The next evolution: Team Liquid welcomes Atelier

September 01 2021




The next evolution: Team Liquid welcomes Atelier







Towa “Atelier” Kuriyama has long been pretty good at Smash. He got ranked 78th in the PGR 100 at the end of Smash 4, the first Smash title that he played competitively. But if we’re being honest, it’s his run in 2021 that made him a name to know and a name that a tier 1 esports org would look to sign.



In the last 6 months, Atelier (pronounced: ah-toh-ree-eh) has beaten kept, kie, rizeasu, Abadango, Raito, VoiD, T, Paseriman, KEN, Kome, Tea, and Zackray. Along the way, he’s made a very convincing argument for being the single best Pokemon Trainer in the word and a dark horse contender for top 20, if the world should ever return to normal and international competition resumes. Still, for the western Ultimate fan, it might all seem “out of nowhere.”







To trace Atelier’s roots, you’d need to look back to the 3DS.



Even if you didn’t want to know where Atelier came from, you ought to look back at 3DS anyways. A quiet success, the 3DS has a bizarrely influential role on Smash 4 as a whole. Technical limitations of the 3DS had lasting impact on Smash 4’s design - resulting in one of the franchise’s most classic characters - Ice Climbers - getting cut from the roster. (In retrospect, given how good grab was in Smash 4, this was probably a blessing in disguise.)



More importantly, Smash for 3DS sold very well - especially in Japan. It sold the 8th highest units of any 3DS game (according to public data from Nintendo) and it had sold 2,190,000 copies in Japan just six months after release. That was compared to just over 2 million in America - which is impressive when you consider Japan has just over a third of the population of the US.



Even more impressive was Japan’s early 3DS scene. The title had a surprising number of tournaments. Often online and with a weird “battlefield or FD only” format, these competitions weren’t the most legitimate - but they were arguably the most accessible Smash had ever been. These tournaments (particularly Tamisuma) were where many of Japan’s new generation got their start - including not only Atelier but Zackray.







It was here that Atelier first found competitive Smash. From there, he followed the classic route of looking up competitive VODS, seeing what the game could be, and suddenly yearning to reach that point. This is what drew Atelier into his local Kansai scene.



Kansai and Kanto are probably Japan’s two best Smash scenes and also two big loci of Japanese regional culture. Kanto is something of Japan’s newer Jewel - where Tokyo (and most of the country’s population) is. Kansai is where Japan’s old capital, Kyoto, is - as well as historical cities like Osaka. The two can have something of a regional rivalry and Kansai even has its own dialect that Kanto folk sometimes poke fun at.



This regional difference and rivalry can strike surprisingly deep. Kansai folk being known for an entirely different kind of persona. That traditional, polite, reserved, picture of Japan comes from Tokyo and Kanto, whereas Kansai people are known to be more outwardly emotional and direct. Once upon a time, during feudal civil wars, the rivalry was even geopolitical. Nowadays, it extends to sports - in form of Japanese baseball and also Japanese Smash.






(Or the Japanese FGC in general - as seen here in this Kansai vs. Kanto Guilty Gear crew battle)



Although, for Ultimate the rivalry is more lighthearted - like the modern East and West Coast Smash rivalries in the US. For Atelier especially, these rivalries mean little. The fast-rising Pokemon Trainer couldn’t care much less about regions and looks more towards individuals he needs to beat. All the same, keep Kansai in mind and you might spot some of the region in personality and play.



~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Do you think your playstyle fits your personality, or do you think they're really different?



I think my style is similar to my personality. I tend to be aggressive and impatient. Therefore, even in Smash, I tend to attack more than necessary, and attack without patience.



When was the first time you played Smash? What made you decide to join the tournament scene?



I have memories of first playing when I was an elementary school student. At the very least, I could do all the moves. Though I didn't really understand high level gameplay theory and I was bad. I remember using Marth and Metaknight in Brawl. Mostly I just played at my friend’s house.



I really started to get into it when I was in middle school, when I was around 15 years old. My parents got me Smash for 3DS and I really got into the netplay 1v1 matches. Little by little, the desire to become good at the game grew within me, so I started looking up stuff on youtube and eventually found videos for the tournament series UMEBURA. In the video, I saw high level stuff that I had never seen before, and thought to myself “I want to be able to play like that, and try to win a tournament.” From there, that's when I really started playing seriously.



