The Neutral Game and Mix-ups
The flow of a Melee match is usually described using 2 phrases: the neutral game and the punish game. The neutral game is when neither player has a clear advantage. Each player tries to get an opening on their opponent, and when they finally do, the match transitions to the punish game, where the player who finds the opening tries to maximize their damage before allowing their opponent to return to the neutral game. These are both very broad terms and not every situation in Melee falls neatly into these 2 categories, but many situations can be simplified down to them. In this post, I want to explore the neutral game at a basic level and explain how I approach thinking about it.
When I talk to a newer player who is struggling in the neutral game in a certain matchup, the conversation usually goes something like this:
(Let’s assume we’re talking about the Sheik vs Marth matchup)
Them: “Man, I can’t ever hit Marth in neutral. He just dash dance grabs me all the time when I try to approach!”
Me: “If he’s retreating with dash dances, you need to aim behind him. Try doing a deep dash attack behind where you’re normally trying to attack.”
Them: “If I do that he’ll just shield it and punish me!”
Me: “If you think he’s going to shield, you can grab instead of dash attack.”
Them: “If he knows I’m going to grab, he can just down-tilt me before I get to him.”
Me: “You can short hop over a down-tilt and forward-air him.”
Them: “Then I’ll get dash-danced grabbed!”
From that player’s perspective, they assume that every option they pick will be instantly countered by their opponent, leaving them with no reasonable options. This is often an illusion when you play against a better player who can pick up on your habits quickly. The truth though is that options in the neutral game and their counters are rarely ever guaranteed. The neutral game is composed almost entirely of mix-ups. I define a mix-up as a situation where a player cannot reasonably react to the different options their opponent has available. At its simplest, a mix-up is just a guessing game. Mix-ups are a core concept across all fighting games, including every iteration of Street Fighter as well as Smash. Selecting an option in the neutral game often requires commitment without a guarantee of the option succeeding. Sometimes you can be reasonably confident that the option you pick will be a good one, but unless you can read your opponent’s mind, there will always be a degree of uncertainty. The need to constantly change your patterns to remain unpredictable while simultaneously adapting to your opponent’s tendencies is what makes fighting games such an exciting and rewarding mental battle.
The underlying concept to this mix-up based neutral game that I’m describing is selecting options that are too fast for humans to reasonably react to. The average human reaction time is often reported as around 1/4 of a second, which equates to roughly 15 frames. Many close range options in Melee are on the order of single digit frames, which is well below this figure. Even though 15 frames is the average reaction time, it’s nearly impossible to react that quickly to something that you don’t expect, so in many situations slower options can work if they catch your opponent off guard. Reaction time also varies greatly from person to person, so some players may be more susceptible to slower options. If you had inhuman reaction times and could react perfectly to all of your opponent’s mix-ups, fighting games would quickly become degenerate, as demonstrated by TAS (tool assisted) bots. As an aside, this is why I always say that fighting games are an uninteresting and easy genre of games for AIs to solve (this has been a popular topic lately with the high-profile Google DeepMind AlphaGo matches).
In contrast to the mix-up heavy neutral game, the pursuit of a perfectly optimized punish game is a popular trend in the Melee metagame nowadays. Certain players have become known for their streamlined, optimized punish games that convert nearly every hit into a kill by reacting to all the opponent’s defensive options with the best option to prolong the punish. With the advent of the 20xx Hack Pack that allows newer players to practice optimizing their punish game in this manner, we see a trend of players entering the competitive scene who can brutally convert off a single opening, but struggle to land such an opening against players with stronger neutral games. I believe the root of the problem is that these players try to approach the neutral game with the same mindset as the punish game and try to find some elusive “optimal” neutral game option that doesn’t exist. Unless you’re playing a highly degenerate character matchup or your opponent is literally incapable of adapting, there is never a single neutral game option that will win you every exchange. The neutral game requires a fundamentally different approach.
The uncertainty of mix-up situations is what turns options that seem to be terribly unsafe in isolation into strong neutral game options. For example, Sheik’s dash attack can lead to big punishes, but it is one of the worst things you can do if your opponent is shielding. If you represent dash attack as one of your neutral game options often enough, however, your opponent will start to be scared of it and may shield more in an attempt to counter it, which you can then counter with a dash grab. Even if you dash attack and end up getting shield grabbed for it a single time, your opponent now has to think twice the next time they stand in your dash attack range. Since your opponent knows that you are capable of using dash attack, the threat of it alone can shift their behavior. If you never, ever dash attack, your opponent may realize that and won’t bother shielding when they are in your dash attack range, which now makes it harder for you to land a dash grab on them. To be able to make this work effectively, you need to pay attention to your opponent’s habits. The frequency at which you can dash attack safely will vary widely depending on who you’re playing against. This need to constantly adjust is what makes the neutral game difficult because there is never a “right” answer.
Since Street Fighter is another fighting game that many Melee players are familiar with, I’ll give a Street Fighter specific example. Newer players are often told that wake-up Dragon Punch is unsafe and a terrible idea. If you never wake-up Dragon Punch though, your opponent is free to pressure their knockdowns with meaty normals as much as they want. To prevent this, you need to represent your wake-up Dragon Punch as an option often enough to make them think twice about sticking out that meaty jab after they knock you down. Sometimes you will guess wrong and get punished, but now every time your opponent tries to capitalize off a knockdown, they’ll second-guess themselves because they know you’re capable of wake-up Dragon Punch.
To start developing the foundations for a solid neutral game, I suggest focusing on a few broad, simple options and figuring out how different opponents counter them. If you try to incorporate every single possible option you can think of at once, you’ll easily be overwhelmed. As a very simple example, with Sheik, you can try approaching a Marth with only either dash attack or dash grab. As you play better players who have solid counters to these options, pay attention to what they do. If that counter can’t be beat by your current arsenal of options (for example, a pre-emptive down-tilt from Marth beats both dash attack and dash grab), figure out what the counter is to their counter (shorthop forward-air from Sheik generally beats Marth’s down-tilt) and work it into your neutral game. Gradually build upon your neutral game options by adding in broad options, and as you become more and more experienced, you can begin to incorporate more subtle variations of these broad options to keep your opponents guessing even more. Watching videos of other players is a great way to discover options you can try. This method of slowly building up neutral game options is something I’m going through right now as a new and inexperienced Fox player.
After reading all of this, I hope it’s obvious now why asking a player “What do I do in the neutral game vs <insert character here>?” is an extremely hard to answer, and frankly, a very terrible question. It’s often much more productive to identify situations that you have trouble with and ask specific questions like “What can I do to counter Sheik’s dash attack?” Learning to ask good and focused questions will help you tremendously when you’re trying to gain knowledge from more experienced players.
In summary, this post was about how I personally think about and approach learning the neutral game in Melee by breaking it down into simple mix-up situations. One of the main reasons I started this blog was to find out how other players think differently from how I do, so feel free to start some discussion with me!