How a Super Smash Bros. Fighter is Designed

Is there any game with a stranger crossover cast than Super Smash Bros.? It is one of the only places where Pikachu, Mega Man, the Wii Fit Trainer, and the blocky builder from Minecraft can all end up in the same fight.

Across five games, six consoles, decades of releases, and piles of downloadable additions, the series has featured 89 different fighters from roughly 40 franchises and dozens of developers. Those fighters are not costumes or brief cameos. They are distinct fighting-game characters with unique animations, movesets, strengths, weaknesses, and tiny references to the games they came from.

That creates a design problem with several layers. Each fighter needs to be recognizable to fans of the original character. Each fighter also needs to be playable inside the shared language of Smash. And once the character enters the roster, they have to sit somewhere inside a competitive ecosystem where no one should simply dominate every match.

A Smash fighter generally goes through about eight major design steps. First, the team has to decide which characters are even going to make the roster. With thousands of possible heroes, villains, mascots, oddities, and historical representatives to choose from, the decision usually comes down to a few pressures.

One pressure is popularity. As early as 1999, only months after the first Nintendo 64 game, Masahiro Sakurai was already asking fans to suggest fighters for a possible sequel. Bowser, Peach, Ganondorf, and Mewtwo all ranked highly in that poll, and all appeared in Super Smash Bros. Melee. Later polls and ballots helped point toward fighters such as Wolf, Bayonetta, and Sora.

Global recognition matters too. A character can be beloved in one country and almost unknown elsewhere. Some candidates have been passed over because the games they represent are not widely recognized outside Japan. Smash is also partly a museum of game history, so the roster has to represent franchises, systems, eras, and strange hardware moments. Ice Climbers helped represent the Famicom era. Mr. Game & Watch stands in for Nintendo's early handheld devices. R.O.B. represents a very particular plastic chapter of NES history.

The team also thinks about balance between franchises. In Melee, Wario was cut partly because the roster already included many characters from the Mario universe. That kind of restraint has loosened over time, especially with the many Fire Emblem representatives, but the underlying problem remains: every character added changes what the whole roster appears to value.

Some choices are practical. Nintendo may want a fighter to promote an upcoming release. A character may be easier or harder depending on budget, schedule, modeling needs, or technical constraints. Wolf was easier to build than Krystal because he could lean on the existing Fox McCloud structure. Ganondorf entered Melee partly because the team could reuse Captain Falcon's animations and borrow a model from a GameCube tech demo.

That is the logic behind clones, alternatives, and echo fighters. A new character model can be paired with an existing rig and moveset to increase the roster without demanding the same amount of work as a completely new fighter. The result may not be as novel, but it can still serve popularity, representation, and production goals at the same time.

Finally, the original creator or IP owner has to approve the appearance. That has become easier as Smash has grown. Early on, Sakurai had to work hard to convince the Pokemon Company to lend Pikachu and Jigglypuff to the first game. Later, developers were actively asking for their characters to be included, with Yuji Naka and Hideo Kojima personally requesting Sonic and Snake for Brawl.

Popularity, history, logistics, and permission all matter. But one factor can override almost everything else: theme. Sakurai places enormous importance on whether a character has something unique to bring to the roster.

A new fighter needs a characteristic, ability, attack, or structure that meaningfully distinguishes them from the others. That can be dramatic, like a fighter made of two characters controlled together, a fighter who transforms mid-match, or a fighter who can alter the stage layout. It can also be more focused, like storing a projectile for later, using different colored Pikmin, or fighting almost entirely through grabs and counters.

To sharpen that uniqueness, Sakurai tries to boil the fighter down to one sentence: a pithy description of what the character adds to Smash. The design documents make that visible. Rosalina and Luma become "Puppet Fighter." Meta Knight becomes "Pointed Sword Dancer." Snake becomes "Heavy Weapons Specialist." Duck Hunt becomes "Third-party Cover Fire."

If the sentence is only "dude with sword," the character has a problem. That is why Chrom did not originally make the roster as a full new fighter when Sakurai was choosing a Fire Emblem: Awakening representative. Chrom did not offer enough that existing sword fighters such as Marth and Ike lacked. Robin, by contrast, could introduce magic tomes, a breakable sword, and limited-use attacks. "Sword and Magic, Limited-Use Attacks" is a much stronger theme.

That single-sentence theme becomes the center of the fighter. It does not solve every balance, animation, or visual problem, but it tells the team what must survive every compromise. Without that center, a fighter can be famous and still feel redundant.

Once the theme is clear, the team can decide how the character should feel in motion. A fighter is defined by a large set of parameters: run speed, traction, jump height, gravity, air control, how far they fly back when hit, and many more.

Those numbers can make one character feel slow and heavy, another nimble and slippery, and another strange in the air. They can also echo how the character behaves in their original game. Mario and Sonic, for example, have jump heights and run speeds that roughly resemble the way they move in their own series.

