewable to everyone. The conclusion with changes done via dialogue through the thread and CL chat will be presented in the future. This exact suggestion without clear numbers was thrown around and considered but it does not seem to be one of the more popular ones.

Now the reason for this post in particular from me is cause I disagree on a philosophical level with some of the replies that talked about the status of lower tiers within the system:

Volatility is a feature, not a bug. Lower tiers by nature will always have to contend with metagame trends in higher tiers and adapt accordingly, they are never meant to be “stable”.

The utilization of "feature" here implies that volatility is something that the system is designed to encourage, which is not the case when talking about usage. Volatility is a consequence of our system, which also means that finding ways for us to reduce the impact of its consequences is still a net positive and something we actively seek out. The fact that our tiers are never meant to be stable is also not correct, but this point might be a difference in semantics. If you consider any individual change to a tier as sign of instability, then I agree that stability is never possible due to drops or rises, but when I think of a tier being stable my definition is much more closely connected to a healthy state of our metagames. This is a spectrum, not black and white. The issue arises when our changes are so drastic that we can no longer move the needle in the correct direction or are forced to do so in excruciating manner every 3 months due to massive losses.

My entire point with this is that I can safely say without feeling like I'm putting words in anyone's mouths that lower tier TLs do seek out stable metas for our tiers and we have often achieved so for many generations in the past. It is also true as stated in the OP here that this has become increasingly more difficult in CG and that we've been presented with new problems in SV.

I also particularly dislike the utilization of logs, which are often taken out of context, and that include people who I'm well aware have expressed issue with current tiering and have shown interest in the subject of changing it for the sake of our tiers. I could go on about how PU in particular has been hurt by usage and how it has worn out council members as well as community members in very different ways, or the stuff I've heard from other leaders and other communities, but I genuinely think that it'd be a lot better if we just let people talk for themselves about these problems rather than utilizing random words from vastly different time periods. Showcasing positive interactions of our tiers or saying that some people are enjoying them is also slightly disingenuous. I am forever grateful that our community has many people that enjoy PU but that does not take away from the challenges that we have come across nowadays as well as our genuine desire to improve that enjoyment further. Someone liking how things are nowadays does not take away from the issues at hand the same way that someone's enjoyment doesn't have to be hampered by someone's negative opinions. This issue is not black or white and I sure am glad that there's people who are extremely happy with lower tiers right now, but I do not see how this proves a point against changing usage.

I wanna touch upon two different points: lower tiers are affected in very different ways across all the way down, but the ones further away from OU are usually the ones that get screwed over the most, which is why PU has lost 15 Pokemon and gained 2 in the last 9 months (not counting immediate QBs and 2 drops that went straight to UR). It is true that embracing of rises is a part of our reality, but there has to be a point where we notice that these consequences are affecting not only our tiers but our playerbases in a very negative way. I do not wish to embrace this!

This may sound a bit cynical but it is a genuine question I have - is usage based tiering supposed to care about “destabilizing” lower tiers?

My second point is in relation to the above question: In the past we've removed quick rises, we've changed the threshold, added DLC changes, all which have served to and work in favor of lower tiers. I do not understand why a system ran by human beings in service of a major community would decide that they should not care about the new issues that are being presented to it's subcommunities. That is not to say that these systems shouldn't be regulated, but they also exist and suffer changes for a reason, that reason relates to how usage affects tiers, these tiers are played by people. This isn't to say that I believe that this specific proposal is one I wholeheartedly agree with either. More on that in the future...

Lastly, I wanna make one final point regarding usage since a lot of people believe that no matter the amount of changes, we will always have "something to complain about". I agree! I think that the issues we had pre-Gen 7 with usage are not the same as the ones we had in Gen 8 or even now, and that's fine! Our realities shift, our issues also do, but that does not mean that we should not look for solutions.