Commentary often says you are the strongest player in Western Japan[/Kansai], but do you have any rivals in East Japan [Kanto]? Do you feel any kind of special significance for matches with East Japan?



I don't really have a rival. I simply want to be Japan’s number one player, so there is a wall of players that I have to overcome, like ProtoBanham and Zackray. I am strongly aware of this, and routinely practice with that in mind. I don't feel a particular rivalry with Tokyo players. Doesn't matter the tournament or the player… I play with the desire to beat whoever my opponent is equally.



[image loading]

Does the Japanese scene view other communities (America, Europe, etc.) as rivals?



I don't think the Japanese scene looks at other countries or regions as rivals, exactly. Rather, I think it is more common for Japanese players to develop rivalries with other Japanese players because they have opportunities to play each other so much.



You used Rosalina & Luma during Smash 4, but in Ultimate you switched to Wolf/Pokemon Trainer. These characters are pretty different, so why did you change?



Rosalina and Luma, in Smash 4, could lead with Jab 1-2, and it would start a yomi situation. This kind of core gameplay was fun and I used it a lot, but in Ultimate she couldn’t do it anymore so I dropped her.




~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



Smash 4, Yomi, and Gatekeepers



It didn’t take particularly long for Atelier to get good.



If he started getting competitive in 2014 or so, then in about two years he’d already started going toe-to-toe with top players and even beating Japan’s own gatekeeper: Hikaru.



Make no mistake, Hikaru is a historically great player who ranked 48th in Smash 4’s PGR 100 (top 50 all time, essentially), and made top 50 in Ultimate’s last ranking too. However, he usually sits just outside the top 10 in Japan's Player rankings (PR) and it’s not often he wins Japanese majors. At the top level, but not quite the very top, beating Hikaru in Smash 4 was a big accomplishment and good omen for a player’s career.






(Mirroring one another again, one of Zackray's breakthrough win was against Hikaru.)



This is not all speaking from fate or reputation either. It’s in part because Hikaru was one of the world’s best Donkey Kongs in Smash 4 and in that way he was one of the world’s hardest patience-checks. In Smash 4, DK and Bowser represented a heavy grappler archetype that could try the patience of the best players in the world. As heavies, these characters took damage and combos more easily and were easier to outmaneuver than top tiers.



But if they grabbed you at mid-percent while they had rage, you were probably dead. In a game with just two stocks, where shielding was very strong, this made them surprisingly good. Being heavy, they took a lot to kill too, hence the patience check. Most top tiers could beat them again and again in neutral, it’s just that if they won a quarter of those neutral interactions, they might win the set.



So, beating Hikaru usually meant that one of these aggressive, young 3DS-generation players had finally learned patience and how to play around a good grab game.







In Smash 4, Atelier played one of the game’s most oppressive top tiers in Rosalina and Luma. Often forgotten due to the raw DLC privilege of Bayonetta and Cloud, Rosa was a demon all her own - in big part due to her jab. Rosa’s jab was insanely low-risk and high-reward and since she could position a surprisingly durable Luma in a lot of places, the jab also gave her insane command of the ground and pressure on the ledge.



Atelier characterizes the jab as “start[ing] a yomi situation.” For the unfamiliar, yomi comes literally from translating the Japanese word for “reading.”



While yomi can be interchangeable with “getting a read,” it can be used as a concept in fighting games, where it gains more depth. Game designer and old school fighting game competitor David Sirlin helped popularize the term and introduced the idea of “yomi layers” in his article “Spies of the Mind.”



The idea is player one starts at yomi layer 0 - doing an option that’s purely good for their character. Then, player two finds an option that counters this and we move to Yomi layer 1. Player one, now getting countered, finds a counter to the counter and reaches Yomi layer 2. Player 2, now getting their counter countered, finds a counter to the counter’s counter and reaches yomi layer 3.



To keep everyone’s sanity intact, many scenarios reset at yomi layer 4, because a lot of times that original yomi layer 0 or layer 1 option will beat the yomi layer 4 option. This is because rock-paper-scissors is at the core of a lot of fighting games, only in form of throw-shield-strike.



[image loading]

(Courtesy of Corgian - original Twitter thread here.)



In this specific case, we could see “yomi layer 0” as Hikaru trying to grab Atelier. Rosa’s jab takes up so much ground space that it becomes a counter to DK's simple run-up, shield, grab, approach - AKA DK's yomi layer 0. In Smash 4, that run up, shield, and fish for a throw or a strike was a very common Yomi layer 0 so, Atelier saw Rosa’s jab as creating a “yomi layer 1” in a lot of matchups - not just against DK.



