What the heck is depth anyway? OR Is Melee deeper than Brawl?
by KishPrime
----------------------------------------

So I totally wrote more than I should have, but it was fun. I'll take suggestions or ideas if you have any before I post it in the main room. I'm also interested in responses.




What do people mean when they talk about the lack of depth in Brawl, what exactly are they talking about? It seems fairly obvious that many people are on a different page. So here�fs one explanation (and unfortunately, not even a thorough one) of depth before we even get into Melee and Brawl. Feel free to skip through if you get bored.

Let�fs go all the way back to the definition of a game. In a game, you have (in this case) two or more players. Both players have an objective, and there are usually many ways to reach that objective. In this case, depth is directly related to the number of effective branches that play can take once it begins. This is directly related to the number of options that each player is allowed to employ. As continuous choices are made, more options present themselves. If you can use your imagination, this creates a branching tree. A depth tree may grow horizontally or vertically. Horizontal depth provides numerous choices at every intersection. Vertical depth occurs when even the earliest decisions in the game continue to have an effect throughout. Both are important to depth. The depth tree grows with every move, but the size of a depth tree is limited at the top. Once an action no longer has an impact on the game, it falls off the tree. Thus, a tree is limited in its growth by the resonating effect of past choices.

Let me give two good examples with competitive depth that exhibit branches. Rock-paper-scissors is itself a display that shows that there need not be many options, so long as they are all effective options. Say that I first throw paper, and you throw rock. Because of the nature of the game, I now know that you threw rock, and because of that, you may or may not choose to throw rock next. I choose paper, believing you to think that I would not throw paper again. I am proved correct, and win a second round. Then it falls on me to choose again, will my opponent continue to throw rock, or will he predict this? This is a game with a truly huge number of paths through which play can travel, but it is limited by the fact that all three options are inherently equal. Starting with rock has the same net effect as starting with paper. Thus, the horizontal depth is limited. Three choices, no more. The vertical depth is also limited to the fourth layer of yomi, after which they begin to repeat. Refer to Sirlin if you have no understanding of yomi.

What happens if options are unbalanced. Say that you are playing rock-paper-scissors, but rock is worth three points on a win, scissors two, and paper one. The goal has not changed, but the route to getting there does. I�fm going to play this game with the hope of scoring a couple rock wins, so I throw rock the first three times. My opponent senses this, and plays paper all three times. I have a decision to make. I am down 3-0, but I am not necessarily defeated. After all, I can play scissors next turn. If my opponent truly thinks I want the three points, I will instead earn two and sneak close to the victory, also confusing my opponent. It is just as valid of a game, and adds in a risk/reward function. Do I go for three points, knowing that my opponent knows that I want those three points? Do I anticipate his prediction and go for two points? Do I play it safe and try to win one point? Again, from the start of the game we have countless paths that we can take, but there are exponentially more paths than In regular rock-paper-scissors, because each of them now has a different �gvalue.�h Throwing rock is not the same as paper. In terms of horizontal tree size, there are still three options, but past choices have a much, much greater effect because the yomi become about pattern recognition due to the different values rather than simple choices. It is technically a �gdeeper�h game.

Here are a few games with little competitive depth. I�fm going to flip a coin. You get a point if you choose the right side. Two choices. Say I get the first call right, as tails. Still, both choices are equally valid. Every new choice has nothing to do with prior choices, as my odds of getting tails on the second throw are identical. There are no decision-making branches that have any effect on anything else. There is no vertical depth to this tree, and little horizontal depth.

An even better example of this is tic-tac-toe. It actually has 9 choices in the first layer of the tree, and shrinks as it goes. Every choice has an effect on future choices. With 9 horizontal choices, and 9 vertical layers, one would think that this game has significant competitive depth. However, we see that in every situation, one choice is always overpowered, leading to each path through the game being fixed. It�fs not very deep at all, because there is only one true choice horizontally.

This is getting very boring for most of you. Let�fs get back to Smash.

