An Open Letter to Nintendo on the State of Zelda in Smash


Smash Apprentice
Feb 9, 2014
EDIT 1: I posted this a while ago in a general forum, but it has now found a home in this thread. Though Smash Ultimate seems mostly complete, it is by no means finished. From the footage we have seen from E3 thus far, the development team has gone out of their way to modify Zelda's moveset, even if only in minor ways. Though I highly doubt they saw this letter specifically, many of the changes I propose here line up with the changes they have given to Zelda already. This makes me incredibly happy. So if you care to read on, feel free. It can only do good.

To Whom it Does Concern,

According to Nintendo’s CSR report of 2017, “We strive to constantly improve the quality of our products by learning from consumer feedback.” This feedback I have provided is particularly detailed and is distilled from a universal and personal response to the state of Zelda in the previous three Smash titles.

In this letter, I hope to accomplish several things. First, I would like to analyze the design of Zelda across all of her Smash incarnations. Second, I would like to provide a critique that has been informed by four years of intensive play and discussion of Smash with several individuals, offering potential solutions to the problems Zelda faces in the course of a game. Finally, I shall explain why it is important that Zelda possess a fighting chance in Smash.

Before I begin, let me provide a brief introduction of myself. My name is Dominic Behler, and I am a student of English literature at Marywood University in Scranton, Pennsylvania. My “tag” is Prynne. I began playing Melee and Brawl as a freshman in high school, about 7 years ago, and was enraptured by the originality of such a fighting game. Some may debate whether or not Smash is a fighting game, but in my opinion, the Smash games are not only a brilliant party game, but arguably the most involved and intricate of all fighting games, combining movement and combos in the most unorthodox manner. In my senior year, I began to pursue the game in a more dedicated manner, by attending tournaments and researching the involved techniques and mechanics of the game. It has been a rather ardent hobby of mine, pouring over internet forums and watching the matches of the best Smash players to understand all aspects of the game and its design. Smash is a party game first and foremost, but there is a competitive community that does exist. Therefore, my analysis is informed from a history of both casual AND competitive play, which are linked in much the same way casual chess between friends is linked to competitive chess between rivals. The goal in both is to win. Otherwise, what is the point of playing a naturally competitive game at all? If Zelda is unsuited to fight in one setting, she will be the same in the other. In my experience, Zelda is neither good in a casual four person game with items, nor is she good in competitive one-on-one tournament.

Princess Zelda has never been an excellent fighter. I understand, this is not terribly surprising, she is a Princess after all, a member of royalty. In Super Smash Bros. Melee, Super Smash Bros. Brawl, and Super Smash Bros. for the Wii U and 3DS, she is substantially underpowered, which has lead to the dismay of many a Zelda main for almost sixteen years. Before I analyze and critique her moveset, however, we have to look at the source material: The Legend of Zelda game franchise.

In The Legend of Zelda series, the Princess frequently finds herself trapped or possessed by Ganon, turned to stone, or in some way incapacitated for the duration of the story, save a few key exceptions, such as in Ocarina of Time, where she transforms into Sheik to assist Link, or Majora’s mask, where she does not appear, physically, at all. She is frequently depicted wielding powerful magic, ranging from creating light arrows to restraining Ganon completely. Additionally, she has shown her skill with a bow on several occasions. She also has musical talent, playing both the Goddess Harp in The Legend of Zelda: Skyward Sword and the Ocarina as well in The Legend of Zelda: Ocarina of Time.