Rosa could take space easily and with little risk using the jab and then it would be up to the opponent to decide what to do. At this point, Rosa usually had some form of counter, to make for an additional yomi layer, and the opponent would have another counter for another layer.






(This is from a later set versus Hikaru that Atelier actually loses, but it best demonstrates how Rosa controls the ground.)



You can see this all play out in Atelier’s win over Hikaru because Atelier’s jab will cut off most ground approaches DK has. If Hikaru attacks Luma, then Rosa often has a chance to run and land a hit. If DK runs in and shields, Rosa can Jab, then grab. Hikaru can try for some big reads and big hits, like DK’s charged punch but it’s not reliable and is often punishable.



So, Hikaru starts to jump in order to open Atelier up. Now, it’s up to Atelier to go to another layer of Yomi by beating these jumps. Atelier does this very well, using a mix of up smashes as well good aerial timings and jump spacings.






(From the 2016 set, Atelier commanding the ground and air.)



In turn, Hikaru tries to return to the ground to create offense that counters Atelier’s jumps. Sometimes, Hikaru will manage it! He does take a stock in both sets. But Atelier is very patient and knows that a lot of Hikaru’s ground options are commitments against Rosa. Hikaru often has to overextend and to return to that first Yomi layer created by Rosa’s jab if he wants to win.



So, Atelier closes the set out by predicting one of Hikaru’s reads. He knows that Hikaru has been getting close, pressuring him to roll, then punishing. In the final moment, Atelier lands in the corner, shields, but rather than rolling in, he dashes right into DK’s forward smash, blocks the forward smash, then retaliates with up smash out of shield.






(It takes about 15 seconds to get to the up smash.)




It’s a smart victory because to beat Hikaru, you had to play smart. If you got reckless, stupid, or over-aggressive - as young, talented, mechanically strong players often get - then you likely wouldn’t beat him.



After beating the gatekeeper, Atelier officially entered the ranks - reaching 20th on Japan’s power rankings (PR) at the start of 2017.



These yomi layers are a strong way for Atelier to keep his offense controlled and well-managed - then and now. That’s especially vital in the Japanese scene because while Japan’s top-level tends to be a bit behind NA, Japan’s average tournament player (mid-high level) is notably ahead of NA’s. Having a good sense of yomi and how to work the layers is a sort of fundamental skill that helps avoid getting upset in a region with a huge talent pool.






(Atelier working the yomi but still getting Bayonetta'd. Japan’s 9B was both an early adopter of Brawl Ice Climbers and Smash4 Bayonetta - a sixth sense for Smash 0-to-deaths.)



Regardless, yomi didn’t mean everything in Smash 4 - especially for high/top tiers that were more elusive than DK - like Sonic, Sheik, ZSS, and Diddy. Or top tiers that gained way too much off of a single correct read (Bayonetta). Atelier’s success didn’t continue linearly from his 2016 uptick. He took a number of losses and eventually fell off the PR entirely, until he had a massive resurgence in 2018, at Umebura 31.



Umebura is one of Japan’s biggest tournament series and even though it tends to have lower entrant numbers, the actual quality of competition is some of the highest in Ultimate. Winning an Umebura at all is a big achievement and Atelier did it by beating T, Kameme, KEN, and Abadango - the 17th, 9th, 2nd, and 1st ranked players in Japan that year.






(A good sequence of defense that leads to a haymaker.)



Atelier cleaned up his play considerably to manage these wins. His game had overall become much smarter, with his disadvantage being safer and more effective. Instead of trying to counter-hit his way out of a lot of scenarios, he’d accept the disadvantage and mostly spot dodge or air dodge. His neutral was also more patient, with him sustaining pressure for longer because he wouldn't overextend as often.



However, his advantage state was the thing to watch. He still had a layer of insane reads but he’d also refined a pretty strong combo game that had surprising lethality for Rosa - a character more known for set play and ledge trapping than combos.






(A gorgeous combo using precise, late-hit aerials, an air dodge frame trap, and a missed tech.)



His tech chases and frame traps also became very sharp, with tech chase up smashes and air dodge frame trap up airs becoming common kill tools. As Smash 4 and 2018 came to an end together, Atelier was in a position to grow his success in Ultimate. A lot of that growth would come from the tech chases and the frame traps he’d been perfecting in Smash 4.




