Restructuring Tier Shifts

Quite Quiet

I need a kitchen knife that doesn't whisper to me
is a Site Content Manageris a Member of Senior Staffis a Community Contributoris a Tiering Contributoris a Contributor to Smogonis a Top Smogon Media Contributoris a Top Dedicated Tournament Hostis a Tournament Director Alumnusis a Battle Simulator Moderator Alumnus
TFP Leader
This is somewhat based on opinion, but I really think lower tiers would be better off returning to the 3-month-only tier shifts and discarding the quick shifts happening every month right now. Either with the release of Sun/Moon, or now.

Back in November/December 2014 during the time just after ORAS had came out, there were problems with usage stats missing and not reflecting the current metagame. So at that time, rather than ignore shifting, a different cut-off was used to shift pokemon based on a single month of usage rather than three. Fair enough, that was special and not a big deal. During 2015 though, this was kept in place to allow for Pokemon to quicker shift down to the metagames where they are "supposed" to be in, and means tiers may shift every single month rather than once every three months.

After nearly half a generation of doing this, and for the full lifetime of a game (it will be nearly two years, if you're wondering) I'm not convinced this is something that should be kept in place. First of all, and perhaps most importantly, this has kept every non-OU tier in a constant state of change with only very short periods of time between every change. Tiers barely get a chance to settle down before new pokemon enter, or staples leaves and the metagame changes again. Constant changes doesn't lend itself to properly allow a tier to explore options and develop, and as a result they will have to develop far slower, or not at all. Three months give a much bigger window for a tier to settle down, develop, and reveal eventual suspects, without risking a new pokemon dropping midway through a suspect test and making it essentially an entirely different metagame than the suspect started in (this has happened before).

One of the big counterarguments I've gotten from this is that it prevents a tier from being stale and keeps the tier fresh. This is purely opinion and there's no proof that the tier wouldn't develop and change either way, even if you didn't introduce new pokemon to it every month. Furthermore, cases like Mawile/Pinsir doesn't exist anymore either, as Megas and their base formes are tiered separately now so this doesn't impact any eventual bans of those anymore.


So yeah, I don't think the quick shifts were a healthy addition to tiering and I'd like to see them retired for Sun & Moon (or now).
 

phantom

Banned deucer.
I think it's fine the way it is. There are very few instances where a Pokemon quick rises/quick drops to a tier that wouldn't have happened anyway as a result of a banned ability/new policy change/new games/megas (in the past)/released event of some sort, and there are none that I can think of that are disruptive to the point where this needs to be abolished. Quick rises/quick drops just allows said drops/rises to happen at a more efficient pace as opposed to certain Pokemon being unnecessarily stuck in tiers longer than they normally would have. For instance, if we didn't do quick rises/quick drops, various Pokemon subject to the mega split policy change wouldn't have settled into RU/NU/PU up until now. The only reason tiers haven been in a "constant state of flux" is because of the mega policy change, and if it wasn't for the quick drops/rises, it would have took certain lower tiers even longer to properly settle when anticipating various drops several months in advance, so the notion that it keeps tiers from settling doesn't seem to hold water when the current system proved to be very useful in this instance. Besides, I find quick rises/quick drops have the potential to be extremely useful in diluting the initial pandemonium brought by new games/policy changes on various lower tiers and counterbalancing volatile shifts in usage that may occur during the third month. It just makes no sense that something experiencing extremely low usage like 1% or less should stay in a certain tier throughout the duration of the three month shift "just because".
 
Last edited:

Nails

Double Threat
is a Community Contributoris a Tiering Contributoris a Social Media Contributor Alumnusis a Community Leader Alumnusis a Battle Simulator Moderator Alumnusis a Past SPL Championis a Three-Time Past WCoP Champion
agreed with spirit. if tier shifts need a change it's staggering the months that the 3 month tier shifts update on so that mons can rise/fall faster, and every tier isn't updating at the same time. tiers shift too slowly as is, it doesn't take 3 months to adjust to the impact of 2-3 extra pokemon.
 

Quite Quiet

I need a kitchen knife that doesn't whisper to me
is a Site Content Manageris a Member of Senior Staffis a Community Contributoris a Tiering Contributoris a Contributor to Smogonis a Top Smogon Media Contributoris a Top Dedicated Tournament Hostis a Tournament Director Alumnusis a Battle Simulator Moderator Alumnus
TFP Leader
I think it's fine the way it is. There are very few instances where a Pokemon quick rises/quick drops to a tier that wouldn't have happened anyway as a result of a banned ability/new policy change/new games/megas (in the past)/released event of some sort, and there are none that I can think of that are disruptive to the point where this needs to be abolished.
On the contrary, there has been drops that have been extremely disruptive to the point it's ridiculous. Gallade dropped to NU in the middle of the Megneton suspect test, essentially changing the metagame to something entirely different than what the first half of the suspect test was working with. I'm not actively on any council, but the current system means that unless tiers start a suspect test within two weeks, any suspect test will not end before a new month, and potential drops that mess up a metagame. A month might be enough time to let a tier settle, but you cannot realistically (unless nothing dropped/rose) expect tiers to settle in a measly two weeks in order to hold a suspect test without putting up with new drops midway through the test?

For reference:
http://www.smogon.com/forums/threads/3537418/page-8#post-6251475
http://www.smogon.com/forums/threads/3539217/
 

phantom

Banned deucer.
On the contrary, there has been drops that have been extremely disruptive to the point it's ridiculous. Gallade dropped to NU in the middle of the Megneton suspect test, essentially changing the metagame to something entirely different than what the first half of the suspect test was working with. I'm not actively on any council, but the current system means that unless tiers start a suspect test within two weeks, any suspect test will not end before a new month, and potential drops that mess up a metagame. A month might be enough time to let a tier settle, but you cannot realistically (unless nothing dropped/rose) expect tiers to settle in a measly two weeks in order to hold a suspect test without putting up with new drops midway through the test?