 

[Lizzie]

I really feel like this is an artificial solution to a real problem that brings its own set of issues with it. My own perception is that the problem here is that every minor shift in higher tiers has a trickle down effect that increases in size the lower you go, and I think the original post largely agrees with me on this. The problem with increasing the threshold for rises, is that you will still get these shifts naturally as metas progress and tiering action is taken as required. You might lessen the effect a bit, but this comes at the expense of tiering accurately portraying usage. As an example, the last 10 mons to rise to OU are listed below:

Blissey
Pecharunt
Araquanid
Zapdos
Iron Crown
Moltres
Garganacl
Deo-S
Iron Moth
Ribombee

Some of these are pokemon who yo-yo up and down a bit depending on trends, but many of them have gone on to have significant ou usage. Not one of these pokemon would've initially risen with a 6.7% cutoff. And while many mons have since exceeded this cutoff, theres no guarantee this would've happened had they not risen originally, as being in a lower tier generally makes things less likely to be used.

To me major tier shifts are inherent in the system, and whether they're a flaw or a feature is irrelevant because you can't artificially prevent them. I hate to say that all you can do is find better ways of dealing with major shifts because I'm well aware many councils are putting a lot of work into doing this and still struggling, but I don't view this as a viable solution.

 

[missangelic]

I really feel like this is an artificial solution to a real problem that brings its own set of issues with it. My own perception is that the problem here is that every minor shift in higher tiers has a trickle down effect that increases in size the lower you go, and I think the original post largely agrees with me on this. The problem with increasing the threshold for rises, is that you will still get these shifts naturally as metas progress and tiering action is taken as required. You might lessen the effect a bit, but this comes at the expense of tiering accurately portraying usage.

The way tiering is done is already arbitrary, and it's been changed in the past (3.41% to 4.52%) with that in mind. I agree that every change in higher tiers reverberates into lower tiers and that it has a progressive effect, which is why I went out of my way to highlight it. I still think that this type of system can be reformed though so lower tiers can have an easier time developing, organizing, and existing as current gen metas. My goal with reforming it isn't to completely abolish that major shifts or changes happening, but instead to make them more manageable while we're working on tiering. My perspective is that this vacuum effect is a mixture of the way DLC2 tiering is done, unforeseen effects of reforming the older tiering system, and the patterns of the tiering system itself being more obscure because they happen over longer periods of time. I get being hesitant to act because it's a change that's deliberately trying to account for this effect, but I don't think that makes it unnatural to tiering when the recent chain of tiering policies related to usage thresholds have been to intentionally make tiering lower tiers easier.

If we used a new system, we'd reframe how we'd look at usage. Under the proposal I suggested, OU would portray Pokemon that had stable and solid usage over a longer period of time (the shifts since they've established themselves as OU) AND Pokemon that had staple-level usage over a shorter period of time (the 3 months before a tier shift). We've adjusted the way we evaluate usage before by raising the usage threshold, making the definition of what's in a tier more elite, and we've also changed from having weighted usage that prioritized the most recent month to just weighting all the months the same.

 

[SoulWind]

I don't play lower tiers enough to have an specific opinion about any of them in particular, but when I do, this always ends up being an irritating problem (mainly thinking about Slam and SCL). You spend 3 months playing a certain metagame only for it to be screwed later by whatever ladder trend is up and then you are forced to reset again.

It's hard to find logical reasons as to why someone would want constant fluctuation (the mentioned above yo-yo effect is more than evident) instead of an stable tier that is improved from time to time without the influence of external factors and I do not really understand why lower tiers are treated different from OU in this regard.

I believe that after a lower tier is formed and a few months have passed, tier should stay frozen and it would be up to the respective TLs and Councils to take any form of tiering action for them.

Tier shifts are straight up a mistake, specially considering it's ladder usage (an extremely awful criteria for competitiveness) what causes them.

 

[Lizzie]

The way tiering is done is already arbitrary, and it's been changed in the past (3.41% to 4.52%) with that in mind. I agree that every change in higher tiers reverberates into lower tiers and that it has a progressive effect, which is why I went out of my way to highlight it. I still think that this type of system can be reformed though so lower tiers can have an easier time developing, organizing, and existing as current gen metas. My goal with reforming it isn't to completely abolish that major shifts or changes happening, but instead to make them more manageable while we're working on tiering. My perspective is that this vacuum effect is a mixture of the way DLC2 tiering is done, unforeseen effects of reforming the older tiering system, and the patterns of the tiering system itself being more obscure because they happen over longer periods of time. I get being hesitant to act because it's a change that's deliberately trying to account for this effect, but I don't think that makes it unnatural to tiering when the recent chain of tiering policies related to usage thresholds have been to intentionally make tiering lower tiers easier.

