On Secondary Ability

[earthflax]

spoo approved
Hi all,

We should make it so that the final Secondary Ability poll ends before the Art poll begins. This can be achieved either through delaying the start of Art Submissions by an as-yet undetermined amount of time relative to the competitive stages, or through setting a hard timeline that the Art poll cannot begin until the Secondary Ability polls close, or both

The main reason that I'd like to see this change is that the competitive aspects of this process should inform the flavor aspects as much as possible, and additionally because the current timeline can make it so that this dynamic is inverted at times. We did not have to approach this issue at all in CAP36 because Ramnarok's framework prohibited a Secondary Ability in the first place, but this inversion of dynamics can best be demonstrated by Shox's process. The first Secondary Ability poll went up right after the last Art poll ended. I'm not one to speculate on the reasons behind the thought processes of voters, nor do I really care how the poll ended on a personal or process level (I actually think that we voted for the best Ability regardless), but I think that you could make the argument that this poll was, to some extent (even if it was very subtle), informed by the flavor of the bullfrog-like design that won the Art poll, with the abilities that made the most sense flavor-wise (Wandering Spirit and Sticky Hold) making the second poll fairly easily. Even if this win was purely coincidental (which I'm inclined to agree with, in this specific case), I think that it can be pretty reasonably concluded that, should the Art poll and Secondary Ability poll overlap, and especially in cases where the former ends before the latter, it can lead to flavor-based voting.

I think that, pretty fundamentally, this is a simple fix to a potentially very bad issue. I think that our philosophy on this front should always be one that assumes the best of all voters while preparing for the worst (or, in other words, even if we don't think that people are voting for the wrong reasons, we should still limit the variables that might influence them to do so). Additionally, with this change, the only other stage that could be informed by the results of the Art Poll would be Movesets, which I don't think is a huge issue considering we have Defining Moves to lock in moves that could otherwise fall victim to flavor-based voting as well. There could be problems with this proposal on the flavor side of things, which I am less familiar with, or from a purely logistical level as far as the timeline for the process and implementation goes, which I'm also not familiar with. The main issue that I can see with this proposal is the issue that got Secondary Ability moved a few years back with the possibility of a determined Secondary Ability forcing artists to change their submissions on very short notice to suit the Ability flavor-wise. I can sympathize with this concern, but considering the philosophy behind CAP, I find it to be much more important for the flavor of a CAP to never influence competitive submissions and votes than vice versa. On top of this, some artists now feel an impetus to suit their design for all of the abilities that are being discussed/slated for Secondary Ability, which is essentially the same issue that we tried solving by moving Secondary Ability in the first place but made even broader. If there are other issues, then do let me know, but I think that, from my perspective, this is a fairly simple fix for an obvious problem.

 

[Zetalz]

I've felt a fair amount of frustration over Secondary Ability for quite some time. It feels like there's no happy-medium placement for it in the process without causing friction with some other aspect of it. Placing it directly after Primary isn't ideal since it's completely uninformed by the stats stage, and essentially just drags the Ability stage out longer than it need be. Placing it after stats and pre-Art, as it was for a long time, is what ended up in the proposal regarding Artist stress that resulted in it's current placement. And while I do agree with flax that competitive aspects should inform flavor aspects as much as possible, I also still understand and empathize with that prior proposal. You could readily see and feel the burnout caused in the artist channels over the stress being caused by Secondary's placement.

I think the op's proposal to essentially revert the previous proposal and move it back where it was Pre-Art will not result in a satisfying solution for anybody. At that point we're just starting up the cycle of kicking a dirty, old, half-deflated ball that nobody likes playing with between flavor and competitive sides of the project once one side gets fed up with it.

But if we're not going to revert the previous change, and continue to exist with this current un-satisfying iteration, then we still need hash out some how to solve the annoying conundrum that is this stage. The rest of this post isn't really structured as a formal proposal or counter-proposal, more a collection of thoughts on the other frustrating aspects about Secondary's problematic existence I've pondered over the years. As a preface; I'm on day 6 of this current migraine, so deepest apologies if this gets kinda rambly or tangent-y.


-The Often Superfluous Nature of Secondary Ability-
In CAP, there is very much a visible culture of preferring to have 'something over nothing', regardless of any confident consensus on the viability of a project. This is noticeable in Movepools, usually manifesting in the form of move bloat, which is a problem we've been slowly addressing over the years to varying success. However, when it comes to Secondary, this mentality is very difficult to escape. I'm not afraid to say I too have fallen victim to it in my time here. The root causes of this mentality aren't hard to deduce. The ever-present fear of a CAP releasing in an underperforming state has haunted many a project we've done in recent memory, and with it comes the overriding desire to give a CAP as many viable tools as possible, to ensure it's success.