For reference:
http://www.smogon.com/forums/threads/3537418/page-8#post-6251475
http://www.smogon.com/forums/threads/3539217/
That's more so the byproduct of the way megas and their base formes were tiered at the time and not solely the result of quick drops/rises. If you notice, nothing like this really happens anymore since the policy change, aside from Gurdurr moving up to RU during the Mega Abomasnow/Sharpedo suspect test where it was quite good at the time. Gurdurr wasn't a particularly game-breaking Pokemon in either tier, so it moving around a little didn't create any noticeable issues. Gallade moving into NU during the middle of a suspect doesn't seem particularly good, but I don't think that quick drops/rises are entirely to blame here given that they haven't caused any issues since megas and their base formes were tiered separately. This is an entirely avoidable situation anyway provided tiers don't start suspect test towards the end of the month. So I don't really see a reason as to why quick drops/rises should be removed with the points you presented given the fact that A) tiers are in control and can avoid their suspects being disrupted by timing them properly and B) there haven't been very many significant shifts and none that are disruptive since megas were tiered separately, and C) there are benefits to quick drops/rises such as allowing Pokemon subject to various anomalies to shift to a proper tier at a more efficient timeframe and allow tiers to settle sooner during new game releases that would be a shame to get rid of.

edit: I confused prior tier shifts my b, but the general premise is still there
 
Last edited:

Lord Death Man

i cant read
is a Community Contributoris a Tiering Contributor
Suspect metas shouldn't be able to change, in my opinion; Gallade shouldn't have dropped (or rather, been usable) until the suspect ended, in my opinion. I think the current system is fine, though staggering it sounds cool, since one thing I currently dislike is how suspects seem to happen in clusters that feels like it encourages uninformed players to share teams and guess.

The only issue with delaying drops is that stuff might not have quite enough usage to stay in the tier, but I think that's a reasonable risk to preserve the validity of the suspect meta.
 
  • Like
Reactions: g

Quite Quiet

I need a kitchen knife that doesn't whisper to me
is a Site Content Manageris a Member of Senior Staffis a Community Contributoris a Tiering Contributoris a Contributor to Smogonis a Top Smogon Media Contributoris a Top Dedicated Tournament Hostis a Tournament Director Alumnusis a Battle Simulator Moderator Alumnus
TFP Leader
Gurdurr is still RU, by the way. It never dropped down to NU again after it moved to RU for the Abomasnow/Sharpedo test, so I don't know where you go that from.

As for the rest: The entire point of the post you quoted was that "not holding a suspect test towards the end of the month" basically means it has to start two weeks into a month not to clash with shifts, and if something moved tiers have very little time to settle before they miss that opportunity.

If stalling drops until the suspect test is what people would prefer, I could accept that I suppose. At least that'd make tiers able to start a test the 25th and not worry about something potentially dropping that ruins the test, which the current system isn't doing.
 

Nails

Double Threat
is a Community Contributoris a Tiering Contributoris a Social Media Contributor Alumnusis a Community Leader Alumnusis a Battle Simulator Moderator Alumnusis a Past SPL Championis a Three-Time Past WCoP Champion
i mean then people are voting based on whether the mon is banworthy in a past format. that doesn't seem like good tiering policy. i think the play is to start suspect tests earlier in the month or have shorter suspect tests. starting a test on the 22nd, having a week of qualification and then a 2 day vote seems really reasonable though. a vote ending on the first or the second of the month seems fine as well, since stats take a day or two to calculate.
 
Three month tier shifts is an old policy and was necessary in its time. Pokemon Showdown has way more battle activity than we had back then; one month now probably has more battles than three months combined had back then. So, I'm going to agree that the monthly tier shifts are fine.
 

Quite Quiet

I need a kitchen knife that doesn't whisper to me
is a Site Content Manageris a Member of Senior Staffis a Community Contributoris a Tiering Contributoris a Contributor to Smogonis a Top Smogon Media Contributoris a Top Dedicated Tournament Hostis a Tournament Director Alumnusis a Battle Simulator Moderator Alumnus
TFP Leader
The only example where this is not desirable that I know of the top of my head:
February 2015: Virizion moved to NU
March 2015: Virizion moved to RU

This is really the biggest issue with the current system; it's too easily influenced by temporary metagame trends. The above may not impact RU in the slightest (since Virizion is still legal there), but allowing a Pokemon to drop and rise right away the very next month is anything but stable and cause serious disruption to the tier below. Even if we're getting 50 times the battles, temporary metagame trends are still a thing and will continue to be so, and the usage will reflect that and potentially give us more situations like the above.
 
I feel like this problem could be addressed by making Pokemon that drops or rises "immune" to changing tiers again the month immediately after. This would cover the Gallade and Virizion case, and cases where Pokemon drop to tiers where they aren't viable and need to drop more quickly are rare enough with the new way we tier megas that they don't matter a huge amount. The other issues brought up in this thread are pretty minor, except the suspect test issue which has been addressed already.

For what it's worth I much prefer the current system to only having tier shifts every 3 months, this way helps break up the monotony of waiting through a stale meta and there are almost never cases where the quick drops / rises have too big an effect because the number of actual tier changes is much smaller than it is for the normal tier shifts.
 

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top