(I don't know if this post belongs here, mods can move it somewhere else if they want but I didn't see a better place for it. This type of thread used to go in IS but IS is dead as fuck, I don't have posting perms in there, and I would want non-badged users to see this anyway.)
When I think of improvement resources in other games, I think of books, video calls with top players, and profitable YouTube channels--in short, money. But there has never been any money in this place, so what do we get out of supporting new players? The only reason I can really think of is that we want to expand the community and gain more perspectives both inside and outside the game. Essentially, this is competitive Pokemon working as a friend-generating machine, with the new players then going on to make their own resources, supporting both older players and even newer ones. Thus, the quality of play, resources, and discourse improves exponentially over time......right?
Analyses and spotlights have a horribly restrictive structure, work within a modular framework unhealthy to both writer and reader, acknowledge "the meta" yet ignore the actual people playing, and are so brittle they have to be rewritten every couple months. With this in mind, it's no wonder that historically most analysis writing has been done by a revolving door of teenagers writing boilerplate in an effort to produce work they can be proud of.
What I mean by "modular framework" is that there are only so many interesting or useful things you can say about a single Pokemon without considering the full team surrounding it (the "Team Options" section, providing little information as to how the full building process should actually work, might as well be called "Good Buddies"). Once you're familiar with what each mon does, even the best-written analysis will not be of much help.
The use of LLMs in analyses has been strictly prohibited, as every Smogon Brand(tm) Content must contain high-level, human-generated, and accurate information about how Heatran loses to Water-types, but the funny thing is that the type of writing we're asking for isn't actually too far off the sorts of things LLMs can produce. The only thing they're really missing is meta knowledge and airtight game mechanic information, and considering how that could logistically work is a fun thought experiment. That said, if such a model could be created, I'm still pretty sure that we could find more helpful things for it to do than write analyses.
I won't say that we should get rid of analyses or anything. They are legitimately helpful for those who are completely new, and the downsizing which has happened lately isn't too far off from what I would prescribe. Those currently in C&C will likely have more relevant and helpful perspectives than me as well.
I certainly don't believe that improvement resources in a competitive game are directly comparable to progression resources in an MMORPG, but you should stop here for a minute and ask yourself how exactly you think those two things differ. I'm serious. Do that right now.
-
Now, to me the essential difference seems as follows: as you progress in an MMORPG you gain new ways of playing, but as you improve in a competitive game you gain new ways of thinking, which can then facilitate new ways of playing. People are drawn to competitive games because inventing entirely new ways of thinking is fun. As players improve, they see the game differently, build and play using rigorous and personal processes alien to even other strong players, and can even identify and change their own outlooks when they fail. Even if meta knowledge itself is both ephemeral and useless outside the game at hand, the process of mastery can be profoundly transformative to one's self.
Getting back on topic, this means that the role of improvement resources is to help players think for themselves in many new ways. How can we do that?
I think samples should be more like RMTs. If they're meant to help new players learn the tier, they should include information regarding how and why they were built, and maybe also have some demonstrative replays. This shouldn't be too much of an ask, as most of the effort that goes into RMTs relates to the mons themselves (this is actually a lie, 90% of the effort is finding songs and images for all the mons). This change would also mean that sample submissions could be taken more seriously as contributions to the site.
Game analysis should be at the core of our resources, not the periphery. I just opened up the OU metagame discussion thread and saw this awesome post + replies. To me, the fact that analyses of single Pokemon are placed front-and-center, yet analyses of games and of the broader meta are left to rot away in threads like this, feels completely backward. An outdated Pokemon analysis is of little use, but an outdated game analysis still holds water; even if the specifics have become irrelevant, the underlying thought process is what counts.
Another great thing about game analysis is that it allows for appreciation of the playerbase. If you're looking, you can spot moments in almost every game where the players made measured maneuvers, hit reads at just the right time, or acted on almost imperceptible long-term plans. This is also incredibly good for social media posts!! If you can explain why xavgb sacrificing his Garganacl was a genius play in language the average player can understand, this puts you on par with BKC, who gets thousands of views on unedited but nonetheless engaging videos. This kind of thing gets people thinking.