However, even when a project DOES have high confidence in a CAP's success, what often ends up happening is this need to ensure viability, transforms into a want for variety. Whether to cover an angle not fully covered in prior stages, or to address a few niche interactions, or just because it's pro-concept; many winning Secondaries have been relegated to being a part of the mon's kit purely for varieties sake.

The stated intention of Secondary to not cause conflict nor outshine Primary Ability means that you will rarely see a truly viable secondary being voted in for a CAP that's already had smooth sailing up until to that point. spoo's absolutely excellent post in Secondary for Obliteryx illustrated this more succinctly than I ever could here. Threading the needle with Secondary is exceptionally difficult. Yet there's a lot less thought and legwork put into the stage in the majority of modern projects that would be necessary to see that through.


An example of the Secondary conundrum that's personally bothered me a lot over the years, is the terrible performance of No Competitive Ability in polling. While it's not often dead-last in most polls, NCA also struggles to win out, REGARDLESS of how smooth a given project has been going. As far as I can recall, there have only been 2 CAPS in the past 10 years, whom were not already restricted in their ability count by their concept/framework (Starters/Book/Ram/Chrom), that won NCA as their Secondary; Pajantom and Astrolotl.

One could point to the high power level of their Primaries, or their general strong builds as the reason why NCA managed to beat out the 'something over nothing' mentality here, but the actual numbers are closer than you'd think. Astro in particular I think illustrates how strong the desire to have something over nothing can be. Regenerator is an exceptionally strong ability, that at the time of Astro's creation, was arguably the single most controversial ability you could bring up in a CAP discussion. And despite this, Poison Touch still nearly managed to worm it's way to victory. On a mon that with hindsight, very clearly would not have used it to any meaningful degree. But even if it was terrible, as long as it wasn't causing any problems, why not try and slap it on? Something over nothing.


On the flipside of this coin, in an a bit of ironic duality, there have also only been 2 CAPs in modern memory who's Secondary processes are brought up as examples of Secondary actually achieving it's stated, intended purpose; Cresceidon and Shox. While I agree completely with the sentiment on Cres' implementation (again, illustrated fantastically in spoo's linked post) , I do very much feel like it is the exception to the rule, as I don't have much in the way of positive to say for the other example in Shox.

Shox's Secondary stage was very clearly an instance of the "Oh shit, we fucked up!" button getting windmill slammed in an attempt to course correct after users realized they tunnel visioned way too hard during Primary. It was not about attempting to strike a balance between viable abilities, but racing to get any viable ability on the mon before it dropped. While one can point to it as an example of Secondary achieving a positive outcome, "saving the mon" in a way, one can also point to it as a superfluous element of the process, given that we have already implemented a stage that is explicitly for the purpose of tuning a CAP after release; Post-Play Lookback. While we've mostly utilized Post-Play as a method to 'release broken, nerf down', it is just as capable of 'saving' a weaker performing CAP, hoisting it up to where it ought to be using hard data gathered from actual playtesting, while also allowing changes to stats and moves. There is nothing more superfluous than having the same positive influence of a stage be achievable almost as easily later down the line, where it is more directly capable and intentional.


All of this old man rambling to say; I really struggle to see the value Secondary Ability brings to the process when compared to the amount of headaches it causes. Honestly, it feels like a part of the process that's here just because it's always been here (more or less), and since it rarely contributes anything of true positive or negative value, it's just been allowed to remain as is. Now becoming the red-headed stepchild of the process, causing fuss with both the competitive and flavor sides of the family. But it doesn't have to be!

So I guess I kinda lied earlier when I said I wasn't going to offer a formal counter-proposal lol, but in the midst of writing this way-too-long post the idea just kinda came to me.


I think there is merit to removing Secondary Ability as a mandatory stage of the process, unless explicitly demanded by the concept/framework, turning it instead into an optional stage that is only discussed upon after finishing Stats, with users needing to provide strong support for it's inclusion. And if in the case of said inclusion, for it to be placed before Art Polls and extending the time between Secondary's wrap-up and the beginning of Art Polling by an amount of time suitable for artists to not pull their hair out.

This, I think anyway, addresses the issue of the superfluousness of Secondary, forcing users to think more closely on the stage, as well as the entirety of the process up to that point, producing a higher quality discussion and outcome than the usual slap-dashery it is now. It also allows artists to be given hard confirmation to not have to worry about it and finalize their designs, or be given a generous heads-up extension to deal with it. The specific amount of time can be discussed by people who are more involved in the flavor process, I think. I don't really feel qualified to throw out numbers here.