If we used a new system, we'd reframe how we'd look at usage. Under the proposal I suggested, OU would portray Pokemon that had stable and solid usage over a longer period of time (the shifts since they've established themselves as OU) AND Pokemon that had staple-level usage over a shorter period of time (the 3 months before a tier shift). We've adjusted the way we evaluate usage before by raising the usage threshold, making the definition of what's in a tier more elite, and we've also changed from having weighted usage that prioritized the most recent month to just weighting all the months the same.

Idk if I'm missing something but my understanding is that your original proposal had no mechanism to allow for rises at a treshold below ~6.78% at any point. You seem to imply here that rises like this could happen over a longer period of time though? I have less opposition to this idea if mons consistently hovering at ~5-6% usage rise after a longer period of time.

 

[ryan]

it has been a very long time since i led and was active in a lower tier, but i've always found them more interesting than the flagship because of the way that they fluctuate over time, including drops and rises. volatility is the price of fluctuation, but that doesn't mean that you cannot work to mitigate one without harming the other. separating ceilings and floors is a very interesting way to address that. it not being a perfect solution is not reason to ignore it, though i'm interested to hear how tier leaders are thinking about the problem.

it is inevitable that as bans occur, higher tiers begin to inflate. if a heavily-used pokemon leaves the tier, its former usage is likely to be divided across a wider range of pokemon, and typical checks/counters and synergistic/compatible teammates suddenly begin to drop off as well. it takes time to experiment and innovate, and doing so requires a healthy community with creative and talented players who don't simply rip off teams from their friends. many people don't fit that bill and prefer a tier that rewards them for sticking around to one that forces them to hang around waiting for other people to solve a new metagame for them every few months. as the playerbase declines, relative skill level falls, which makes innovation all the more challenging and creates two separate metagames: the one that good players play amongst one another and the one that everyone else plays.

all of this limits the ability for a tier leader to embrace volatility, even if you can as an individual player. some players thrive in those metagames, but there's serious risk to running a high-friction tier when so many people want to log in, click find a battle and have fun without getting destroyed for not having done their homework.

 

[ThatOneApple]

I would like to speak in support of this change, rises are always kind of a pain to lower tiers as many of said before. With the best example given being PU losing a lot of tier staples. I do not believe rises should stop as a whole however, as if a mon is good and gets solid usage in a higher tier it should be in that higher tier to reflect that. However, as others have pointed out, there are some scenarios where mons that (at least seemingly going off of VRs) yo-yo around based on the ladder trends rather than any major viability. Examples pointed out by others include Meowscarada and especially Blissey in OU. Another scenario is that in lower tiers where the activity may not be super consistent, one person who is super dedicated can drag things up such as the example Heatranator brought up where three (arguably four but indeedee was at least viable) mons got brought up to UU that really did not deserve the placement at all. To cite another example however, there was also the time where really early in the gen, Hisuian Zorua rose to NU. This did not cause any metagame problems for the lower tiers but kinda illustrates the fact that singular people actually can cause issues where mons rise to tiers they dont belong in, and that doesn't even get into the whole hitmontop situation from SS.

Generally speaking i dont think volatility from shifts is a good thing, not sure what else i can say as im mostly just repeating what others have said tbh. Regardless, even if i dont play the tiers where this really happens, tiers should not have to rebuild themselves due to a ladder trend in a higher tier. Keeping the threshold for drops the same is also a good idea because you could argue that drops can bring some volatility even if its not as bad. Separating the thresholds for rises and drops also kind of gives the impression that a mon would need to actually prove its usage to rise back up rather than just, as others have said, yo-yo'ing up and down due to constantly being around 4.52%. Honestly not sure I really added much to this conversation but I agree with the proposal.

