At some point in every game developer's career, they are legally required to write an essay explaining why Super Metroid is the gold standard of game design, because when a game is so groundbreaking that it names an entire genre, it deserves to be examined. While my previous article took a look at Final Fantasy VI and pointed out a long list of game design "don'ts", this one is going to tackle the "do's". But instead of following same format as the FF6 article, I'm going to start out with a single point - the thesis of this piece, if you will - and then I'll spend the rest of my time coming back around to it.
"People like playing games, but they don't like being reminded that the game was designed by someone."
Now, that's going to sound kind of hypocritical coming from me given how much Brave New World leans on the fourth wall, but there's a huge difference between the initial impact of that statement and what I'm actually talking about here. Self-awareness in games is not a bad thing - the greatest thing about Hideo Kojima (or, for a more contemporary example, Toby Fox) is his unrivaled ability to remain one step ahead of the player, anticipate their actions, and react accordingly. Although my point does concern the narrative of your game as much as it does the mechanics, the intended takeaway is that it should always be the player initiating the action rather than the game itself.
Of course, a game does need to give the player some direction or else you end up either lost or, in the worst case scenario, with what Penny Arcade once referred to as a "quicksand box". The goal is to direct players without making them aware of the fact. Playing a game is a personal experience, and so players are naturally inclined to want to make the game their own. Being a part of someone else's vision places you in a box with restrictions and limitations, so it's of great importance that those limitations are kept hidden from the player at all costs. It's because of this that I prefer to avoid "artificial" boundaries wherever possible, such as an invisible wall at the edge of a map or a game that's designed for the player to hit a hard level cap instead of a "soft" one. Anything that highlights a limitation in what the game is or has to offer will pull a player out of the experience just as much (and arguably more) than a bad joke will, and the best-designed games are the ones that make you feel like no matter how much you've discovered, there's always something more just out of reach.
So, back to Metroid. The original game pioneered the ideas of open-world exploration, upgrades that open up new areas in addition to making you more powerful, and backtracking through old areas to access the new ones. It dropped the ball quite a bit due to the hardware limitations of the time - the entire game was a whopping 128 kilobytes - and the fact that it was the first of its kind. There were only a few upgrades that allowed further progression and so a lot of the world opened up very quickly, leading players to get lost more easily. This was compounded by the repetitive map design that, again, existed due to hardware limitations. Damningly, the game failed to telegraph its secrets well and made the fatal error of gating progression (which will be discussed just below) behind one such secret.
(Fun side note: did you know that the origin of Samus's iconic "morph ball" is that the developers of the original game couldn't be assed to come up with a "crawling" animation that wouldn't set the NES processor on fire? Goes to show you how good ideas can be born out of limitations, even if that limitation is laziness.)
See, there's a difference between figuring something out from a clue and being given no clues whatsoever and stumbling across the answer by, as Ben "Yahtzee" Croshaw once put it, "...carting a truckload of miscellaneous knick knacks around, patiently rubbing them all one by one against everything else in the hope of hopping on to the train of logic unique to the game's designer." One gives a sense of accomplishment, and the other is literally brute-forcing your way through every possible option until you land on the right one. Super Metroid made an important compromise in this regard by adding a map system, which in the hands of lesser developers would ruin the core exploration aspect of the game but was instead used to great effect to hint at the game's secrets rather than laying them bare. Compare how the entrance to Kraid's Lair is discovered in Super Metroid by means of the in-game map system revealing its location, but not how to get there, to the original game requiring you to bomb nondescript tiles at random to find your way into lower Norfair.
One might be critical of Super Metroid's above-mentioned approach for making things a little too easy to figure out, but bear in mind that this occurs very early in the game in what can still be considered to be its "tutorial" segment. Later puzzles are more subtle, such as the broken glass tube in Maridia hinting at the player to blow up the other one in order to continue(*). The important thing to note is that the game never flat-out tells you where to go, but rather invites the player to explore the path leading forward through its level design. Where the original game stumbled with this by opening up too much of the game too soon, Super Metroid presents new players with a deceivingly linear path for its first act, only opening up the bulk of its world after it's taught you how to properly explore it.
(*The Mandela Effect shows up here seeing as a lot of people - myself included - recall the game pulling a Hideo Kojima on this one by showing off the solution in its "attract mode", which it doesn't. It does, however, act as a comprehensive tutorial of the game's basic and intermediate mechanics for a new player, which is pretty damn ingenous.)
Of specific note in the above paragraph is the term "new" players. In the beginning of its second act, Super Metroid teaches you two tricks - not by text, but by literal example - that can be used to break the game. And while there's nothing that kids enjoy more than breaking things, what's extremely important is that at no point were we actually told to do so. The game simply gives you a pair of neat tricks that the astute player will realize can be used to skip the game's "intended" sequence of events. And one of the most beautiful things about Super Metroid is the ambiguity about where the line lies between what its developers intended and what paths the players have since forged for themselves.