~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


Is there a player you consider a rival or someone you want to beat?



I want to beat Tweek. He is the reason I started using Pokemon Trainer, so as a player, he is special to me. [...]



Why did you choose Pokemon Trainer? Was this character difficult to master?



I saw Tweek use Pokemon Trainer at EVO 2019, and thought it was so great that I decided to pick him up. I don’t think I have mastered the character, but when I first started using Tweek’s as an example, I felt I improved quickly just by copying his play.



Where would you put Pokemon Trainer in your tier list?



I think he is in [the] top 10.



Your Squirtle is awesome! Lots of top players say, Squirtle is the strongest of the three. What do you think about that? What do you think about Squirtle’s potential?



I think Squirtle is the strongest of the 3 pokemon. Ivysaur and Charizard can become difficult when the opponent understands them and the match dynamic, but Squirtle is so fundamentally high mobility and fast with attack startup, that even if opponents understand, there is no problem.



Squirtle has yet to reach his highest potential. For example, there is no player in the world yet who can land the perfect highest damage down throw combo every time.



~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


Choose Your Starter // Professor Tweek



In any fighting game, there are characters that really excel in early metas. For sake of theme, we’ll call them starters.



These starters tend to have simpler execution, a wide array of tools for lots of scenarios, and a few neutral patterns and advantage options that take time to understand. With a cast of over 70 characters and a pretty new engine, Ultimate’s starter meta was a great deal different than what we see now. MKLeo was dominating with Ike all while Pichu and Olimar were menaces to most of the cast. However, the true perfect trio of starters in NA was Palutena, Lucina, and Wolf. In the early days, it felt like every top player had a pocket pick or main of one of these three characters.



Of the three starters, Atelier chose Wolf.



Wolf fits Atelier pretty well overall partly because Wolf has tools for most things and fits nearly everyone pretty well. But Wolf also allows Atelier to create similar Yomi situations, covering the ground with a good blaster shot, covering air approaches with great air speed and aerials, and threatening shield with a very rewarding grab. With Wolf, Atelier could pretty easily pick up where he left off with tech chases, juggling pressure, and a lot of tools to hop around Yomi layers.



Wolf also had an added advantage of playing well to Japan as a region, because the starter trio was a bit different in Japan. Mainly, Japan swapped Palutena out for Zero Suit Samus (ZSS) - a character Wolf handles pretty well. Wolf's fast air speed, strong projectile, and good anti-airs give him at least one tool that can answer any of ZSS’s.







Atelier’s Wolf was really proficient too. He placed 14th on Japan’s first PR basically being the country’s second best Wolf, right after Zackray (a top10 player worldwide). It helped that he could take apart prominent ZSS mains like Yamanyon, Shky, and Kuro.



While Atelier did well on Wolf and still uses the space privateer here and there, the character was not quite his soul-main. Much like so many other top players who once used one of the starters, Atelier would find a new main. Only, for Atelier, inspiration came from a surprising place:



Tweek’s EVO 2019 run.







Atelier directly attributes picking up Pokemon Trainer to Tweek’s run at EVO. At EVO, Tweek would not only make Pokemon Trainer look busted, but he’d do it by knocking half of Japan’s top players into the Losers Bracket.



He hadn’t only found his new main, but also a guidebook on how to play the character in a way that could beat the players in front of him. Watching Atelier’s recent big runs with Pokemon Trainer, you can clearly see shades of Tweek within it too.



In his EVO run, Tweek used Squirtle’s tools to create better yomi situations than any Trainer had done before. He’d use Squirtle’s super fast aerials and great mobility to basically shield camp his opponents. While in shield, he could punish any unsafe move very well. It’s something Atelier would later perfect in his own play.







Tweek would also chase his opponents really far into the air with either Squirtle or Ivysaur. Tweek’s Trainer was insanely hard to land against because of how he applied pressure with Squirtle’s insane speed and high-reaching up-b and Ivysaur’s huge aerials. There’s even an inherent mix-up dynamic within, where Squirtle’s pressure is very burst-heavy and fast, while Ivy’s is slower but longer-lasting. Switching between the two makes both more effective.







Even some of the smallest bits of tech, Atelier would take from Tweek, where he’d tinker it into something even more useful. Specifically, at EVO Tweek would air dodge and then immediately use Squirtle’s side special Withdraw. It created a way to break out of disadvantage that’s hard to visually read, surprisingly low-risk, and surprisingly effective. Tweek used it occasionally, where Atelier now uses it consistently.