I'm certain I didn't mention a few things I wanted too, or provided more examples for context, but I really wanted to get this post out in a timely fashion. The struggles with Secondary are something that's bothered me for a long time here, and I'm very keen on seeing them addressed in some form.

 

[Spammernoob]

Figured I'd make a post here (speaking personally, not as a mod).

I'll reference the original Relocating Secondary Ability thread (https://www.smogon.com/forums/threads/relocating-secondary-ability.3705301/) as I think much of the reasoning there is relevant here.
The original proposal as outlined in the thread was:

Hide

Proposed new pipeline in the situation that primary ability is central:

v
1) Primary ability stage
2) Stats stage
3) Primary ability centrality eval
4) Art poll
5) Secondary ability stage
v


For the last point in this section I wanted to highlight some areas in which Secondary Ability before Art could be seen as necessary:
1) The concept has a high amount of depth, and the Primary Ability touches on a very specific niche inside it- with much more room for exploration. Example: Jumbao
2) The primary ability is mild, and doesn't define the mon. There is clearly a power budget for more abilities here. Example: Cyclohm
Its also worth noting that both of these concepts had 2 abilities in mind before even reaching the secondary stage- Cyclohm's was outlined in the concept, and Jumbao we decided to do a sun and a sand ability prior.

In summary, if we expect a CAP's secondary to see no competitive play, then we relocate Secondary Ability stage to after Art Poll. Notably, this proposal allows for Secondary Poll to take place before Art Poll if we expect the secondary ability to see competitive use. The criteria for when this is the case are outlined in the proposal above.

The problem that this proposal aimed to solve was to avoid (when possible) designs essentially getting invalidated by the choice of Secondary Ability. It conceded that competitive should take precedent over flavor, but in the case where the secondary would see no use anyways, there is no point in making artists' lives harder.

The proposal that ultimately got implemented was to move Art Poll before Secondary Ability Poll in all cases, bypassing the need for Primary Ability Centrality Evaluation. Given that CAP rarely has competitively used Secondary abilities, and such a stage would likely see very little engagement (see: activity during Threats stage), this was definitely not an unreasonable solution, and saw no objections at the time.

What has changed since this proposal? This proposal passed after Saharaja's process, at the end of Gen 8. Since then, we've had 6 CAPs, 3 of which (Cresceidon, Shox, Obliteryx) with competitively used secondary abilities. Realistically, I don't think any of these processes has been negatively impacted by Secondary Ability Poll taking place after Art Poll. However it is not unreasonable to worry that this might become an issue in the future.

That being said, I am against a full revert of the proposal. The impetus for the original change was a concrete issue artists had, and, while similar concerns regarding flavor impacting competitive stages were raised at the time (here), they haven't been borne out in any recent process. I think the tradeoff of solving a hypothetical problem by reintroducing a known one just doesn't make any sense. The OP brings up that artists may need to adjust their design for the range of discussed secondaries, but this seems like a lesser evil compared to designs being invalidated right before polls, by an otherwise inconsequential Secondary Ability choice. I'm not an artist, so I'd need artist input on whether moving Secondary Ability after Art has actually made life easier (Pipotchi).

Additionally, it's not clear to me that this proposal solves the problem as outlined. Even if Secondary Ability Poll occurred first, I think it's impossible to avoid some level of influence from Art; if my favorite design is incompatible with a Secondary Ability, wouldn't I be incentivized to vote against said Secondary Ability? Ultimately, it is a matter of degrees, and I don't think we have enough evidence that the design actually affects Secondary Ability Poll to a degree that warrants this change.

As an aside, I will bring up the logistical side, as that is probably the part I am most familiar with. We currently have multiple flavor stages (Name, # of Prevos, Sprite, and Shiny) that take place after Art Poll, and only one competitive stage (Secondary Ability). Moving Secondary Ability before Art basically extends the process by the duration of the Secondary Ability Poll, but more importantly, increases the time where the competitive side is basically just waiting for the flavor side to finish. This point is relatively trivial, so I left it at the end of the post, but it nudges me towards keeping things as-is.

In conclusion, I think things are basically fine the way they are now. However, I am also fine with the original proposal, where we decide whether Art comes before Secondary Ability based on whether we expect the secondary ability to be used.

EDIT: New post while I was writing this. I don't hate the idea of making Secondary Ability optional. That is, during Primary Ability Centrality Evaluation, we decide whether we have a Secondary Ability Stage (in which case, it occurs before Art Poll), or we decide to skip Secondary Ability Stage and just get 2 flavor abilities afterwards.