Zelda’s Melee design was ingenious. The implementation of the three spells that Link uses throughout Ocarina of Time was a peak of creativity. Her Lightning Kicks were also a nice touch, providing her threatening moves to zone and punish opponents at close range. The brilliance that was “Transform” allowed one player access to two characters, at the cost of switching their style and adapting. All of these came together to create fun, magical character. Unfortunately, all of her specials suffered from a lack of power, speed, and overall reliability. Several of her normal and aerial moves could be downright useless. Her jab came out on frame 11, and was surprisingly weak for the long startup. Down Tilt lacked for range and would not combo opponents until very high percentages, leaving Zelda at a disadvantage even if she succeeded in hitting her opponent. Down aerial was far too weak and often left her at a disadvantage even after landing it. Her up aerial was surprisingly weak for its speed (or lack thereof). Her Smash attacks were of considerable strength, but would frequently fail to land their strongest hit, usually because the opponent would fall out of the multi-hits before the final launching strike. In versions 1.00 and 1.01 of Melee, they worked slightly better due to a change to Smash Directional Influence mechanics in the final 1.02 version (which governed the properties of multi-hitting attacks below 1%), as well as the PAL region release. This problem would also occur in Brawl. To make matters worse, she also suffered a frame 12 grab, which although had decent range, was still inferior to the grabs of faster characters with better throws, such as Marth, Fox, Falcon, and Peach. On top of all of this, she was slow, had a larger than average hurtbox in that she was very tall, had a very slow fall speed to hinder the use aerials near the ground, had poor aerial mobility, and died extremely early to attacks. In all, playing Zelda felt like an uphill battle to all players, and Sheik, the character who was not supposed to kill, ended up having more reliable kill power than she did.

According to Sheik’s trophies in Melee, the intended strategy for playing Zelda and Sheik was to switch between them as needed. Sheik had the ability to rack of damage quickly, while Zelda had access to powerful KO moves. However, Sheik did not have any trouble netting KO’s without Zelda. To transform into Zelda for any reason was an enormous liability.

These flaws are understandable considering the game was given the very small development window of about a year. Therefore it is unfair for me to nitpick Melee as I am now. However, Super Smash Bros. Brawl for the Wii, was delayed several times to ensure its quality, making her faults in Brawl less excusable.

There were some exceptional changes made in Brawl, and I really commend the designers for addressing some of her issues. Transform returns, and because of Sheik’s reduced KO power, Sheik had to rely on Zelda to net KO’s at a standard percentage. However, the increased loading time made this an impractical strategy in battle. In the 2.5 seconds during the transform time, an opponent could easily react to the sound clip and punish you for switching.

This time, Din’s Fire actually allowed Zelda to attack at range for the appropriate amount of risk. It functioned more in line with other projectiles in the game, despite leaving her quite vulnerable still, and putting her into freefall. Her grab was faster, if only by two frames. Her kicks were reimagined as potent finishers, rather than the spacing tools that they were in Melee. Her down aerial gained a powerful sweetspot. Up aerial was sufficiently powerful for its startup time. All of these were right on track, but she became less used when players came to understand her repeated shortcomings. The hitbox placement of her smashes was superior to the Melee versions, yet characters could still fall out of them. Four of her aerial attacks now killed, but the difficulty in landing them was still impractical given Zelda’s terrible mobility, both on the ground, and once she jumped into the air.

I looked to the future, where, with 13 years of time to examine the mistakes apparent in Zelda’s design, she may be improved.

Moving on to Super Smash Bros. for the Wii U, I must admit that perfect balance is impossible, and that not all characters can be equally good. Even so, some characters are bad by the virtue of their initial designs. Bowser, Ness, Yoshi, Luigi, and Mario were all, at some point, unplayably bad in past games. Through various fixes and redesigns, their considerable flaws were mitigated in SSB4. It seems evident from the balance patches that the game designers were acutely aware of large discrepancies between characters’ capabilities. Still, Zelda suffered from tremendously crippling design flaws.

Her combo game was far below average when compared to the rest of the characters. Her Forward Smash had too much ending lag and start up time for only slightly above average kill power. Her forward tilt lasted only 2 frames and still lacks for range. Her Farore’s Wind had a tremendous amount of ending lag, and was difficult and risky to control, especially when aimed into the stage from the air, as one must travel the full distance when teleporting, often sliding off platforms or the stage and falling to their death, where before, they would end safely on the stage. As for the substantially powerful Farore’s Wind “elevator” kill combo, the first disappearing hit frequently failed to combo into the second reappearance hit, even at a percentage that it should net a KO.