However, none of this learning came overnight. If Tweek was the professor, then Atelier was still a new Trainer, just stepping into the tall grass. Atelier would actually fall down in rank on the Japanese PR after picking up PT. He’d experience a handful of poor results, including a pretty notable upset where he lost to a Kirby - one of the lowest-rated characters in the game.



While learning from Tweek would help Atelier get better, he’d ultimately get his breakthrough by surpassing him and by further optimizing the character.







~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



During 2021, what was the reason for your improvement? What was especially important for your growth?



With Smash 4 and Ultimate in the beginning, just from playing based on intuition, I felt myself improve but I increased my video review and study opportunities and I think that helped my results. When I watched my replays, I asked myself why and how I took damage, why I got hit, and tried to practice the next day and test with small improvements. I thought a lot about what the differences are between my Pokemon Trainer and Tweek’s Pokemon Trainer and in those points where we differ, which is more correct.



You have three pokemon to adapt to a variety of situations, do you think that kind of character suits an adaptive playstyle well?



No, I don't think it fits that well. I think if you want to adapt to your opponent that way, high mobility, basic characters like Sheik or Wolf are easier to work with.



However, I think it is easier to win in tournament by having your opponent adapt to you rather than adapting to your opponent. Therefore, with three characters you can use, you can toy with your opponents. This gives Pokemon Trainer a high sense of value.



Lots of Trainer mains change pokemon based on opponents, stages, or matchups. However, you will continue with the same Pokemon for a long time. Are there situations where you choose your Pokemon based on opponent playstyle?



Squirtle 50%, Ivysaur 40%, Charizard 10%. I try not to deviate too much from this standard.
For example, if there is a matchup where I think I have to use Ivysaur 90%, in that kind of matchup it doesn't make sense to use Pokemon trainer, in my opinion. In that case, I would use Wolf or Hero.



It seems like you often stick to the same pokemon to maintain match pace and advantage. How does pace and stage position influence your decision to switch pokemon?



This is putting it REALLY simply, but if I feel the situation is “even” I usually run Squirtle. If I have the advantage, I run Ivysaur and attack aggressively. If they are at percents where they can be killed easily, I use Charizard.



If the stage has a platform, in “even” situations I can also run Ivysaur. If it is a situation where I can knock off the opponent and interrupt their return to the stage, Ivysaur is good. If that is impossible, I can run Charizard. For this kind of strategy discussion, there is no end to small details or particular situations like that, so I will stop there.



I think it would be good if I can make a video where I can explain this kind of stuff.



You are really good at parrying. In this game, how strong do you think parrying is? As players come to understand the game better, do you think players will parry more?



Parry is strong but it also carries risk! Therefore, I don't think that everyone should necessarily be trying to parry more, or that with more game understanding will come more opportunities for people to parry. However, I think moves with long startup that are commonly used in certain situations (Yoshi’s Fair, ROB’s Nair, Charizard’s Bair, etc.) will be parried more, as more players develop their understanding of the game.




~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



Next evolution



It was only natural that Atelier’s PT would evolve past Tweek’s. After all, Tweak has been off the character for a long time now and Atelier has recently come into an insane streak of wins. Even as Atelier’s streak has calmed down, he’s still maintained a top 8 level in Japan’s nationals and could very well be Kansai’s best Ultimate player.






(The final stock against Protobanham demonstrates how Tweek could build percent and control space very well, but struggled closing out stocks enough that it nearly lets Protobanham mount a comeback.)



Reaching and maintaining such a level in such a competitive region demands growth. Even Tweek’s 2019 Pokemon Trainer would have issues that would get exploited by Japan’s deep talent pool. Namely, at that time Tweek had labbed out early and mid percent damage scenarios very well but against top players, getting the kill was a bit awkward.



At the time, getting the kill for a lot of characters was awkward since the set play of the game was still coming together. So it was fishing for a kill confirm at certain percents and if the opponent didn’t fall for the bait, then fishing for a kill move. Even with Ultimate being online a long time, players have discovered more pathways to take a stock off of a tech scenario, ledge trap, or even when someone’s pinned in shield.






(Tweek fishes for razor leaf, down throw, but Leo knows how to counter these options.)