 

[Marty]

This conversation has been had before: https://www.smogon.com/forums/threa...gher-rise-cutoff-the-hitmontop-issue.3701773/

In my view, suggesting that the rise and drop thresholds be different from each other is a complete non-starter unless major changes are made to the fundamentals of the whole tiering system. Arcticblast touched on the absurdity early in this thread, but please also read my two posts in the thread I linked that go into more detail about it. (https://www.smogon.com/forums/posts/9214693/ and https://www.smogon.com/forums/posts/9215164/)

 

[missangelic]

This conversation has been had before: https://www.smogon.com/forums/threa...gher-rise-cutoff-the-hitmontop-issue.3701773/

In my view, suggesting that the rise and drop thresholds be different from each other is a complete non-starter unless major changes are made to the fundamentals of the whole tiering system. Arcticblast touched on the absurdity early in this thread, but please also read my two posts in the thread I linked that go into more detail about it. (https://www.smogon.com/forums/posts/9214693/ and https://www.smogon.com/forums/posts/9215164/)

The thread of the previous conversation also discussed changing the tiering system and included a proposal to raise the rise threshold, but the OP outlines that it was for a different purpose. The previous topic of whether "bad" Pokemon should or should not rise isn't relevant to this conversation. Bad Pokemon being in higher tiers is not as far-reaching or important as the consequences of rises destabilizing tiers. We're having to have this conversation again because we're seeing the same effects of Generation 8 of rises destabilizing tiers when the follow-up policy of freezing rises altogether wasn't brought forward into SV. This is a pretty noteworthy gap in the tiering policy for Generation 9, where we're doing the same thing again as though we're not going to get the same result or worse with the longer Generation lifespan.

One of the big questions you brought up in your posts that you linked to is "how do you explain [a Pokemon not being in a tier and still having higher usage than a Pokemon in a tier] to a new player, given that Smogon claims to tier by usage?" The answer to how this would be explained if rises and drops are decoupled: any tierlist includes the groups of Pokemon that either have major recent usage or consistent long-term usage.

Most new players frankly do not immediately grasp what usage even is. This entire point of the analysis is to demonstrate that even people who are actively involved in tiering aren't conscious of the multi-tier and long-term trends of tiering. The way tiering is set up is not intuitive, and that's just considering current gen. "Pokemon in a tier are the ones a player on a competitive but not too competitive slice of the ladder has a 50% chance to encounter in a selection of 15 battles. It used to be 20 because that was the average number of games a player would be expected to play in one session, but we switched to 15 because it made lower tiers more stable. Also it's averaged across three months, even if the tierlist wasn't the same throughout that three month period because of tiering action and quickdrops. Also quickdrops are a thing, but not quickrises, because they're destabilizing." Not only is the way we tier not intuitive to newcomers, but intent of "we tweaked it to make this system better for lower tiers" was and is already baked into multiple parts of the tiering process.

Tiering gets even less intuitive if you compare SV to any old gen. SS had its rise-freezing policy for the sake of lower tiers, one that was in complete contradiction to the fundamental of tiering that Pokemon with higher usage would go to higher tiers. DPP to SM had a different threshold for changes, and they weighted average usage towards the most recent month. RBY to ADV aren't even usage-based and still undergo active tiering. However, we don't do rises for usage-based lower tiers in old gens, even when old gen OU ladders are more active than current gen lower tier ladders. We still reflect Pokemon that are "OU by technicality", but the UU community doesn't want them to drop at this stage.

I get being concerned about the optics of making the system more complex. However, I don't think making tiering, a system that is already beyond most newcomers, simple to understand is more important than the stated intent of tiering policy to make stable and enjoyable lower tier metagames for the people that actually play them.

 

[BigFatMantis]

However, I don't think making tiering, a system that is already beyond most newcomers, simple to understand is more important than the stated intent of tiering policy to make stable and enjoyable lower tier metagames for the people that actually play them.

