Plateaus exist at all levels of play in competitive gaming. Rather than lack of practice keeping players from improving, it is usually a lack of knowledge about how to practice intelligently. MIOM has had some articles about how to improve, and I hope to add to those by going through a basic flowchart of how to advance in Melee. The optimal way to improve is to learn things in an order based both on ease and usefulness. Though the list is stratified, the transition of focus is gradual. As you start to master one step, you will slowly shift your focus more and more to the next, until everything you learned from the previous steps is subconscious.
Though I feel strongly that this sequence is the best way to improve, it’s important to remember that during each of these steps, there are other things to practice. Your available practice partners will impact which things you can practice effectively. If you don’t have other people to practice with, coming up with new technology or highly situational techniques may be the only thing you can practice. Or maybe you just find solo practice to be a ton of fun. There are reasons to practice contrary to how I suggest. You as an individual will have your own weaknesses that will require different training. Spacing is a good example of something that, despite being deficient in many players, does not require specific attention. Someone could actually have good or bad spacing at any step. Though it may be the most important and fundamental aspect of the game, people normally pick it up over time without having to dedicate attention to it. Conversely, bad habits are something that are not shed the same way that bad spacing is. You should continually evaluate your habits at every step. It can be something simple, such as “My wavedashes are always shorter in game than in practice” or “I jump immediately whenever I get hit,” or something as complicated as “I always get mindgamed when Marth crosses up my shield, do something stupid out of shield, and get punished by pivot fsmash.” Old habits die hard, and this as true in Melee as any other game. Muscle memory is essential for getting your SHFFLs down, but autopilot is the enemy of mindful improvement, and thus muscle memory is often the biggest hindrance of growth.
It’s been said time and time again, but the first thing any new player should practice is tech skill. This starts with short hops, SHFFLs, and wavedashes. Once these are out of the way, more character specific technology awaits. Samus players might choose to practice Up+B OOS. Falcos need to learn reverse lasers. Falcons should master pivots. Different characters have different tech skill requirements, but no character is free from requiring tech skill. This phase is about learning basic tech skill and does not touch the truly difficult stuff, like haxdashes or invincible waveland usmashes. Nonetheless, to a new player, it would be supremely impressive to get past this phase in even in a few months. Experienced players often forget how hard it is to have any control over your character when starting out. Keep in mind that mastering a technique in solo practice is only the first step. After this, you have to learn to do it under pressure, and then you have to learn how to apply it. The application of any given thing is generally learned just by spamming it and then figuring out which situations it’s useful in through trial and error. You will start to have an idea of when wavedashing and some other techniques are good, but there’s no way to come close to understanding the application of everything at this step.
The next thing to focus on is a punish game. Though the neutral game is arguably more important, the punish game is easier to focus on. Neutral game improvement involves understanding of mindgames, positional advantages, etc. Knowledge of most of these things actually comes quite naturally with experience. The punish game is something that, like tech skill, is largely contingent on your own execution and not on interaction with the opponent. Waveshine usmash, stomp knee, chain grabs, frame perfect aerials, etc. can be practiced just by focusing on yourself during a match. Master your own hitboxes, learn your frame data, and learn to be precise with your character. This is the time when good players can most easily give good advice, so you should solicit it as soon as you feel your tech skill is solid. Bad move selection is a glaringly obvious deficiency in most new players, and it’s pretty easy to say “You should have usmashed there” or “You need to learn to use your invincibility to ledge guard that.” Most people plateau during this phase because they don’t give it adequate focus–there is a vast amount of progress to be made in this step. Improving your control and punish game is not only one of the easiest ways to improve but also one of the most gratifying. There are few other times when your improvements will be so immediately apparent both on the screen and in your results. When starting Melee, the first year, or maybe even two if you aren’t dedicating much time to it, is best spent learning to have perfect control over your own character.
The third step is DI. Although one may think that DI is a simple input and could be learned earlier, DI takes focus on the opponent. You have to watch them. Really, if you are a competent player, you should actually be watching your opponent all the time and never focusing on yourself. But this takes too much concentration to practice for a new player, and focusing on it sooner will impede improvement of other things that are more easily practicable. Moreover, after you start to notice and understand DI, this will help expand your repertoire of punishes. Many new players go for their killing moves in the first possible setup. Experienced players will do autocombos or bread and butter extensions into their killing moves. Great players will trick their opponents DI to artificially extend combos (PewPewU and S2J are two players who really seem to know when to go for their finisher and when to extend combos). But for any new player, it starts first with auto combos, then with standard but not automatic combos, and then somewhere down the line DI mixups come in. As you learn to watch your opponent and not yourself, you will start to gain some insight into the neutral game, though this knowledge may be subconscious. In order to have good DI, you need to understand when your opponent wants to attack.