Although she was faster than before, it was not by a significant degree. She still suffered from the same attribute problems as before, which are her floatiness, lack of weight, great hurtbox height, and lack of speed. Zelda was still unnecessarily slow for her power. She was still incredibly light, meaning that she will die much earlier from kill moves than other characters. Zelda would often die at substantially lower percents, which is illogical, considering that she does not have the mobility or kill power to justify this balancing decision. These poor attributes, in tandem with her poor moveset, created a very bad time for the Zelda player, to the point where one might question her inclusion in the game at all.

Din’s Fire was far worse than it was in Brawl. It could not reliably control space as it must in order to provide the Zelda player with a sense of control. It was weaker, dealt less damage, and was less maneuverable - all departures from its previous design. On top of all of that, when Din’s fire was used in the air it still sent her into “freefall,” a problem that has plagued Zelda since Melee. Was this move really so powerful or useful that it should send Zelda to her death when used offstage? It was not a recovery move. It offered her no mobility or protection. What about this move was so powerful in the air that it must send her to her doom if used offstage? There was no reason for this move to prevent Zelda from performing actions after its use, nor is there any reason for it to send her to the depths of hell if used offstage. If Bayonetta can still act after using two comboing recovery moves, Zelda should be actionable after using her basic fireball attack.

Zelda’s new Down B was underpowered, and although it may do some decent damage if both parts of the move connect, this was exceedingly unlikely to occur. It takes a long time to charge, can be reflected, and only takes a measly 13% before being destroyed, after which one must wait a whole 9 seconds to use it again. Compare this to Rosalina and Luma, who possess some of the best range and power in the game together. Luma can take 50 damage before disappearing, and the Rosalina player needs to wait 3.5 seconds more before Luma returns. Perhaps it would have been better if one could hold the charge or cancel out of the move like Samus can with her charge shot. The phantom could be deflected just like a charge shot, so why not? Allowing for a reflection was a poor choice and an additional limitation to an already underwhelming move. Phantom Slash did not fill the hole that was left from losing transform, the only way that Zelda could access her one true chance in a fight – Sheik.

Zelda’s grab was substantially slower than the grabs of a majority of the cast, coming out frame 10 of the animation, as opposed to Mario’s frame 6. Mario’s grabs, specifically down throw, gave him great damage from repeated up tilt attacks. Zelda’s down throw was the only throw she could combo with, giving her a free neutral aerial on most characters at low percentages. However, the rest of her throws are too weak and deal too little damage to warrant a frame 10 grab. Only her back throw will net a KO, and only at unreasonably high percentages, even with rage. A frame 10 grab would be understandable if each throw did 12% or more in damage, with at least one reliable kill throw, similar to Ness’s Back Throw.

Finally, I would like to talk about Lightning Kicks, and there is much I must say. These two aerials are Zelda’s most iconic moves according to many players. They possess speed, strength, and style in droves. In previous games, her kicks had substantial power to compensate for their risk. In Melee, the risk was lesser, but it was acceptable, for the rest of Zelda’s available options were also lesser. In Brawl, the kicks were harder to land, but slightly stronger. In SSB4, these two aerials were Zelda’s worst moves.

First, let us discuss their speed. The startup time was only one frame slower than previous installments, but the ending lag on each of them was enormous. Not only would they fail to auto cancel/smooth land from a short hop (the lowest possible jump height), they have absurd amounts of ending lag. I once heard a friend joke that Zelda will land on the ground after a kick and still be in lag at the results screen. In previous games, Zelda was able to perform two Lightning kicks in one short hop. I understand that for this game, a conscious decision was made to prohibit certain characters from doing this, while also making it riskier to use powerful aerial KO moves, as shown in similar lag on Captain Falcon’s Knee Smash.

However, the lightning kick was already sufficiently difficult to land without the need for extra ending lag. The hitbox placement is comparable to Brawl, if not slightly smaller, with only one frame of sweetspot. The Lightning Kick was harder to land than Jigglypuff’s rest. Although it would make sense to make a powerful kill move from the air risky by giving it high ending lag, this seemed to be unwarranted given the vast array of limitations already imposed on it, with a small hitbox and the smallest possible window for timing.