Is this really the intent of usage based tiering though? If this was the intent of tiering policy, we would do away with usage based tiering altogether, since it obviously does not cleanly satisfy that intent. The point of usage based tiering appears to be much different, and having a stable and enjoyable lower tier metagame out of it seems to be secondary to the purpose of the usage based system itself defining tiers by usage.

 

[missangelic]

Is this really the intent of usage based tiering though? If this was the intent of tiering policy, we would do away with usage based tiering altogether, since it obviously does not cleanly satisfy that intent. The point of usage based tiering appears to be much different, and having a stable and enjoyable lower tier metagame out of it seems to be secondary to the purpose of the usage based system itself defining tiers by usage.

From Hogg's Tiering for Generation 8 post:
"last generation there were serious concerns with how disruptive tier shifts could be to lower tiers. While drops could always be handled via bans, rises could completely destabilize a tier, causing major impact... impact that then needed to be reset in a few months when the risen element fell again. Even a single popular sample team could end up causing major ripples in the tiers below due to nothing more than a passing trend. This was especially true if it occurred during the months weighted more heavily. While rises are always going to be a part of any usage-based tiering process, we wanted a solution that made them a bit less frequent... the end result will be smaller tiers overall, where rises are less frequent but drops are more common."

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."

It's clear from the stated intents of these policies and their implementations that the goal was to make lower tiers more stable and easier to tier while still maintaining usage-based tiering, and that would be my goal with a policy proposal too, even if it doesn't turn out to be decoupling rises from drops.

 

[Oculars]

More rises than drops is a feature not a bug, the tiers start out at a power level too high and as things are banned to ubers the next best stuff from UU rises up and then RU to UU and so on, tiering has always leaned on the more bans then unbans side and unless that somehow changes more rises than drops will be the standard across all usage based tiers. I dont think this is something that needs messing with as we start to lose the definition of what lower usage based tiers are supposed to be if we go tweaking the numbers to favor one side of the equation rather than the other.

 

[missangelic]

More rises than drops is a feature not a bug, the tiers start out at a power level too high and as things are banned to ubers the next best stuff from UU rises up and then RU to UU and so on, tiering has always leaned on the more bans then unbans side and unless that somehow changes more rises than drops will be the standard across all usage based tiers. I dont think this is something that needs messing with as we start to lose the definition of what lower usage based tiers are supposed to be if we go tweaking the numbers to favor one side of the equation rather than the other.

There’s a difference between “rises > drops is a feature” and “rises > drops is an emergent effect”. Rises outpacing drops is not a feature, our tiering policy describes intentionally pursuing more drops than rises. The way tiering is set up is already arbitrary, so it’s actually a good thing that we can change it to benefit lower tiers and their playerbases. Several users in this thread and the usage-based tiering discussion thread have already come forward with their frustrations with this pattern on all sides, including tiering councils, tournament players, and metagame resource contributors, so I think this is worth addressing instead of throwing our hands up and going “there’s nothing we can/should do about it because it would mean change”.

 

[Lucario]

Let's be real, raising the threshold for rises makes things too complex. I'll use the most recent usage stats (specifically the stats from March) for my example. In OU for last month we see Galarian Weezing and Scizor above the current 4.52% threshold, being at 5.45% and 5.40%, respectively. If this was going off of one-month shifts then we would see them rise to OU. However, if it was a one-month shift with a 6.78% threshold to rise then they would remain UU, despite having more usage in OU than some actual OU Pokemon, like Enamorus, Walking Wake, and Iron Crown. It does not make sense to have those latter three Pokemon be OU while not letting Galarian Weezing and Scizor into the tier.

Another argument I've seen is re-introducing the freeze on rises. This is a terrible idea for the current state of SV tiering. The reason it worked in SS is because we knew Gen 9 was right around the corner, so freezing any rises would stop the tiers from randomly becoming unplayable as old gens. We currently have no release date for Gen 10, in fact, we know nothing about it. We can only assume it will release in November of 2026, which means we have at least 19 months left of SV, which is another 6 shifts. Stopping any rises from occurring now will mean that the tiers will be unrecognizable by the time SV ends its run as current gen.