Every Pokemon analysis should include one or more high-level replays of the Pokemon in question doing something (even if no commentary is included). Metagame thread OPs should link to high-quality posts. If I'm getting into a new tier, I want to be able to find every game analysis anyone has ever done for it. Fuck it, bring back the warstory forum.
One last thing that I wanted to point out: I feel like we enforce a sort of faceless, voiceless "Smogon image" on official, outward-facing resources, which I think has had devastating consequences for said resources. This is not a solved game, and nothing interesting at all can be said if you do not embrace the subjectivity and eccentricity of your playerbase. The problem is that no one can advocate for sets not unanimously agreed to be viable without being policed and having their skill level or commitment to making "good content" called into question. Some of the most bland, noncommittal, conformist writing I've ever seen has come out of a website for young people without any financial incentive. Make it make sense!
To start off, we need to take a closer look at metagames that are very distinctly not Smogon. I remember having a surprisingly illuminating experience playing Shitmons in 2021. This is not a balanced metagame--it doesn't take a genius to see that Beldum is practically uncontested--but playing a metagame with such flagrant disregard for balance was not something I'd ever done before and I think everyone should at least try it. If we understand these extreme metagames (typically a much easier ask than understanding standard ones), we can much more easily make statements on what we want from this game and why we want it.
I only played a couple games of Shitmons, so I would say my understanding of all this is pretty low and I don't really have any answers. I do, however, have a few questions. Answer My Questions. Play Fishmons and MIA BH (shoutout Nihilslave) for ten years.
You're free to disagree with any or all of the post and write it off. My arguments are pretty imperfect and I'm not good at appealing to large crowds of people. But if you take anything from this post, take this: science and philosophy and mathematics must not be things which are locked up in some sterile ivory tower. You would do well to practice them regularly in every space which is important to you, even if you aren't formally trained in them. And it's healthy to ask yourself what you are really doing here.
Thank you for reading.
Expansion of the Community
When I was reading that sample teams thread, something that stuck out to me was this unspoken belief that we, the Content Creators of smogon.com, should do our best to help newer players get better at the game. Don't get me wrong, I agree with this, but we should be asking ourselves why we believe this.When I think of improvement resources in other games, I think of books, video calls with top players, and profitable YouTube channels--in short, money. But there has never been any money in this place, so what do we get out of supporting new players? The only reason I can really think of is that we want to expand the community and gain more perspectives both inside and outside the game. Essentially, this is competitive Pokemon working as a friend-generating machine, with the new players then going on to make their own resources, supporting both older players and even newer ones. Thus, the quality of play, resources, and discourse improves exponentially over time......right?
C&C
As someone who's worked in C&C in the past, I want to start off by saying that I truly have no ill will for anyone involved in the C&C process, from writers to QC to GP to mods. All of them are working hard and I respect that.Analyses and spotlights have a horribly restrictive structure, work within a modular framework unhealthy to both writer and reader, acknowledge "the meta" yet ignore the actual people playing, and are so brittle they have to be rewritten every couple months. With this in mind, it's no wonder that historically most analysis writing has been done by a revolving door of teenagers writing boilerplate in an effort to produce work they can be proud of.
What I mean by "modular framework" is that there are only so many interesting or useful things you can say about a single Pokemon without considering the full team surrounding it (the "Team Options" section, providing little information as to how the full building process should actually work, might as well be called "Good Buddies"). Once you're familiar with what each mon does, even the best-written analysis will not be of much help.
The use of LLMs in analyses has been strictly prohibited, as every Smogon Brand(tm) Content must contain high-level, human-generated, and accurate information about how Heatran loses to Water-types, but the funny thing is that the type of writing we're asking for isn't actually too far off the sorts of things LLMs can produce. The only thing they're really missing is meta knowledge and airtight game mechanic information, and considering how that could logistically work is a fun thought experiment. That said, if such a model could be created, I'm still pretty sure that we could find more helpful things for it to do than write analyses.