In Tweek’s case, he’d build to high percents and fish for a razor leaf into vine whip confirm. If the opponent fell out of the percent range for that confirm, he’d switch to Charizard and try to land one of several raw kill moves - often via ledgetrapping. While not necessarily bad, Tweek’s relatively linear pathways to take a stock was a notable part of his losing to MKLeo in Grand Finals at EVO - not because Leo lived to insane percents but because Leo could more easily catch on to what Tweek wanted to go for and adapt.



Atelier builds on the groundwork that Tweek laid by fleshing these pathways out. One of the biggest areas of improvement comes in the tech chase - particularly with razor leaf. In 2019, a lot of Trainers used razor leaf in neutral or advantage, poking at the opponent, hoping to hit them while they approached or took space, then getting a confirm.







Atelier uses razor leaf more as a tech chase option - and it makes the kill confirm a lot more reliable. In the clip above, you can see that Atelier starts a tech chase with Ivy’s dair, then slips down from the platform, uses the razor leaf to catch Zackray’s tech roll, leading into a guaranteed vine whip.



What’s extra cool about these tech chases is that sometimes they could be a reaction - not a read. Ivy’s razor leaf’s hitbox lingers for a while, so it can last through a tech roll’s invincibility window and hit as soon as the enemy’s tangible. As a side special, it’s also easier to see which way the foe rolls and select either side (since you don’t need to b-reverse) as Ivy falls through the platform. Atelier hits these kills so routinely that it feels less like a read and more like a fairly reliable, reaction-based kill setup that Atelier’s perfected.







This tech chase, and all things Atelier perfects within Pokemon Trainer, come right back to the Japanese prodigy’s past. The tech chases, the frame traps, and even the yomi situations we see from his Rosa are now entering into his PT. For example, that Ivysaur dair tech chase against Zaackray's ROB pretty closely resembles a Rosalina dair tech chase Atelier perfected in Smash 4.



In Smash 4, Atelier had also gotten pretty good at frame trapping and juggling in a way that you can see in his Trainer now. With Rosa, he'd throw opponents up, wait for them to air dodge a hit, then using Rosa’s big, lingering up air hitbox to chase and punish the air dodge.







In Ultimate, he does something similar by threatening with razor leaf and vine whip. Vine whip is generally a strong kill option that chains out of throws and razor leaf, but at a lot of percents it’s possible to air dodge out of it. PT mains have relied on razor leaf confirms since Leffen’s run with the character. It’s something most players will look for in the matchup.



Atelier leverages that ingrained sense of matchup memory by building follow-ups to punish the air dodge most players use to get out of the confirm. In the clip below, Atelier uses this frame trap at the very end of a long string of beautiful aggression. Atelier threatens with razor leaf, falls, traces Akakikusu falling pattern, grabs, and gets the down throw kill confirm.







And then, there's the yomi layers.



In the Q&A above, Atelier says he prefers a character that pushes the opponent to adapt and creates yomi situations. In Smash 4, Rosa was well suited to that mold because she could stuff a ground game and run-up shield/grab style of approach that was all too common. In Ultimate, at the highest level, Pokemon Trainer is well suited for Atelier - but this time because of the air game.



In Ultimate, the common “yomi layer 0” is somewhat reversed to Smash 4. Rather than safe shielding on the ground, Ultimate’s abusable initial layer is the safe aerial. Every character in Ultimate (except Kazuya) starts their jump in 3 frames and usually has at least one aerial that’s under 10 frames and that can be auto-canceled to have less landing lag.



That means many characters can 1) get in the air quickly, 2) put out an aerial quickly, and 3) reduce the lag at the end of the aerial so it’s harder to punish. This is the widely complained-about “safe aerial” meta. Like Smash 4’s run-up shield, it’s a common gameplan a competitor can use when they know nothing about the opponent and just want a good, safe way to take space.





To create that yomi situation in Ultimate, it’s not as much about stuffing out the ground game as it is stuffing out the air game.



In Pokemon Trainer, Atelier has 3 different movesets - each of which has powerful ways to command air space and force the opponent into moving up yomi layers. In the above examples of tech chasing, you can see how Ivysaur can command space with huge hitboxes and very threatening confirms that would make players rather stick to the ground. Atelier notes that he often goes with Ivysaur in advantage or near platforms. That’s because Ivysaur’s threat zone is hourglass-shaped - very potent directly and diagonally above and below the cabbage dog.



