Some ramblings on Terastallize, where the hat you wears matters.
I think a lot of the discourse on Tera has been focused on the wrong things so far — debates on its power level or whether it lets mons blow past conventional checks or whether it contributes to making stuff like Roaring Moon busted or what-have-you. All of these are important things to discuss, for sure, and it's possible Tera is overbearing on one or more of these points, but I think they're all missing the forest for the trees. All of these things are tangible manifestations of the subtler real problem.
To me,
the fundamental concern with Terastallize is its effects on identifying and progressing towards win conditions.
Anyone who's played mons at a passable level, or at least follows players who do (e.g. keeps up with tournament replays or whatever), knows that winning at Pokemon is rarely about making better turn-by-turn predictions or getting the upper hand in individual exchanges
[1]. Rather, it's about being able to
identify win conditions and how to make progress towards them. This is a big concept to get the hang of since it's core to analyzing gamestates, which is the main mechanism by which skill differences manifest at a high level
[2]. When Heavy-Duty Boots were first released, a lot of players identified that they'd make Rocks-vulnerable mons like Volcarona and passive walls like Blissey a lot stronger. This is true, but their effect on the gamestate — one of essentially halting an entire mechanism of making progress (hazards) — ended up being a lot more influential on the metagame overall
[3]. This took a bit longer to identify since it was a bit subtler, but when people started arguing to ban Boots (something which had very serious support for a while despite not going through), it was not because it made Cinderace broken or pushed WishPort over the line. Hell, Boots isn't the only thing that this argument has been applied to. Everything from Aegislash to Ditto has had people arguing for a ban on these grounds: "it's not broken but it warps and reduces the skill requirement of every game it participates in"
[4].
Of course, Boots didn't end up banned, and I think a major reason for that is that its effects on progress became accounted for. Hazardstack fell out of favour, with most teams that used Spikes also running a gratuitious amount of Knock Off. Sticky Web went from "fringe archetype that could be okay in another meta" to something that wasn't even good enough to be a meme
[5]. And just in general, we saw a lot more teams focus on other ways to make progress as the generation progressed. Knock Off has lost distribution this generation so it's unclear whether this one will follow a similar trajectory, but in any case, I think players were able to "adapt" to how Boots affected their mechanism of making progress
[6].
But Terastallize is not Boots. Boots had a clear and consistent effect on how games are played and won. They reduced the amount pivoting could be punished, and meant that hazard stack wasn't unambiguously a way to make progress towards a "clean" unless supported by Knock Off or other ways to punish bulky pivot cores
[7]. Tera does not come with this luxury.
Tera has negative effects on the gamestate, but these negative effects are fundamentally
unpredictable [8]. You have no clue if your opponent will be able to Tera into a type that stops your sweep. You have no clue if your opponent's sweeper will pull out a surprise Tera Fairy that makes it impossible for you to revenge, or a Tera Electric that blows past the mon you thought would "counter" it. Indeed, I would argue that
we cannot analyze Generation 9 in the same way we analyzed previous generations for as long as Tera exists, since it is currently impossible to consistently recognize win conditions no matter how good your metagame knowledge is.
Let me take an example
[9]. You identify that your opponent has a Dragapult on their team, and you recognize that DD Dragapult is potentially problematic to your team structure. A half-decent player would now know to prioritize keeping their checks healthy — mons that can switch in and prevent it from setting up, or mons that can Sucker it to revenge it, or whatever. Similarly, your opponent knows that their job is to WEAKEN those checks. This establishes the fundamental framework the game is played around: your opponent wants to set up their win condition and you want to stop them
[10]. But with Terastallize, this sort of analysis is
impossible. I am not exaggerating; this one mechanic single-handedly makes this sort of long-term planning not something that can be done (at least, not with what I currently understand about the game — but from talking to strong players, I get the sense that they share my frustration here, even if they don't formalize it in these terms). Here are some examples of ways Terastallize complicates the above win condition analysis:
- Dragapult's Tera type is unknown, so it's very unclear what actually walls it. Sure, most Dragapults are Tera Ghost right now, but there's nothing really forcing that.
- Dragapult's Tera type is unknown, so it's very unclear what actually revenges it.
- Dragapult's teammates also have unknown Tera types, so they could lure in the counterplay to Dragapult you spent so long trying to keep safe and healthy and disposing of it.
- The Pokemon Dragapult should be able to "sweep" have unknown Tera types, so you could see a random Tera Dark or Tera Normal or Tera Fairy totally halt Dragapult's sweep, or at least force some uncomfortable 50/50s. This point is particularly problematic to me, since it means that — even if the win condition is correctly analyzed and flawlessly set up — its success could potentially come down to whether you predict the Tera type correctly and choose the correct move/switch/whether to Tera or not/etc. Even as we get more "used to" Tera and get a sense for what Tera types are common on what mons, this will be a perpetual problem to a far greater extent than Aegislash's 50/50s ever were [11].