The translation cannot be literal. Bowser would be almost unusably slow and heavy if he matched his appearance as a Mario boss. The goal is not physical accuracy. The goal is a version that feels truthful enough while still working inside Smash.

Even before moves are added, stats can create sharp differences. Incineroar sits near one end of the speed spectrum, while Sonic sits at the other. Jigglypuff's floaty jump gives a completely different feeling from Falco's. A roster starts to become legible when the numbers make each body feel distinct in the player's hands.

In one sense, every Smash fighter is the same. The series was built around approachability, so characters share a standardized control scheme rather than having unique fighting-game inputs for every roster slot. That constraint is one of the reasons Smash can host so many wildly different characters without becoming impossible to learn.

Within that formula, there is still a huge amount of room. A basic smash attack can differ in damage, knockback, range, area of effect, startup, recovery, and status effects. Special moves can become projectiles, counters, traps, charge attacks, command grabs, random RPG selections, copied abilities, stat changes, vehicles, tools, or stranger one-off ideas.

Many moves come directly from the character's home series. Kazuya and Terry lean heavily on their fighting-game identities. Mega Man references robot master powers. Diddy Kong gets peanut popguns. Mario carries the water-spraying companion from Super Mario Sunshine.

But faithfulness still needs adaptation. Robin can use dark magic. Ness and Lucas can use PK Starstorm. Snake can jump even though that is not how Metal Gear Solid works. Captain Falcon and Fox needed hand-to-hand moves because their original games put them in vehicles, not platform-fighter duels. The reflector shield and Falcon Punch are Smash inventions that make the characters work in this new context.

Moves also have to be scaled to Smash's speed. A Tekken attack that feels fast in Tekken may be far too slow inside Smash, where the quickest moves launch in a fraction of the time. Translation is not copying. It is finding the version of the character that reads correctly under a new set of rules.

With stats and moves in place, the fighter has to be balanced. Sakurai is not interested in perfect equality across the entire roster. Some characters can be deliberately underpowered, like Pichu in Melee, because winning with them creates its own kind of style. Still, no character should be able to dominate every match without serious answers.

The tempting solution is to sand down every advantage until the roster becomes fair. Sakurai generally avoids that, because it can make the whole game feel flat. If every strong move is softened and every unusual trait is normalized, the fighters begin to blur together.

Instead, Smash often accentuates a character's strength and pairs it with a clear drawback. Ganondorf hits brutally hard, but he is slow and has limited range. Mewtwo has strength and rapid-fire tools, but it is light and easy to launch. Link's projectiles are powerful, but his walk speed and jump are limited. Simon's whip has huge reach, but the recovery time leaves him exposed when he misses.

There are hundreds of parameters that can support this kind of tradeoff. Banjo and Kazooie's Wonderwing is a powerful charge attack, but it can only be used a limited number of times per stock. Charizard's Flare Blitz has strong knockback, but the character takes damage from using it. An attack can be made stronger by giving it a longer wind-up, a longer recovery, a smaller hitbox, a limited resource, or a risk that appears only when the move fails.

That is the useful lesson: balance is not always about reducing power. Sometimes it is about making power specific, readable, and costly.

A Smash fighter needs a huge number of animations. Attacks and special moves are only the beginning. The character also has to walk, run, jump, take damage, grab characters, grab ledges, taunt, celebrate, and perform a Final Smash. It is easy to see why a single fighter can take roughly a year to produce.

Sakurai starts with a master list of required animations and written instructions for how the character should move during each action. The team may use illustrations, screenshots from the original games, and even poseable reference figures to communicate the intended stance or motion.

The most important pieces are the key poses. An attack can be understood through four major poses: idle, wind-up, attack, and follow-through. The idle pose shows the basic stance. The wind-up signals that an attack is coming. The attack pose is the damage moment. The follow-through shows the character recovering and unable to move immediately.

Each pose needs to be dynamic and informative. The wind-up must tell the attacking player what they have triggered and tell the opponent what is coming. The attack pose may linger during hit-stop, so it needs to sell the impact. The follow-through must show vulnerability. The animation can use squash, stretch, exaggeration, and rig distortion, but clarity comes first.

That clarity matters because Smash is fast, visually busy, and often viewed from a pulled-back camera. The silhouette has to read immediately. A move cannot ask players to parse subtle body language while particles, items, platforms, and three other fighters are all moving at once.

Responsiveness also shapes the animation. The wind-up often snaps into place on the very first frame after the player presses the button. That can look abrupt, but it makes the game feel immediate. Transitions may be sped up rather than smoothly interpolated so a kick feels sharper. King Dedede's hammer takes a long time to swing because the animation needs to sell its weight. Zero Suit Samus's neutral jab can skip the wind-up almost entirely because its purpose is instant speed.