Squirtle is the foil to Ivysaur that elevates Pokemon Trainer. If you hear a player rate PT in the top 10 or top tier, it’s often out of faith for Squirtle. Where Ivysaur’s big hitboxes and straightforward confirms made the character a menace in the early meta, Squirtle will likely be nuts in the late game meta for its rapid movement, bonkers frame data, and robust combos.






(If you can shield camp Tea’s Pac-Man, you can shield camp anything.)



Atelier uses Squirtle to cover the horizontal area that Ivysaur can sometimes struggle with. Where Ivysaur is great at catching full hops, commanding platforms, and juggling, Squirtle can do well at stuffing out aerials, beating short hop options, and commanding the air up close by shield camping opponents.



With every single aerial starting up in 6 frames or less, Squirtle quietly has one of the best out-of-shield games in Ultimate. These insanely fast options allow Atelier to be hyper-reactive because his moves are so fast that they punish most options on shield, are mostly immune to cross-ups, and can reach far enough to cover a surprising amount of aerial space.






(Squirtle’s great out-of-shield game also makes for great ledge trapping.)



PT is far from unbeatable - all these options have their counters, just like Rosa’s did. However, this command of the air means that Atelier can push the opponent to adapt to him.



In this way, Pokemon Trainer feels like the natural evolution for Atelier’s Smash career. The character may even fit him better than Smash 4 Rosa did. A Kansai player at heart, with a more aggressive personality and playstyle, PT gives Atelier's offense more reward, while also making his attack safer and more reliable.






(Very clever use of withdraw.)



Beyond just the aggression, PT gives Atelier a lot more to work with in terms of movement and defense too. Atelier uses Squirtle’s tiny body (and hurtbox) to evade lots of hits and Squirtle’s withdraw to not only effectively swing out of disadvantage but sometimes pressure shields. Not to mention, for whatever reason it seems like Squirtle’s up-b goes twice as far when Atelier uses it.







He also finds a ton of purchase through using parries, which further punish that initial layer of fast, easy aerials. The parry is difficult to execute and high risk, requiring you to drop shield within a few frames of a move hitting, but it makes certain aerials open to much bigger punishes. For PT, this is another improvement in offense and defense - making the shield game all the scarier and opening up for new kill avenues, like a parry into a Charizard up-smash.







In the realm of edgeguarding, he’s really mastered Ivysaur’s command of the ledge. He’ll use vine whip cancels to feint like he’s snapping to the ledge, then throw out a razor leaf or just float there menacingly. It’s a lot of weird micro-movement that overwhelms the opponent and lets him extend his time around the ledge. It’s part of why a lot of Atelier’s opponents will opt to jump and recover high against him even though he routinely hard-punishes these high recoveries too.






(Vine whip cancels and razor leaf movement. About 45 seconds later, he kills Zackray with a high Squirtle up-b.)



When you add all these adjustments together, you get what is probably the consensus-best Pokemon Trainer in the world and some of the most finely-tuned aggression in the game. When you consider how young Atelier is, you have one of Ultimate’s most promising players as well. Especially since he hit a stride like this in Smash 4, only to be cut short by the game’s last breath. Now he’s hitting that stride with plenty of Ultimate left to play.



For all those improvements, Atelier still doesn’t have any great laurel to rest on. He’s beaten some of Japan’s best, but they’ve come back to beat him too. Tea reset the Grand Finals at Kagaribi 3 off of an advantage state just as cracked as Atelier’s.



And Zackray, Tea, and Protobanham are only the first three of an elite four. If Atelier wants to beat Tweek, he’ll need to work the yomi and the character deeper than he ever has before. The shiny new tier one sponsor is mostly the chance to compete full-time and travel. To realize that chance, there will need to be no end to Atelier’s evolutions.



Writer // Austin "Plyff" Ryan
Graphics // Felipe "Humanmgn" Magaña
Translator // Jonathon "Majin Obama" Metoyer








Postscript: Dogs, tags, poker, new characters



When you play Smash, outside of winning, what gives you the best feeling?

I can't say it is the best feeling, but it makes me really happy when I learn how to use a character that I couldn’t use before.



You often retweet pictures of dogs. Do you have a dog? What kind?



We have a chihuahua at my house. You can’t feel a sense of malice from dogs, so they are one of my favorite animals. They are really precious.






How did you decide on Atelier?



Atelier is the name of a cafe I used to go to often when I was really young.



Do you have any interests or hobbies outside of smash?



I'm just a beginner, but I'm interested in poker.