Now, contrast the next two games in the series. Fusion looked at how Super Metroid could be broken, scoffed, and then proceeded to railroad you so Goddamn hard that you had to cough up two hundred dollars every time you landed on it. Following massive fan backlash, the following game went to the opposite extreme but continued using the same flawed method of heavy-handed delivery. Rather than taking Super Metroid's subtle approach, Zero Mission beat you over the head with both directions and sequence breaking tools to the point of requiring them to get the best ending - which really begs consideration of whether it even qualifies as sequence breaking if the developers overtly intend for you to do it.
One thing that Zero Mission shoots for with this approach, however, and succeeds in accomplishing is highlighting the "easy to learn, difficult to master" gameplay for which the series is equally known, but rarely lauded. Super Metroid is very approachable to the casual player, but it is one of the most brutally unforgiving games ever when played for score. And that specific word is very important, because the idea of score as a measure of performance in games has been a joke since the early 90's. The modern era has seen "challenge" runs, often of the self-imposed variety, replacing the bragging rights of yore from holding the high score at your local arcade. Upon completing Super Metroid, it presents the player with a screen detailing the percentage of items collected and the time taken to finish, thus providing both the basis for and a means of tangibly rewarding what would otherwise be completely arbitrary challenges. More importantly, it does this without ever explicitly issuing a challenge, thus going back to my point about the difference between the game initiating action versus the player doing so. This is by far the biggest area in which Super Metroid suceeds where Zero Mission fails.
It's worth discussing the purpose of challenge in video games. Something that I've written about before is that many ROM hacks fall into the trap of making challenge their primary - and often only - notable feature. By contrast, Brave New World was developed to focus primarily on the fun of character development while its challenge is simply a means to an end: an obstacle that warrants that development to overcome. A good game is designed as a vehicle to present the player with opportunities to use the tools they're given, while a great game creates a seamless divide between the level of skill with those tools it demands in order to complete it and the level of skill that it inspires in order to master it.
A good (albeit not infallible) litmus test for how well-designed a game is in this particular aspect is to look at how many of its tools, skills, and/or abilities remain useful in its final stages. A proper climax should be a culmination of everything that you've learned in a game up to that point, testing all of your skills and knowledge in more complex and varying ways than before. A typical role-playing game will fail this test miserably: players spend the entire game collecting a plethora of skills and abilities only to spend its closing moments spamming nothing but the strongest ones. The idea of "situational" skills that are only useful when the circumstances allow are often overlooked by developers and players alike, leading to games (RPGs most egregiously) that marginalize them since the ones that aren't situational end up being all that you ever need.
The Metroid series as a whole avoids the above-mentioned issue entirely by designing the game around the core concept of Samus's abilities as a means of both combat and exploration. Going back to the "challenge" aspect mentioned above, it's entirely possible to complete Super Metroid with only a handful of its upgrades, but doing so requires great skill and mastery of the ones you do collect. And even on a casual level, the "fun" of any Metroid game is primarily derived from two sources: exploration and the player's movement. These two things go hand in hand; backtracking through previously-explored areas in order to progress is one of the chief defining characteristics of the Metroidvania genre, which in turn draws extra scrutiny to the means by which the game is traversed. Done poorly, this can be seen as padding meant to artificially increase the game's runtime. Done well, however, a combination of tight player controls and well-crafted level design can turn traversing old terrain with new abilities or from a different perspective into a completely fresh experience.
Another thing that Super Metroid is particularly well-known for is the laconic nature of its player character; aside from a brief introduction at the outset of the game to set the stage, Samus does not speak at all and the entirety of the game's plot is presented through subtext. Now, I'm of the opinion that stories in video games are like panties: they can be interesting, but I'm much more interested in what they're covering up. Super Metroid, for all of its sparsity, managed to present one of gaming history's most memorable plot twists completely inline with its gameplay. Eight years later, Metroid Fusion showed us why Samus should never be allowed to speak. Fast-forward another eight years and Other M showed us that nobody on Nintendo's current writing staff should be allowed to eat with a fork.
In summary, Super Metroid highlights something I believe to be not only a good game design principle, but a valuable life lesson, as well: show, don't tell. People will rise to far greater heights if simply given the tools and encouragement to do so rather than being explicitly shown the way. It was true in 1994, and it's still very true now 25 years later. Case in point, this article that been floating around in my head for awhile now, and it was playing a certain modern game that manages to encapsulate and build upon so much of what made Super Metroid great that inspired me to get off my ass and write it. And what more could an artist possibly hope for if not for their work to inspire others?
(Oh, yeah... groupie sex. Definitely the groupie sex.)