The fourth step is the neutral game. This is when you are actually starting to play Melee in a way comparable to respected players. This is when you learn everything about every character. You learn their ranges, their recovery options, their frame data, their positional weaknesses, etc. It’s time to start theorycrafting and to truly understand the utility of all your options. You’ve probably been watching videos for your whole Smash career, but now is when you really learn to analyze them. Learn to understand why players choose to do what they do, not just admire how sick their combos are. Watch videos with analysis by top players and make sure you understand everything they say. You will now get hype over things that you didn’t notice before, and things you got hype over before will now seem standard. Though we have amazing resources like smash labs, where Melee’s elite have shared their insights going into their thought process, I find myself now thinking that the neutral game has some ineffable quality to it that makes it difficult for me to give useful advice at this stage. Once you have mastered control over your own character, and you have learned to watch the other player, things will start to click.
The beauty of the game doesn’t really shine through until you are at this stage. There are a few things that come to my mind that puzzled me for years. I always heard about stage control and about momentum of the match. Yet it was difficult to see how control of the match influences decisions about when to attack and when to evade. The lack of understanding of these concepts is obvious in new players because of how frequently throw out attacks from defensive situations. As your ability to understand other characters blossoms so will your ability to mindgame other people. After hearing “rolling is bad” for years, step 4 is when you should finally have enough situational understanding to bait and punish them. Once you understand the options a character has, the brilliance behind the decisions of top players becomes clear.
The last stage that I feel qualified to talk about is innovation. If you are at this step, you have watched many videos and you understand your own limitations. You are making few bad decisions when you play, and when you do make one, you either understand and regret it or solicit advice to deal with the situation. From step 2 onward, you have been expanding your bag of tricks. The movement options you practiced are now clear in their application. You don’t drop combos anymore and you have all the expected skills of a competitive player. Now you have beaten some respectable players, but you probably still aren’t at a level where you are a serious threat to your regional bosses. Your own style will have started to emerge in step 2, but this is where you harness your own strengths and introspect to search for small optimizations. As an aspiring player, you won’t be able to beat people better than you just by copying what they do. I believe it was Prog that spoke in the documentary about how no one can truly teach you Melee. They can show you 90% but this is that last 10%. Polish everything and incorporate every new thing into your game that you can. Deign to have truly difficult tech skill as consistent as easy tech skill. Make sure the most useful things are the things you dedicate the most time to be consistent with. This goes double for spacie players. If you aren’t beating good players yet, your multishines aren’t going to change that. It’s a waste of time to spam multishines until you are deep in step 4. Even then, remember that Melee has an absurd amount of borderline useful tech. Multishines are cool but SDI is way better.
Until you are in step 4, you aren’t really in a position to innovate. You can go do all the solo practice you want, and you can learn or invent some new technology, but you will almost certainly not be able to apply it in a match. The necessity for new movement options or any new technology is what will drive you to implement new things in your game. You need a deep understanding of positioning and situations that occur in game in order to implement new things. Step 5 is something that helps you break plateaus that you reach in step 4. It’s how you get new options to get out of sticky situations and how you trick other people with options that they didn’t expect. Remember that, in a close tournament set, if you use something once and it gets you one kill, it was worthwhile (and likely crucial).
These final two steps are the most blurred in their lines. No one ever truly masters the neutral game because it isn’t truly masterable. There are always mind games and new styles of play. Melee is a game with beautiful depth, and if you practice mindfully, your rate of improvement should generally increase as you get better. It’s like practicing violin–it’s going to take years to make anything worth listening to, but after some years, you are able to pick up new compositions rapidly and express yourself more fully. Learning matchups is terribly difficult if you are still in step 2, but it’s something you will pick up rapidly if you are in step 5. I would go as far as to say that, with the right partner, a matchup could go from being one of your worst to one of your best in a day. Most players face plateaus throughout their development, but it’s just because they don’t have direction. Now, with all the resources ranging from MIOM to Smashboards to Sirlin, there is a massive amount of content to help direct your practice.
As a final note, I’d like to say that, for not only the sake of fun, but also for your growth as a player, learning teams should happen during steps two and three. If you can precisely place attacks, you can play teams. DI is easier in teams because holding in is good most of the time. Rolling and bad tech habits are less problematic. At this level of play, you will also get a lot more out of playing continuously and practicing control over your character than you will losing every game you play in a four man rotation. The intricate dance of the singles neutral game can be entirely circumvented. All you need is to play your role. If you are a new player but have a partner who is good and with whom you have good synergy, you can definitely beat two great singles players who lack synergy. As a low level player, your chances of beating a good player or even a pair of good players is immensely higher in teams than in singles. I’m sure new players would love to have the accomplishment of beating two good singles players with their buddy in teams, and there is no reason why you shouldn’t be able to do this. Though the skillset is different in doubles, it’s actually much easier to do your role and be a good teams player than it is to learn the subtleties of the singles neutral game and its spacing.