To make matters worse, the Lightning Kick was less powerful than it has been before, now with only average kill power. I have landed lightning Kicks on average weight opponents at 95% and they have survived, even near the edge of stage. Worse still, I have landed lightning kicks on opponents at 110% and they have survived. For the risk involved in using this move, a Lightning Kick should kill noticeably lower, like that of similar moves in difficulty, such as Captain Falcon’s Knee Smash. Therefore, Lightning Kicks must be one of the following:

A) They were too weak to have a one frame sweetspot.

B) They were too risky to have a one frame sweetspot.

C) They were too risky to be as weak as they were.

A change to any one of the above would have gone long way. I believe that the landing lag of the move should have been cut considerably, and that both forward air and back air should at least autocancel from a short hop. An increase to the size of the hitbox or the window of opportunity would also be necessary.

This is not to say there were no good changes. On the contrary, Zelda now has a small but reliable combo game, which is better than it ever was before. Her smash attacks are more reliable, and I found many of the modifications to her ground moves and aerials refreshing, such as her up aerial and down aerial. Her up tilt is now a combo move, which is sensible, and her Up Smash is far more powerful. Down tilt is now a wonderful combo starter, leading frequently into grab. Her jab is far superior to its previous incarnations. Her neutral aerial is very well designed, despite the smaller hitbox on the final hit. Opponents rarely fall out of neutral air once it has begun. Her teleport functions like most risky up-B kill moves now, and although unreliable, can net some early KO’s. However, all of these adjustments amount to little overall improvement.

When we saw that Zelda was returning as a fighter, we had great hopes that, perhaps, the third time would be the charm.

Several characters have already been improved through the balance patches, including Shulk, Mewtwo, Link, Marth, Lucina, and Robin. Zelda desperately needed a wide spread of buffs as these characters have received.

As of patch 1.1.7, Zelda received only minor improvements. According to many players, she is not only considered the second worst character in SSB4, but also the most consistently horrible character across all Smash games. Even without consulting “tier lists,” any sufficiently dedicated Smash player could tell that Zelda is weak. With years of my life dedicated to observation and analysis of the balance of the game, it is clear now that this has become a real problem. Many players have decided to move on from Zelda and stop playing her. In fact, many Zelda players have abandoned SSB4 altogether.

Coincidentally, the closest equivalent to Zelda in past games has been Sheik. Sheik trades Zelda’s powerful knockback for mobility and high damage combos. Sheik can control the pace of a game at a distance using her needles, as Din’s fire may. Additionally, Sheik’s normals and aerials do what Zelda’s should, zone out the opponent through their sheer size/duration and begin combos. Through the game mechanics themselves, I was told not to play a character that was included in the game as a perfectly legitimate choice. Those who would play Zelda often play Sheik instead, as Sheik can do all that Zelda can, but better. It isn’t as if players dislike Sheik, far from it. The fact of the matter stands that Zelda is in the game, and is borderline unplayable even in a casual setting. If a character is put in a game, it stands to reason that they should have a fair fight with the rest of the characters.

Zelda’s design is remarkable, for it is perfectly true to form. Zelda has not been shown to fight for herself in any Zelda game until Breath of the Wild was released. Through the mechanics of the game, players feel like victims when playing Zelda. They feel trapped and unable to fight back, for every other fighter is provided with greater moves and attributes.

But things have changed since then. For one, The Legend of Zelda: Breath of the Wild was released, showcasing a scholarly Zelda with some serious magic. This design for Zelda aligns closely with the conception held in the minds of those who consider her their “main” character.

With the fifth installment of Smash on the horizon, I had to act quickly. Three times in a row, Zelda has been underwhelming and un-fun to play. However, it is clear from the small trailer released during the March Nintendo Direct of this year, 2018, that the version of Link that appeared in Breath of the Wild will be coming to Smash for the Switch. This opens many possibilities for the design of Zelda, ranging from a simple visual overhaul with her moveset tweaked in accordance to the information provided here, or a complete redesign of her moveset to reflect her character in Breath of the Wild.