I also want to acknowledge the lack of understanding behind tier development in the OP. I play NU so I will be using NU as an example. Yes, we lost 2 Pokemon in April, 7 in July, 3 in October, and 1 in January. Obviously, all of those shifts changed the tier, but they changed the tier in a positive way. We first lost Chansey and Overqwil, the latter of which was really strong on Rain teams, which weren't too popular at the time, but seeing how Sun broke the tier, Rain could've done the same. Then in July we lost 7 pokemon, one of which was Slowbro which had just survived a suspect test. Shortly before the shifts people started realizing that Reuniclus and Magnezone were really good, Thundurus was also gaining more traction that time. It's safe to say that 5 of those 7 Pokemon would have ended up getting banned at some point, but instead of wasting a lot of time tiering them individually, RU stole them, letting us develop a new meta. Since those rises, NU has been able to develop greatly, banning several threats while seeing new tech arise from NUCL 2. I'm saying all this because rises aren't strictly negative, if you look at them that way then your understanding of lower tiers should change.

Lastly, I want to use this as an opportunity to bring up a different approach to this. Instead of having shifts every 3 months, lower it to every 2 months. The NU community has known for a while now that we would be losing Registeel, and as of a couple hours ago, it became reality. A tier knowing it will lose a Pokemon just 1 month into the 3-month cycle sometimes halts development and lowers overall enjoyment and moral. "Oh no, we are losing Registeel in 2 months, I hope you're ready for x and y to take over the tier," is really annoying to read. It's also not a fun experience seeing "x could drop from RU to NU" every month just for it to not drop. For a while now NU has been expecting Pokemon that "could have dropped" to drop just to see none of them. Azelf, Reuniclus, Politoed, Barraskewda, etc. have been under the threshold several times just to go back up in the final month. If a Pokemon has little to no viability in the tier they're in while also teetering over and under the 4.52% then it's clear it should drop, but that cannot happen because people will look at the potential drops just to "save" these Pokemon. 2-month shifts means tiers can develop more quickly while also getting more drops.

 

[missangelic]

Let's be real, raising the threshold for rises makes things too complex. I'll use the most recent usage stats (specifically the stats from March) for my example. In OU for last month we see Galarian Weezing and Scizor above the current 4.52% threshold, being at 5.45% and 5.40%, respectively. If this was going off of one-month shifts then we would see them rise to OU. However, if it was a one-month shift with a 6.78% threshold to rise then they would remain UU, despite having more usage in OU than some actual OU Pokemon, like Enamorus, Walking Wake, and Iron Crown. It does not make sense to have those latter three Pokemon be OU while not letting Galarian Weezing and Scizor into the tier.

I don't think using one-month shifts to describe a three-month shift is a good fit here, but I'll still address this. If rises and drops were decoupled, you'd have to look at the bounds of tiers differently. Yes, there would be periods of time where Pokemon not in a tier see higher usage than some Pokemon in said tier. This would be because the definition of a tier would be Pokemon that have staple usage in said tier or Pokemon that once had staple usage in the tier and still continued to have adequate usage.
We already have these contradictions in the 3-month average system. A Pokemon can see greater usage than another Pokemon across one or two months and still end up not being in the same tier during the 3-month average tier shift because it dipped in usage during one or two of the other months. The way we explain these situations is "the Pokemon did not get the necessary average usage, and that's what we tier by". The explanation for why Iron Crown would be OU while Galarian Weezing wouldn't be under the decoupled system would be "Iron Crown has stable and adequate long-term usage in OU and had staple-level usage, while Galarian Weezing did not have long-term or staple-level usage, and that's what we tier by." This really isn't more complex than everything that already goes into usage-based tiering that I already listed in my reply to Marty.

Another argument I've seen is re-introducing the freeze on rises. This is a terrible idea for the current state of SV tiering. [...] Stopping any rises from occurring now will mean that the tiers will be unrecognizable by the time SV ends its run as current gen.

I agree that freezing rises at this point in time is not be the move for SV. However, I do have to point out that keeping more Pokemon in place, especially Pokemon that would otherwise rise, would make the metagames at the end of SV more similar to the metagames now compared to having rises to the end or near the end.