The observant among you might argue that all of these boxes are ticked by something already. Sure, it's unclear whether you can revenge Pult with Sucker Punch, but that's always been the case — DD Pult has always been able to run Substitute! Sure, it's unclear what walls what, but plenty of mons get weird coverage options or niche sets or whatever — not to mention Z-moves weren't banned when those existed, and they were arguably more problematic in terms of what can be lured and what can be broken
[12][13]. And besides, isn't part of the skill of Pokemon adapting to complications in the gameplan anyway? Aren't the most exciting games the ones where a player has a "key" wall lured out and disposed of early, but goes on to find some miracle comeback?
The problem is that Tera does this all at once with effectively no opportunity cost
[14]. Skilled players can account for individual complications and disruptions, and in fact, the existence of such disruptions is a great opportunity for creativity and skill expression
[15][16]. But Tera makes
every Pokemon a potential source of win condition disruption from
multiple angles. I do not know
any of the types of my opponent's Pokemon until they click Tera. I do not know what they wall, what they kill, even what status I can use against them. I do not know whether I can Spin on them. I don't know
anything, I can't plan around
anything.
It's like every OU game is a game of OU Blitz where you just try to make the best decisions on a turn-to-turn basis because there's no real room to plan out long-term — except Blitz's time constraints are something a skilled player can overcome by getting good at thinking fast and focusing on what's important. So my question, then, is:
How do we construct a long-term gameplan focused around making progress towards win conditions in the "Era of Tera"?
I'm still not 100% certain I want action taken on Tera, since I think it's possible that this question has an answer... or even
perhaps that I'm asking the wrong question. Perhaps thinking of generation 9 as a traditional generation, with traditional development towards traditional win conditions, is a mistake. We've already basically thrown out the idea of "countering" Pokemon this gen because nothing "counters" anything with full consistency — instead, it's all about hedging the most likely Pokemon to wall you, to stop you, to sweep you
[17].
I use the jargon "Era of Tera" because I want a term for this hypothetical new framework of analyzing games of Pokemon. We can't think about counters, and we can't really think about consistent win conditions
[18]. So perhaps I'm just thinking about games wrong. Perhaps there's some post-Tera way of analyzing gamestates
[19] that requires us to toss out a lot of "traditional knowledge"
[20]. This is a solid argument for being hesitant on Tera, for waiting for players to figure it out. Maybe we should give it more time, healthy development of the metagame (and the ability to make sensible bans) be damned.
But SPL is coming up soonish and I have genuine concerns that Tera continuing to warp the tier will negatively disrupt metagame development — for now it's "too soon to ban Tera", in 6 months it'll be "Tera is too entrenched in the metagame for us to ban it", so what's the point at which we can conclude that, in lieu of finding another way to "play" Pokemon in this "Era of Tera", that Terastallize just removes a lot of opportunities for competitive skill expression from the metagame?
I don't know.
But at the moment, I think Tera is unhealthy. It deemphasizes long-term planning, it reduces the skill cap of the tier, and it makes games too focused around getting key predicts on key turns
[21]. I'm not optimistic towards our chances of finding a solution to this that holds this tier up to a similar standard of competition than it would be without Terastallize.
Apologies for how long and disjointed this is, I hope I got the point across. This mechanic is a problem, but maybe it's one that'll fix itself like Boots kind of did
[22]. It has a lot going against it, though.
——————
Footnotes
[1]: which is why the concept of "momentum" is so nebulous and has kind of fallen out of favour as a way to analyze games since the era of U-Turn-spamming Scizors in pre-Fairygens
[2]: Indeed, a lot of "ladder players" who aren't good enough to make it into Smogon tournies — myself included — struggle because of precisely this point. They have very good gamesense and metagame understanding and are often very skilled at predictions and at threat identification, but struggle to put all the pieces together into a conclusive gameplan. They have, by and large, gotten the hang of making "safe/hedge/midground" plays, but not quite at the overall gamestate. This is what distinguishes them from truly "good" players, to me (and yes, by this definition, I am not a "good" player).
[3]: Or in the case of Regenerator abusers like Toxapex and (to a lesser extent) Torn-T, they bolstered an existing form of reversing progress, i.e. healing off damage taken.
[4]: To be clear, the people who argued against Ditto in this way were a very, very small minority, but they did exist briefly near the end of gen 7 — and they were, by and large, "high-level" players. It wasn't the ladder heroes arguing for this, it was a small subset of SPL players who regularly posted in Policy Review and who understood the game on a high level.
[5]: Sticky Web has been playable in Ubers in the past.
[6]: Trust me, I despise "just adapt" as well as anyone, but there are cases where it legitimately applies — especially when we're talking about a characteristic of an entire metagame (whether it be an item or a move or a mechanic) rather than an individual threat or two.