Zelda is first and foremost a magician, meaning that she ought to have a great number of total options through virtue of magic. Providing her with a teleport, a reflector, and a remote projectile were all fantastic to this end. Zelda fights with distances in mind. Her moves seem to suggest that one must be mindful of their position relative to their opponent, and choose these options accordingly. Therefore, it is a safe assumption that she can be kept relatively immobile as a character. Her movement is not overtly important, as she seeks to fight at range, or in instances where she can punish an opponent’s uninformed attempt at attack.

Din’s fire should be much faster in all regards so that she can bring opponents to her. Instead of a slow teleport that can potentially kill early, she should have a teleport with high startup and low recovery, which would be enough to provide her with a basic positioning tool such as the one she had in Melee and Brawl. These changes to these specials would go a long way to helping her keep opponents at a safe distance. Her immobility was further compounded when she was unable to protect herself. Therefore, she must have a protective reversal ability, such as Nayru’s Love to alleviate the overwhelming pressure available to faster characters.

An array of ranged normals is necessary to stop approaches, poke opponents, and potentially begin combos. As for her throws, they can be kept relatively the same. Forward and Back throw are rarely strong enough to kill, but have good knockback to reset the neutral, and up throw and down throw are good for starting combos, both with decent damage. Her grab, however, should be no slower than that of Mario’s. Mario possesses greater speed, damage, and combos than Zelda ever had, and does not pay the price imposed on Zelda. As we have discussed, Zelda is slower than a majority of the cast. If she has a grab that is only slightly longer than most characters, at the cost of many frames of startup, she will rarely be able to grab at all. Characters with faster grabs often have better dash speeds as well. Knowing this, it will be virtually impossible for Zelda to get a grab without a standard number of frames of startup. These changes to her ground game make her more interactive and less like a “wall” to fight against. Once one can get past her defenses and zoning, one should be able to kill her easily due to her low weight. This is the standard shortcoming of long range fighters, and translates well for Zelda.

However, Zelda is also supposed to possess moves with powerful knockback to compensate. Therefore, her strongest kill moves should appear in the form of her aerial attacks. However, Zelda should not be completely defenseless in the air. To make her useless in one space of the game would be counterintuitive, since we already established that she should have an interactive (rather than advantageous or one-sided) ranged game with her opponent. Her neutral aerial is a perfect combo tool, which can be left as is so long as it functions properly. This is where the Lightning kicks come in. Zelda’s lightning kicks should retain their virtues of speed and power, but they ought to cover more space in order to adequately threaten opponents who would seek to approach recklessly or juggle her unhindered. Lightning kicks, if they are to accomplish this control of space, must have a larger sweet spot that lasts longer than two frames. As a result, they will need to be slightly weaker. Her up aerial and Down aerial are well designed, but suffer due to their startup and Zelda’s low airspeed. In order to provide this threatening aspect to Zelda’s lightning kicks and other killing aerials, she must be able to maneuver in the air properly so that she can land them. By increasing her airspeed and air traction, if even slightly, the lightning kicks become much more potent. Since Zelda has a low falling speed, she can occupy airspace for longer than her opponents. While she is there, it makes sense that she should be able to take advantage of her attributes, rather than be punished by them. Overall, changes such as these will turn Zelda into a good fighter at range and a stronger one up close, at the traditionally heavy cost of her speed and survivability. Zelda is often described as a “glass cannon.” With these changes, she will have the cannon to justify her form of glass.