I won't say that we should get rid of analyses or anything. They are legitimately helpful for those who are completely new, and the downsizing which has happened lately isn't too far off from what I would prescribe. Those currently in C&C will likely have more relevant and helpful perspectives than me as well.
Why play competitive Pokemon?
I assume some of you, in response to the earlier question of "Why should we provide high-quality resources", thought something along the lines of "this helps new players, and I like to help people". To this, I offer the following anecdote taken from a post about teaching Pokemon that I wrote a couple years ago:I certainly don't believe that improvement resources in a competitive game are directly comparable to progression resources in an MMORPG, but you should stop here for a minute and ask yourself how exactly you think those two things differ. I'm serious. Do that right now.
-
Now, to me the essential difference seems as follows: as you progress in an MMORPG you gain new ways of playing, but as you improve in a competitive game you gain new ways of thinking, which can then facilitate new ways of playing. People are drawn to competitive games because inventing entirely new ways of thinking is fun. As players improve, they see the game differently, build and play using rigorous and personal processes alien to even other strong players, and can even identify and change their own outlooks when they fail. Even if meta knowledge itself is both ephemeral and useless outside the game at hand, the process of mastery can be profoundly transformative to one's self.
Getting back on topic, this means that the role of improvement resources is to help players think for themselves in many new ways. How can we do that?
Sample Teams
I do think the way samples are structured currently doesn't encourage players to think for themselves, so I really do understand where peng was coming from in their OP. However, making the samples less viable isn't a great solution to this. I won't dwell on this case as others in the thread made many good counterarguments.I think samples should be more like RMTs. If they're meant to help new players learn the tier, they should include information regarding how and why they were built, and maybe also have some demonstrative replays. This shouldn't be too much of an ask, as most of the effort that goes into RMTs relates to the mons themselves (this is actually a lie, 90% of the effort is finding songs and images for all the mons). This change would also mean that sample submissions could be taken more seriously as contributions to the site.
Game Analysis
Recently I've been making game analysis posts (example) and while there's a lot of effort involved, this has felt like some of the highest-quality competitive writing I've ever done. Going back to the theme of helping players think in many ways, when you're really getting into the guts of playing Pokemon you'll introduce a highly personal decision-making process which other players can understand and in some cases disagree with.Game analysis should be at the core of our resources, not the periphery. I just opened up the OU metagame discussion thread and saw this awesome post + replies. To me, the fact that analyses of single Pokemon are placed front-and-center, yet analyses of games and of the broader meta are left to rot away in threads like this, feels completely backward. An outdated Pokemon analysis is of little use, but an outdated game analysis still holds water; even if the specifics have become irrelevant, the underlying thought process is what counts.
Another great thing about game analysis is that it allows for appreciation of the playerbase. If you're looking, you can spot moments in almost every game where the players made measured maneuvers, hit reads at just the right time, or acted on almost imperceptible long-term plans. This is also incredibly good for social media posts!! If you can explain why xavgb sacrificing his Garganacl was a genius play in language the average player can understand, this puts you on par with BKC, who gets thousands of views on unedited but nonetheless engaging videos. This kind of thing gets people thinking.
Every Pokemon analysis should include one or more high-level replays of the Pokemon in question doing something (even if no commentary is included). Metagame thread OPs should link to high-quality posts. If I'm getting into a new tier, I want to be able to find every game analysis anyone has ever done for it. Fuck it, bring back the warstory forum.
One last thing that I wanted to point out: I feel like we enforce a sort of faceless, voiceless "Smogon image" on official, outward-facing resources, which I think has had devastating consequences for said resources. This is not a solved game, and nothing interesting at all can be said if you do not embrace the subjectivity and eccentricity of your playerbase. The problem is that no one can advocate for sets not unanimously agreed to be viable without being policed and having their skill level or commitment to making "good content" called into question. Some of the most bland, noncommittal, conformist writing I've ever seen has come out of a website for young people without any financial incentive. Make it make sense!