Obviously, [rises from NU] changed the tier, but they changed the tier in a positive way. [...] It's safe to say that [5/7 rises] would have ended up getting banned at some point, but instead of wasting a lot of time tiering them individually, RU stole them, letting us develop a new meta. Since those rises, NU has been able to develop greatly, banning several threats while seeing new tech arise from NUCL 2. I'm saying all this because rises aren't strictly negative [...].

You're missing the point of stability in this discussion. Yes, some rises are neutral, positive, or manageable for tiers. However, most rises inarguably change the metagames they come from. Whether change leads to a better or worse metagame afterward still has the stressful effects I've described previously.
It's worthwhile to take action on something that has negative effects even if it has some positive or neutral effects (sort of like in tiering!). "Some rises were good, in my opinion" is myopic when this discussion is about the overall cross-tier trend of rises driving long-term instability and the lack of tiering policy modification going into Generation 9 after seeing how Generation 8 turned out.
As I talked about before in this thread when addressing rises out of NU, it's not a good argument to say "rises took away Pokemon that were problematic, making the metagame worse, or would have gone on to become banworthy". It's the responsibility of tiering councils to act appropriately when there are Pokemon they recognize as being actionable, so it should not be the job of rises to take away Pokemon that are overpowered for the tier. That's all before considering that Pokemon that are problematic or banworthy in a lower tier don't always go on to rise to a higher tier.

[...] Instead of having shifts every 3 months, lower it to every 2 months. [...] A tier knowing it will lose a Pokemon just 1 month into the 3-month cycle sometimes halts development and lowers overall enjoyment and moral. [...] If a Pokemon has little to no viability in the tier they're in while also teetering over and under the 4.52% then it's clear it should drop, but that cannot happen because people will look at the potential drops just to "save" these Pokemon. 2-month shifts means tiers can develop more quickly while also getting more drops.

Making tier shifts every two months is only going to make lower tiers even more volatile because months would be weighted more heavily for shifts, similar to how they were earlier in pre-DLC2 SV during 1-month shifts. 2-month tier shifts themselves would actually be even more vulnerable to players who want to rubber-band declining Pokemon into higher tiers, and that's putting aside that there are Pokemon that just genuinely do fluctuate around 4.52% instead of having orchestrated #SaveTyranitar campaigns. When the pattern in the post-DLC2 period is that there are more Pokemon are rising up to higher tiers than are dropping to lower tiers, having more tier shifts is only going to make the gap between rises and drops larger. I don't think you understand the short- or long-term outcomes of what you're suggesting with your proposal.
Given the responses to the effects of instability people have shared in this thread, I don't think 2-month tier shifts would be popular or useful. More volatility would place even greater burdens on lower tier communities. Councils would have to rush tiering action in the first month after shifts, as the second month could already anticipate Pokemon trending out of the tier. Metagame resource contributors would have to make even more updates, and most lower tier tier C&C groups are already strained as is with 3-month tier shifts. Tournament players would have to overhaul teams and builders more often because tours would be much more prone to having a tier shift intersect with them.
However, I do strongly agree that guaranteed-rise or near-guaranteed rise situations hurt enjoyment, activity, and morale (not that quickrises are a viable solution to this). This is why I think doing something about the destabilizing pattern of rises is important, as these lame duck periods are demotivating.

 

[PowerOfMemes]

I believe that rises are almost always disruptive to lower tiers and am in support of this proposal.

However, the OP’s point on preparing lower tiers for how drastic the drops might be feels weak, especially with Lucario bringing up lowered morale, which is very common in these situations. Additionally, people will simply either play the current metagame with the slightest of theorymonning (and usually then, it’s in reference to drops) for shifts or do little to nothing till shifts.