At this point, you must be asking “Why change Zelda at all?” The question is a good one. Why would I write almost nine full pages of analysis and critique based on my experience? For my own benefit? No, I have nothing to gain from knowing these things. One gains little from knowing that his or her favorite character is bad at fighting. I have written this as a logical argument for the improvement of Zelda, not just for myself, but for any person who says to themselves, “I like Zelda, I would love nothing more than to have her as a main in the next Smash!” By now, the answer has been hinted at several times. Smash is a game that reaches a wide range of people, old and young, from all manner of backgrounds. Smash is a game based on bringing together characters from all different genres, cultures, and so on. Though there is a bit of exclusivity with who does eventually get picked, but overall, Smash is made with inclusion in mind. It is a huge game, that is very important to many people. However, there are some people who like to play as Zelda, and who have chosen to play as this one character for all three games that she has been playable. In all three games, she has remained relatively the same, and received little to no attention to her design in order to improve the experience of these players. Since she has not received the careful modifications granted to characters like Bowser, Yoshi, and Ness, it is safe to say that at some point during development, Zelda went largely ignored, like a similar character, Jigglypuff. When a character is consistently poor across three games where other characters have received buffs and attention, players begin to feel that they are unwanted in a particular game. These players are given inferior tools for wanting to play a particular character, and perceive this as a slight to themselves. The connection between player and character is one that can be traced back to the connection between humanity and archetype. The individuals who find a connection to Zelda are one fraction of a larger audience, the entire community of individuals who purchase and play the Smash games. By providing Zelda players with a weak version of their favorite character over and over again, Nintendo risks alienating a sizeable fraction of their audience. This repetition of weak design says to these players, “We do not care enough about you or your character to provide you with the same quality of experience as other players.” That is a slap in the face to some of Nintendo’s most stalwart supporters.

We love Zelda, we love Nintendo, and I do hope this will find its way into the right hands.

Make Zelda viable.

With best regards,

Dominic “Prynne” Behler

EDIT 11/28/18: I completely forgot when originally posting this to thank otheusrex otheusrex for assisting in the editing process, many years ago. You’re the best, Elf!
Last edited:


Smash Cadet
Aug 9, 2014
And while you're at it, please return transformations. I mean, several elements that didn't make it to 3DS and Wii U due to the former's limitations are now making a return appearance, like Pokemon Trainer and the Ice Climbers, so why not return transformations? By now, it should be clear that the Nintendo Switch is perfectly able to handle all these things without a problem, just like the Gamecube and Wii were, right? There's great potential right there, so why miss it?
Last edited:


Smash Rookie
Dec 14, 2014
And while you're at it, please return transformations. I mean, several elements that didn't make it to 3DS and Wii U due to the former's limitations are now making a return appearance, like Pokemon Trainer and the Ice Climbers, so why not return transformations? By now, it should be clear that the Nintendo Switch is perfectly able to handle all these things without a problem, just like the Gamecube and Wii were, right? There's great potential right there, so why miss it?
This is perhaps my biggest gripe about the game. If a character like Ryu has the ability to change between weak and strong attacks based on how long you hold down the button, there's no excuse to keep away the transformations. Tap down-b for the transform, hold down-b for the shadow knight thing (from the worst mainstream Zelda game ever). The timing-dictates-the-move can also be applied to Sheik's Floppy Fish as well as avoiding the loss of Charizard's rock counter (which annoys me and I won't miss it, but I'd rather see a system that gains better transforms than loses the rock).

And obviously, there are other practical ways to implement it. Bring back transform.
Last edited:

L2 Sentinel

Smash Apprentice
Jan 24, 2014
If I may offer a dissenting opinion: I absolutely love that Zelda no longer transforms. I have no interest in playing Sheik, so the transformation, to me, means I have one less special to use as Zelda. And I'm sick of the "Zelda's best move is Down B" memes. I want her to step out of Sheik's shadow. Lastly, Sheik doesn't appear in ALttP or TP, so it makes no sense for those Zeldas to transform.
Apr 9, 2014
Behind you.
Nice read but Zelda did some fighting in Twilight Princess, The Wind Waker, STracks, also a tiny bit of magic in OOT and MC
Last edited:
Nov 29, 2013
Unfortunately there was no direct line that I could send it in through. I settled for leaving it open for all to read and posted it on Reddit as well.
I have heard stories of people sending such letters through tech support and actually getting an appropriate response. It may be worth a shot, as it is an issue of dissatisfaction that has plagued a character for years.