Tiering
In case you didn't notice, we have a new Tiering Policy Framework, which is mostly the same as the old one. My main issue with the Tiering Policy Framework is that it basically says "This is how Smogon works" without really giving a rigorous reason for why we do things the way we do. Some luck is okay, but too much is bad? Some team matchup is okay, but too much is bad? How has this masqueraded as some sort of bastion of objectivity? All we're doing here is giving players ammunition to talk past each other in suspect threads: "I think this element causes too much luck and team matchup!" "I disagree, I think this amount of luck and team matchup is perfectly fine!" Neither side really ends up reaching the other.To start off, we need to take a closer look at metagames that are very distinctly not Smogon. I remember having a surprisingly illuminating experience playing Shitmons in 2021. This is not a balanced metagame--it doesn't take a genius to see that Beldum is practically uncontested--but playing a metagame with such flagrant disregard for balance was not something I'd ever done before and I think everyone should at least try it. If we understand these extreme metagames (typically a much easier ask than understanding standard ones), we can much more easily make statements on what we want from this game and why we want it.
I only played a couple games of Shitmons, so I would say my understanding of all this is pretty low and I don't really have any answers. I do, however, have a few questions. Answer My Questions. Play Fishmons and MIA BH (shoutout Nihilslave) for ten years.
- A game is considered "solved" when you can consistently maximize your odds of winning with either an unchanging pure strategy (always bring X team) or unchanging mixed strategy (40% chance to bring X team, 60% chance to bring Y team). Is it desirable for a metagame to be solved? Why or why not?
- Are more unbalanced or uncompetitive metagames generally closer to being solved? Is this always the case, or are there cases where a very unbalanced metagame is unsolvable?
- Consider the metagame NU + Arceus. Due to Arceus's extreme versatility, it's difficult to come up with a specific winning strategy. Can a more vague strategy such as "Limit opposing Arceus with X and Y measures, and build around any good Arceus set" be considered a solution? If this strategy is generally sound, does this limit the amount of ways in which you can think about the game? Does "use viable Pokemon, and knock out all 6 opposing mons" limit the amount of ways in which you can think about the game?
- Monotype is a very unique metagame. Sorry if I'm wrong on this, but I've gotten the impression that Monotype teams are generally easier to build than non-Monotype ones due to the matchups you need to cover. However, this doesn't tell the whole story: your selections will naturally be limited when you're pulling from only 1/18 of the pool. If every type had as many viable options are there are viable OU Pokemon, would Monotype be easier or harder to solve than OU? Could such a metagame even be stable, or would the vast majority of options quickly become unviable?
- If you use the mixed strategy of taking one of the 17 sample teams at random, is BW OU a solved metagame?
- Are the people playing whichever hax-removal mod is currently in fashion playing an objectively better game than we are? Ok, I hear you, you say those things can't be objective. Are they playing a subjectively better game than we are?
- Let's introduce another word, "complexity", and define it as "the amount of computation it takes to solve a game". So if a metagame is easily solved, it isn't complex, and vice versa. If complexity is desirable, is more complexity more desirable? Why or why not? Is there any limit? If so, why is it there?
- Let's introduce a completely hypothetical character, Sally. Sally has won three different team tournaments which awarded custom avatars in years other than those in which she won. Sally is calm and rational. Should Sally be awarded a custom avatar?
- Can an in-battle position be more complex for one player than for the other?
- Can a battle become more complex over time?
Conclusion
In some respect I think the friend-generating machine is the only pure part of the whole community process and this website and game are but polluted machines that get in the way. Such machines will always be at least a little bit polluted (this website maybe more than a little bit), but what I aimed to do here was reduce the pollution--empty signifiers, unmaintained mechanisms beneath the surface, and worst of all, people not questioning any of it--so that the machine could run more efficiently and waste less people's time.You're free to disagree with any or all of the post and write it off. My arguments are pretty imperfect and I'm not good at appealing to large crowds of people. But if you take anything from this post, take this: science and philosophy and mathematics must not be things which are locked up in some sterile ivory tower. You would do well to practice them regularly in every space which is important to you, even if you aren't formally trained in them. And it's healthy to ask yourself what you are really doing here.
Thank you for reading.