I will say first that the vertical depth of any Smash is mostly not the issue. All actions in a Smash match tend to affect those following, as pattern recognition is a huge part of the yomi of competitive play, though there is one example of vertical depth which I refer to later. Therefore, let us focus on the options that create horizontal depth. Fighting game options are generally in two categories, offense or defense. Let me classify offense as simply the one who moves first and defense the second, as obviously offense can counter offense and that confuses the issue. The main way that a fighting game lacks competitive depth is if there are too few effective offensive or defensive options. The balance of these options is important, though let us remember that imbalanced options are not necessarily bad. It is often fine for there to be superior options as long as they are inherently more difficult and/or risky. An imbalance is not a sign of a broken system, but there must also be a simultaneous rise in risk with every increase in power.

In Melee, grabs and attacks both had good combo potential for offense, and left you equally vulnerable afterward if defended properly. Vulnerability could be minimized, but never eliminated, which is why there was room for mindgames (to apply an overused term, literally to create confusion or paralyze one�fs ability to react) to properly cover oneself. Enhancing the options for offense was the diverse movement system in Melee, including wavedashing and dash dancing, which affected the timing of your attacks. Because defensive options required timing as well, movement combined with grabs and attacks allowed one to neutralize some of the vast number of defensive options.

There were numerous defensive options available, which while hard to execute perfectly, when done correctly would allow for a proper counterattack dealing significant damage or a kill. Different defensive options allowed different counterattacks. Different offensive options would create different amounts of damage. The independent offensive moves also increased the amount of horizontal depth. The offensive player will always try to land his most damaging combo/kill move, the defensive player would always attempt to employ the defense which allowed the greatest counterattack. Both players know this, and try to use that information against each other. The potential punishment was great, which is one of the reasons the game was exciting. By the end of the game's life, every mistake was devastating, which drove the top players to achieve one of the most high-level environments of play for any video game ever.

Brawl, quite simply, has much less horizontal depth. The removal of dash dancing and wavedashing as offensive and defensive options were devastating to the delicate system of options in the game. Let�fs look at the lists. Defensively you still have these options: camp the ledge, run away, jump, air dodge, sidestep, shield, and outprioritize. This might not seem like a bad list, as there are still many choices here. The problem comes with offensive options. Instead of being able to alter the timing with a dash dance or a wavedash, you are restricted to only three options now, which are to poke, attack, or to grab. Two of those options are practically the same. This is IT, at a basic level. Because defensive moves are mostly timing-based, with no way to alter the timing, you have no way to neutralize any of the defensive options. A problem is immediately apparent when you look at these two lists. However, it would not even be that big of a deal if offensive play was able to deal a decisive blow when it in fact landed an attack and defensive play did not allow this option. After all, if offensive play is more risky, then a better reward would still lead some players to choose it even though it was less successful in general.

Sadly, as we all know, this is not the case. The ability to punish is now a moot point, as the ability to do significant damage is drastically reduced by Brawl�fs hitstun system. With both offensive and defensive options leading to roughly the same result, the painful end is that players will most often pick the side that wins more often, and logically this will be the side with more effective options. This limits the options to defense and reduces horizontal depth further. After all, if all options are good ones, and you are picking between two sides to be on, one will usually pick the side with more options. This is why defensive play wins out the majority of the time, and to make matters worse, it is now possible in more situations to camp for some kind of infinite on the counterattack. The depth of the core is removed, and when the depth of the core is removed, offensive options become extremely limited to only those moves that are safe, and defensive options are abused. As the number of choices (usable moves) plummets, so also does the depth.