 

[shnowshner]

Extremely 3:00-AM-stream-of-consciousness type post below. There's a TL:DR at the bottom.

The first Secondary Ability poll went up right after the last Art poll ended. I'm not one to speculate on the reasons behind the thought processes of voters, nor do I really care how the poll ended on a personal or process level (I actually think that we voted for the best Ability regardless), but I think that you could make the argument that this poll was, to some extent (even if it was very subtle), informed by the flavor of the bullfrog-like design that won the Art poll, with the abilities that made the most sense flavor-wise (Wandering Spirit and Sticky Hold) making the second poll fairly easily. Even if this win was purely coincidental (which I'm inclined to agree with, in this specific case), I think that it can be pretty reasonably concluded that, should the Art poll and Secondary Ability poll overlap, and especially in cases where the former ends before the latter, it can lead to flavor-based voting.

This whole tangent does not sit right with me. You spend half the wordcount backpedaling about how this situation is very likely completely fine and pointing at nothing, and use it as a baseline for the proposal on the basis it "can lead to flavor-based voting" regardless? It just isn't. Wandering Spirit has nothing to do with Shox, at all. Immunity, the second highest placement, does not either outside of the Normal-type being shared with Snorlax and Zangoose. Sticky Hold sort-of makes the most sense with frogs typically seen as "slimy," or illustrating the triboelectric effect of static electricity, but its still a stretch if you want to compare it to all other Sticky Hold users.

Look, the ordering of the top voting list (Sticky Hold - Immunity - Wandering Spirit - Seed Sower - Heatproof) definitely lets you draw conclusions about flavor-based voting. Sticky Hold and Immunity are more broad and rank above the other three, but at the same time, Wandering Spirit and Seed Sower make far less sense for Shox's design than Heatproof does, and excluding NCA, that one placed dead last!

Here's the thing about flavor-based voting: it's inescapable. We can't control how people vote, and anyone who can post can vote. People are voting for reasons beyond our goal of competitive-first focus the moment we have open the Concept Poll of a CAP.

Additionally, it's not clear to me that this proposal solves the problem as outlined. Even if Secondary Ability Poll occurred first, I think it's impossible to avoid some level of influence from Art; if my favorite design is incompatible with a Secondary Ability, wouldn't I be incentivized to vote against said Secondary Ability? Ultimately, it is a matter of degrees, and I don't think we have enough evidence that the design actually affects Secondary Ability Poll to a degree that warrants this change.

This is a very good take and highlights the difficulty of trying to control the potential impact flavor-minded decisions can have. There are many factors that could influence how someone decided to vote, and we have no surefire way of knowing what they are for any given individual. It could be competitive, it could be flavor, it could be outright trolling, maybe someone wasn't even sober and threw some letters together. Just as someone might vote for Sticky Hold because it makes sense for Shox, they may also vote against it because it doesn't align aesthetically with the Pokemon that currently have Sticky Hold. This is still voting for flavor, but in a way reverse to how its been presented. That's what we're up against.

Let's talk typing for a sec:

• We're 39 CAPs deep at this point and not one has retread on a typing used by a previous CAP.
  • To elaborate, I looked through the first Typing Poll of every CAP between SM and now, and the amount of times a Typing on said poll was used by a prior CAP was 4: Steel/Flying for Obliteryx (last place), Fire/Grass for Jumbao (last place), Steel/Flying again for Equilibra (second to last), and Flying/Steel yet again for Chromera (second to last).
• Of these 39 CAPs, 17 of them have a typing that was unused by any official Pokemon at time of release. (Syclant, Revenankh, Pyroak, Kitsunoh, Cyclohm, Colossoil, Voodoom, Tomohawk, Necturna, Mollux, Aurumoth, Volkraken [2014, Volcanion was released 2015], Plasmanta, Crucibelle, Kerfluffle, Miasmaw, Hemogoblin)

What does this tell me? People vote for unused type combos frequently, and avoid previous CAP typings as much as possible. How confident can we be that the voterbase is picking the most competitively-relevant typing when the data is skewed so heavily for "unique" typings?

Even this argument I don't hold too much value for. From a competitive standpoint, a unique typing offers a unique niche in a format, and given CAPs tend to be somewhat relevant in their own metagame, overlapping your typing with one could put us at risk with competing for space. Furthermore, this does not account for all the instances where proposed unique typings failed to succeed, or the fact that in many cases these typings had strong merit for the CAP in question, hence why they were allowed to be on the poll at all!