Frame data then becomes part of balance. A move's startup and recovery are not just animation details. They are numbers the designers can tune to decide how powerful, risky, punishable, or responsive a fighter should feel.

The fighter still needs to look right. That is harder than copying a character model from another game. Outside of extreme outliers like Mr. Game & Watch and Steve, Smash wants its fighters to feel as if they belong in one shared universe.

If every character simply matched their official design, the roster would become visually incoherent: realistic humans, cartoon mascots, tiny creatures, giant monsters, old sprites, modern models, arcade drawings, and anime heroes all colliding without a common language.

Several techniques help unify the cast. Brightly colored characters are often desaturated so they do not clash with more subdued designs. Proportions are changed so Ridley, Olimar, and everyone else fit within a workable size range. Simple characters can receive extra surface detail, while realistic characters may lose fussy detail so the whole cast sits at a similar level of visual complexity.

That is why Olimar's suit gains pockets, straps, and small widgets, while Donkey Kong gains rounded teeth and toenails. The point is not realism for its own sake. It is cohesion.

At the same time, those changes cannot betray the original character. Ness keeps his simple, toy-like eyes. Pokemon generally look like plastic models rather than furry or scaly animals. Sakurai has said that he did not want to betray the original creators, a conviction shaped partly by his own discomfort with certain depictions of Kirby.

The team also has to decide which version of a character to represent. Pac-Man has appeared as a pixel circle, arcade cabinet art, cartoons from different eras, and some deeply strange designs. Smash uses a version closer to the original sketches. Mega Man uses an NES-era design, while Sonic uses a more modern depiction.

Some characters barely have a modern reference at all. Before Brawl, Pit from Kid Icarus was a tiny NES sprite, a manual illustration, or a peculiar animated version. To update him, Sakurai and the team looked at Link: another NES-born hero who had evolved across generations. They asked what Pit might look like if he had followed a similar path, then built a modern design from that imagined trajectory.

Once the fighter is complete, the team still has to reveal them. The Smash reveal trailer became its own design object, and the idea came partly from a production disappointment.

Brawl included expensive CGI scenes for the Subspace Emissary mode, but many of those scenes appeared online before the game released. After that, Sakurai stopped using that kind of cinematic material inside the games and began using it as a marketing tool to introduce new fighters instead.

Like many other parts of Smash, the trailers often begin with Sakurai. He writes the sequence of events, then works with a studio through storyboards, rough previz, and final production. The goal can change from trailer to trailer. Some are built to trick the audience, such as a reveal that initially appears to be something else entirely. Others are built purely around hype and the moment of recognition.

External constraints still matter. The Ridley reveal had to be toned down because an early version was too intense. The Minecraft reveal had to feature Steve and Alex equally, following guidance from Microsoft. Even the celebration of a fighter is shaped by tone, brand, audience expectation, and approval.

The whole method becomes easier to see through one fighter: Little Mac from Punch-Out!!

He represents an important moment in Nintendo's arcade history. His theme is clear: a super-offensive, up-close brawler with a unique Power Meter. His stats support that identity with outstanding ground speed, helping him close distance and stay dangerous in the pocket.

His moves are built around boxing. His balance comes from the contrast between those powerful punches and his terrible aerial ability. He has a weak jump, poor recovery, and bad aerial attacks, so the character is frightening on the ground and vulnerable off it.

His animations are short and snappy to sell the force of his punches. His visual design is largely based on the Wii game, but with added detail so he fits the Smash roster. His reveal trailer uses a comic book style, partly because it fit the character and partly because it kept production manageable.

That is a complete fighter design in miniature: history, theme, stats, moves, tradeoffs, animation, visual adaptation, and presentation all pointing toward one readable idea.

There is one more constraint that affects the roster: the age rating. Super Smash Bros. Ultimate received an A rating from Japan's CERO board, meaning it was suitable for all ages. That is surprisingly strict for a game that includes characters from series with much older ratings.

To keep that rating, some designs and attacks had to change. Mythra's outfit was adjusted with tights and a fuller shirt. Snake does not use knives or guns. A King of Fighters stage includes many famous brawlers but leaves out Mai. The Wii U version of Smash even risked schedule trouble because models for female characters, including Palutena, needed repeated revisions.

A crossover game does not simply import characters. It imports the problems around them: audience expectations, ratings boards, brand rules, animation readability, roster balance, visual cohesion, and the need for a single design language. That is why a Smash fighter is such a delicate balance. The team has to honor the character's legacy through visuals, movement, attacks, and references while still making them a balanced addition to the roster and a coherent part of the Smash universe.

Most game creators will never put 89 famous characters into one fighting game. But the process still has useful lessons: define the character's theme before the details, make strengths strong enough to matter, pair those strengths with readable weaknesses, treat animation timing as design, and adapt visual style without losing identity.