There is one example of vertical depth that is stifled in Brawl that seems obvious to me. As perhaps the worst of its flaws as a competitive game is the simple fact that mistakes are simply not punished. One or two hits as an attack or counterattack does virtually nothing to a player in Brawl, and quite often after hits are exchanged the match is simply reset to the beginning conditions, two players standing and looking at each other (even if it doesn�ft last long). The vertical depth is then cut off, and the match begins roughly anew with only the yomi layers present. Even if you have someone in the air, the defensive player still has too many options available that can take him to a reset position. He can jump, run (or move) away, air dodge, fast fall, or outprioritize. With this many options available, it is a virtual reset to initial conditions, and yomi layers are again all that remain. Yes, the offensive player has gained a slight advantage, but he is primarily focused on yomi as a means of continuing his assault, as the fact that he just hit his opponent has little effect on the mechanics of the game.

Let�fs get away from the strategic depth in the game for a second to look at another important part of any video game. In all games, there are what I would like to label as physical levels of depth. In board games such as Chess and Go, one�fs ability to manage your energy level and mental stress would be considered physical depth. Obviously, losing these battles will cost you in the game, even though they occur outside the game. Yet they are indeed things to be mastered. Video games have even more levels than the two mentioned, such as a player�fs reaction time and finger speed.

So let�fs look at the physical depth in Brawl. The overall speed of Melee added much physical depth. In Brawl, the speed was reduced, more or less reducing the physical depth as well. L-canceling was a strict physical skill, and though I have seen many people argue that it was unnecessary, I choose to disagree, because I value these physical levels of skill as well. These were levels on which good players could differentiate themselves from great players. I never did have the fingers to play Falco. In the end, Brawl limited the depth of potential achievement. A �gperfect game�h physically in Brawl is, simply put, much easier in nearly every way than in Melee (except for a few techniques). So while this does not technically discuss the depth in the game itself, it is an important part of measuring how far there is to grow.

Now, what might be wrong with my argument? Unique characters! They mess this whole thing up really good, because their offensive and defensive properties are all different. You might hope that they serve to balance the mechanics. But in reality, they merely have served to hide the lack of depth in the game mechanics as discussed in the article. Characters with overpowered attacks or better speed lead to the impression that offense and defense might be balanced, where the real story is that characters with legitimately balanced attacks are really suffering or that some characters' defensive properties are simply weak. Individual character properties for defensive options further murk up the waters, as some characters have horrible sidesteps, rolls, air dodges, etc.

It is very hard to get through this appearance of balance. Still, at its core, we can see that the game has some mechanics that are severely lacking. The options of equal value prove perilous when one side dominates the number of options. So do I hate Brawl? No, not at all. But I do see it as competitively inferior to Melee. Horizontal depth is more limited, vertical depth is more limited, and even physical depth is more limited. With this all being the case, it is hard for me to say anything else.



Side topic...
On a related note, balance can be defined in many different ways. In fact, move or character balance is near irrelevant to the question. The example in the initial paragraph perfectly illustrates the way that imbalance can actually be part of the yomi layers. I know that you know, etc. I want to land my most powerful move, you know that I want to land said move, etc. Move balance serves only to preserve or eliminate depth from the core of the game. If moves are imbalanced in their attack/defense ratios, what really happens is that they eliminate options, which removes depth. Character balance cannot fix a broken system, though I believe that is the argument players make with MVC2 on occasion (I know nothing of this, I have just heard many a player say that the only reason MVC2 works is because there are multiple broken characters).

Ultimately when we talk about balance with regard to depth, we want each option to be balanced in terms of risk/reward, which is far different from our standard definition of balance. Every choice made in the game has this risk/reward premise. If the opponent perfectly times some defense of my move that must be frame perfect, then I could get grabbed to death. If he fails at his defense, I will kill him. This is an example of balance. There are intermediate factors. Perhaps the defender chooses a safer, easier option, which avoids the danger, but allows for a lesser counterattack. Perhaps the attacker is leading and sees no reason to risk a stock, and so does a safer poke. Perhaps the attacker employs a mindgame, expecting that the defender will defend the wrong attack, leading to a safer or more effective counterattack. All of these things are great examples of a balanced scenario. In Melee and Brawl, with a million different attacks and ways to string them together, every one of those scenarios is tweaked for every move.


There are a lot of random thoughts in there. It is not one of my best articles...sigh.