That's the thing about CAP: we don't want people to vote for flavor, and thus we've implemented measures to minimize it as much as possible. It's a major part of what the TLT is there for. Sticky Hold doing better than Seed Sower in the Secondary Poll because of the winning design for Shox is a significantly less impactful reason compared to Sticky Hold being a generally good, straightforward, and practical ability for the sort of competitive role Shox was seeking to fulfill, compared to something more awkward, variable and potentially harmful like Seed Sower is. I think it's unfair to assume, even half-heartedly, that Sticky Hold did well because people felt it worked with the design more than how it worked with our toolkit.

Sorry if this comes off as overly bitter, but it's hard to read that paragraph and not feel like you're trying to draw conclusions off of information you outright recognize at inadequate. Please do not do this. I would know from experience.

I think that our philosophy on this front should always be one that assumes the best of all voters while preparing for the worst (or, in other words, even if we don't think that people are voting for the wrong reasons, we should still limit the variables that might influence them to do so). Additionally, with this change, the only other stage that could be informed by the results of the Art Poll would be Movesets, which I don't think is a huge issue considering we have Defining Moves to lock in moves that could otherwise fall victim to flavor-based voting as well.

The whole of your argument could have really just been this block right here with zero mention of how things went down with the Shox Poll. Even if the likelyhood of a finalized design influencing voters is mostly inconsequential, ideally we limit how much reach flavor has on competitive stages as much as possible so that the CAP process is as good as it can possibly be. Pushing Art Polls to start after Secondary Ability, when applicable, is a step towards further optimizing the CAP Process ideal. Spammer brought up a good point with regards to timing, the flavor stages can already drag a process out quite heavily, and this furthers that annoyance. I don't find needing to wait longer is too concerning, even if it does suck.

Personally, I've always held onto the belief that CAP would be better off if every flavor-related stage only got going the moment we finished up Movesets: get the whole functionality skeleton out the gate and let the artwork side of things come afterwards. I have been told that this isn't a good way of doing things due to both the length of such a process, and Showdown-related implementation hurdles: were this not a concern, I'm certain we'd have gone down this route a long time ago. So any step that gets us closer to this ideal is something I am broadly in support of.


This is where we arrive at Zeta's proposal of just axing Secondary if there's not enough qualified intent to pursue one. What exactly are we using this stage for? Primary Ability already gives us direction and strength to build around, so we're not in need of that (at least, we shouldn't be). Is it to further our mission to explore the design of competitive Pokemon and a chosen concept? Our results in this case are rather mixed imo, some CAPs can certainly choose between two Abilities, but most stick to just one. And if its an emergency button to reroute a CAP people are losing faith in, we already have PPL to handle such things, or upcoming stages to use as well. I understand the sentiment completely.

Now, as far as the OP is concerned, this isn't solving anything. Secondary being optional only affects the potential for overlap with Art Polls to encourage flavor-based voting when there's no poll to be had. So we've kinda solved the problem, except when we do want to explore a Secondary. Now what? When do we hold this stage? If it still occurs after Art Poll, we're losing optimization again. And if we don't have the stage at all in favor of addressing it during PPL, well that's after Art too.

_______________________________________

Here's my final thoughts on the matter, based on what I said above and in response to what others have mentioned:

I don't think this is going to bring meaningful change to how polls proceed. There's too many variables attached to what makes a person vote a certain way. CAP already has a lot in place as-is to encourage users to think about the competitive side when voting for such stages.
I would agree that this is the correct path from an optics view. You said it best, "even if we don't think that people are voting for the wrong reasons, we should still limit the variables that might influence them to do so." This should be the modus operandi of CAP.
We would be reverting a change that was made to avoid other problems people had. I am of the opinion that flavor-related aspects of CAP should always come second to competitive ones, and your artwork doing worse because of an ability choice is completely irrelevant to how the CAP will perform. It just sucks for you and you'll need to live with that. Ultimately, however, the flavor side of the CAP Process is as large as the Competitive side at minimum, and such a mindset comes off as particularly elitist and cold. We're all doing this for fun, after all, and what's certainly not fun is feeling your design you spent weeks on got owned because we decided last minute the Bug/Fighting type should also have Water Veil.
Secondary is often not a necessary stage and we could partially solve the problem OP confronts by not having it as much. There's more to be done in terms of the why and when we should have a Secondary Ability stage at all, but that's not something I'm looking to answer right now. Strictly speaking, if there's no Secondary Poll to overlap with Art Poll, there's no worry of flavor impacting our Ability choice.