On the argument of arbitrariness or complexity, I disagree with that point. Definitely, there comes a point that you have to ask yourself “is this a usage-based tier, or is this entirely fabricated for people’s enjoyment?”. To me, though, changing the rise percentage but keeping the drop one doesn’t hurt here. You can argue it’s weird why a lower-tier Pokemon becomes OU when an OU Pokemon doesn’t need to prove itself just as much, but I see it as rising through the ranks. You work hard to get noticed and are treated accordingly. But, now, there’s not much else to do, and because of that, you won’t be kicked back down for not doing as much. Significant drop-offs should be noticed and acted on accordingly, though. If the disparity in the percentages between rises and drops truly is the issue, I am in support of raising both. While raises are painful, drops are never as bad. This is usually due to how councils often do their best to tier them as soon as possible, whether that be banning them or simply placing them on the tierlist. More drops would only require this to happen on a slightly larger scale. However, I do realize this can take up more of people’s time, and that’s why I am in support of keeping the rise percentage higher than the drop one.

To argue with people in favor of rises possibly being positive, I am on the bandwagon that councils are supposed to act on a Pokemon’s brokenness. I believe that Slowbro should’ve been banned and many others had the same sentiment, which is why we were content that RU was taking it anyways when the suspect test fell through. I could see how some of the other Pokemon could grow to be broken like Lucario says, but if they became so, then we act on them. We should take Occam’s razor into account. What’s more likely? A Pokemon being used a lot in the tier above, or a Pokemon being broken in one tier while being used a lot in tier above? The first, clearly. Sure, you could say I’m excluding neutral rises, but those are nearly always negative in nature (no matter how little so), since they remove a building option. For example, in the same tier shift that brought Slowbro to RU, Noivern rose there as well. Most people said it was a neutral rise, or not that important. However, it becomes painful afterwards when you want something that has the same attributes it did. In this case, building became more restrictive entry hazard control-wise because you were unable to use a hazard remover that also provided the bevy of resistances that the Dragon-type offers along with the wallbreaking Noivern’s moves and stats brought.

Is changing the tier shifts to have a higher rise percentage while keeping the old drop percentage an example of us influencing tiers’ rises and drops? Slightly, but what’s the point of a tier if building isn’t enjoyable and playing’s not fun? I’ve seen plenty of complaints on certain tiers having a million threats that aren’t outright broken but definitely extremely problematic, making you either have to go offensive to cover up the fact you can’t take opposing firepower too well or simply accept you have a so-so matchup into every threat. In these cases, there’s usually an impasse, but the best solution is to ban something and move forward until you hit a healthy metagame. Here, the issue is just a different side of the same coin: tiers are more volatile because they’re losing useful Pokemon simply due to them being used a bit much by the above tier. Obviously, if that Pokemon deserves to be there, it should be there, though Pokemon-of-the-week and hovering ones are all too common.

I’d like to end this off on clarifying that I come from the view of “if it’s bad / wrong, we can fix it later” when it comes to bans and the like, meaning—if this gets implemented and actually is pretty detrimental—I am in full support of going back and changing this. I don’t know how practical that is, though. And, unfortunately, all we can do is theorymon beforehand. Speculating myself, having a little test period (for preferably a few shifts, or simply one) would be nice, though, to see how it plays out, and it would let us be more educated on this for better input. Additionally, setting that precedent wouldn’t be that problematic, as it’s not like posts such as this are omnipresent. Once again, I know nothing of the practicality of that, but it’d be nice to see.

edit: grammar

 

[TheDuckChris]

It feels like every so often a post like this is made begging lower tiers to become these stable places where nothing ever rises and you can develop that one same meta to your heart's content.

Guess what guys OU, Ubers, past gens exist. Ever since lower tiers started tiering by usage in gen 5 it's been a feature that each shift (3 months, enough time to run a full tournament) you get a slightly different meta. Sure this can cause problems but it can also solve problems. Lower tiers get away with less bans because usage brings up the problematic mons.

Councils can do whatever they want to mitigate these issues. Faster suspects, Koko method, quickbans are all valid strategies. Additionally if the concern is about tournaments, then simply freeze the tier for the course of the tournament or even better, schedule your tournaments so they happen between tier shifts.

Apologies for the dismmissive tone but it really just feels like these posts take 1000 words in order to say "I choose to play a tier with shifts but would prefer for there to be no shifts"