This is part 3 of our 6 part in-depth series devoted to exploring Smash's neutral game.
Part 1: Intro/Options and Positioning
Part 2: Endgame First
Part 3: Two Types of Approaches (You are here)
Part 4: Relative Frame Advantage
Part 5: Reads
Part 6: MangoPuff - Profiting from Risk
Part 3: Two Types of Approaches
Attacking playstyles are exciting to watch and gratifying to play with. Especially in the context of a 1v1 fighting game, there’s a very special feeling that accompanies enforcing your will on the opponent and discovering cracks in their defenses just as quickly as they can get them up. This is especially true in such a high-octane game as Melee.
Despite its intuitive appeal, approaching is commonly misunderstood on a conceptual level. You cannot simply press forward, do something pseudo-random, and hope not to be countered or adapted to. For aggressive approaching to be a valid strategy it must be designed for long-term success against a strong opponent. This means that the attack itself and the timing of the attack cannot be predictable or easily countered. However, it also means that the reward for successful approaches has to outweigh the punishes for unsuccessful punishes. You can see this play out over Genesis 2 Grand Finals. Despite the frequency with which Mango successfully landed an approach, Armada’s lengthy punishes on mistakes were more valuable.
Reaction Time and Unreactable Range
This being said, there’s a two-part logic that can be used to conceptualize and optimize approaches that maintains effective in-game and psychological aggression. Before getting to what the two types of approaches are, we have to understand the biggest difference between Melee and Chess: Reaction time. If humans had sub-frame reaction time, it would be pretty easy to optimize approaches. Melee would turn into a very very weird game with strange and unintuitive priorities like port-priority. Basically 20XX. In reality, though, humans have a natural lag in their response time caused by the need for the brain to process information. Reaction time can be improved with practice, but there is a genetic limit.
Now we know that there is a timing consideration that must be taken into account when making plans. Because there is a limit to how quickly we can react to things it is frequently impossible to cover every option on reaction and a certain amount of guesswork enters into our gameplan. But because different options have different risks and different rewards, optimizing takes some amount of game theory. I say this not to intimidate but to point out just how tricky unreactable problems are.
The relevant (and fun) bit for us here is applying this information about our timing limitations as a spacing consideration. Let’s say that your personal reaction time is 20 frames. That means that if you stand precisely outside where your opponent can have a hitbox out after 20 frames then you are safe. The timing issue is actually a spacing issue! Everything outside of that range can be flowcharted and executed on reaction. Provided that your reactions are consistent your success will be exactly as good as your flowchart. In contrast everything within that range depends not only on the strength of your flowcharts, but also on how good you are at guessing.
Now that we’ve established the sort of invisible line that we’ll call the Unreactable Range, two types of approaches emerge naturally: the Incremental Approach and the Unreactable Mixup.
In this approach, you aggressively move forward to the very edge of the unreactable range then stop. Because for the duration of our approach you maintain your capacity to react to anything your opponent does, it is completely safe. However, because you are rapidly closing space, this movement is still hyper-aggressive and often baits a pre-mature response. Your goal is then to punish their impulsive response on reaction. If they don’t commit to anything punishable you have still gained ample stage control. The incremental approach is most difficult when your opponent decides to run at you at the same time. When this happens it is more difficult (but still reasonable) to stop or to dash dance at the correct spot.
Note the very precise and consistent location where mango shields/starts his jump. Mango will often flash shield to stop in place when an incremental approach is his intention.
0:40, 1:18, 3:10, 3:41, 7:48, etc.
In this approach, you aggresively move into the unreactable range and initiate a mixup. An unreactable mixup should be specifically designed to cover options. For example, Fox can mix up between running shine (beats shield, CC and other in place options), deep nair (beats a WD/dash away/jump) and a WD back (beats an attack). A Fox player that actively chooses between these three options can beat most of his opponent’s options. He can alter this set of mixups to account for anything outside as the matchup demands, but this is the rough design. It’s as close as melee gets to rock paper scissors. The player initiating the mixup is either a) confident that he can consistently out-guess and out-punish his opponent or b) probably shouldn’t be initiating the mixup.
At 1:14, 1:32, 1:40, 2:21, 2:35, 2:52, etc.
Just as the Unreactable Mixup requires the Fox player to mix up between his options to avoid being predictable, approaching at all requires him to mix up between Incremental Approaches and Unreactable Mixups. A player that never enters into the unreactable zone does not represent a real threat when moving forward. A player that never stops short is not abusing their stage control and will quickly lose to a player with a strong defensive game and good risk-reward from the corner. In order for either type to maintain effectiveness against a strong opponent, the threat of the other must always be present. Every option that isn’t covered is an exploitable hole in your strategy. But as long as these threats are sustained and executed well, approaching is not just intense and thrilling, it’s a winning strategy.
Next time we will look at how the neutral game is influenced by the concept of relative frame advantage and how we use it to make unsafe options safe.