[7]: I would even argue that "cleaners" as we traditionally understand them were basically killed by Boots. I mean, sure, mons like Kartana demonstrate similar traits and patterns, but how many generation 8 games are actually won by a Scarf Pokemon spamming resisted moves against weakened threats, a la the "cleaners" of old like ScarfChomp or ScarfKeld? A lot more gen 8 games ended with scrappy endgames involving Tornadus or whatever. LO 3/4 Attack Pokemon were almost extinct by the end of the generation, and priority cleaners were practically nonexistent.
[8]: For the purpose of this post, I will not be discussing proposals like "force Pokemon to reveal their Tera types at team preview". Maybe this would work, but it unsettles me from a policy perspective — and in any case, before discussing "solutions" like that, I'd at least want us to come to a general agreement of whether Tera is a problem or not.
[9]: I stress that DD Dragapult is just an example, and any win condition could fill in this place. In fact, part of the skill difference between "good" and "great" players is being able to identify less "obvious" win conditions, like "my defensive core can't be broken except by x threat, so my win condition is to cripple x threat — but y pivot lets it come in for free, so my opponent's win condition will be to get in y pivot a lot in order to get x threat in a lot and get a free kill every time". You don't need a mon explicitly marked "wincon" to have a win condition. But Terastallize is a problem for
every form of win condition, not just sweepers.
[10]: Of course, this is symmetrical — you probably have a win condition of your own you want to set up — but for the sake of this example, I'll take a one-sided view.
[11]: To be clear, Aegislash's 50/50s were not the problem. They are not why Aegislash was banned. Hopefully the fact that it was unbanned and healthy in generation 8 should demonstrate this, but I still see so much historical revisionism on this point that it's kind of ridiculous.
[12]: ...though there's an argument that Z-moves
should have been banned, that isn't super relevant either way here.
[13]: Unlike Z-moves, though, Tera has Tera Blast, so it doesn't require a matching-type move to lure things and blow past them anyway. Not to mention that Z-moves were one-time activation and otherwise were a waste of a move and item slot, whereas a successful Tera continues to pay dividends throughout the entire game. This direct comparison is potentially misleading, though, so I suggest you not read too much into it.
[14]: I barely count Tera Blast as an opportunity cost here, since if you spend your Tera on it, it just becomes a good STAB. But yes, if you don't Tera, then Tera Blast is an opportunity cost of sorts.
[15]: There's a reason a lot of the most memorable tournament moments are things like an unexpected Z move lure or some crazy Scarf Final Gambit play (if you know the game I'm talking about, you know).
[16]: ...and for what it's worth, I think generation 9 is a very... uncreative? metagame, even in light of Tera gimmicks. Perhaps that's the wrong word — this is a very nebulous concept, after all — but a lot of the new designs feel obviously minmaxed towards a single goal. Esparath is literally CM + Speed Boost + STAB Stored Power, a lot of the Paradox mons are minmaxed offensive stats + near-unresisted STABs, you get the idea. The sets kind of build themselves. It's frustrating — I don't expect to see too many cheeky innovations like Vincune or SubPunch TTar emerging when so many of these Pokemon sets basically build themselves. This is a side point, though, and one I don't think council tiering action can realistically fix.
[17]: and I'm not opposed to this hedging on principle. In fact, competitive Pokemon is arguably a game about balancing probabilities and hedging the odds. It's always possible to lose due to a badly-timed miss or freeze, after all — predicting what checks what in the Era of Tera, while not
explicitly RNG, is arguably its own sort of "hedging" and "randomness". Adding this layer of unpredictability isn't a priori unhealthy.
[18]: Again, win conditions have never been 100% consistent — it's Pokemon, after all, and that ambiguity is arguably a good thing for the game's skill cap — but Tera, at least from what I've seen, brings down the consistency to unprecedentedly low levels to the point where planning around a "win condition" goes wrong more often than it goes right.
[19]: I would be moderately concerned if we had to totally reevaluate how competitive Pokemon works every single generation with each new passing gimmick, but oh well.
[20]: and yes, I recognize I'm being hyperbolic here.
[21]: the fact that you can only Tera once is, in my opinion, a flaw rather than a feature — it means that "predicting correctly" with Tera is way more important than it has any right to be. Again, I hate banning things based off of "forcing prediction" since I think a lot of less experienced players conflate their own inabilities to analyze gamestates and pick good mid-grounds with "losing 50/50s", but if any mechanic forces a ridiculously high burden on prediction, it's Tera. (Perhaps we will see Protect popularized as a way to exploit this, though — who knows.)
[22]: Hell, maybe it's a mechanic that adds skill to the game — hazards are arguably "broken" in an overcentralizing sense, but I don't think anyone would argue that they're unhealthy for the game (though perhaps they're a bit too good in some metagames). They add a relevant dynamic to the game and complicate a lot of gamestates in an interesting and competitive way.