For years, I've been thinking about how some of the stats in game might not be equal to each other. This would result in that the same amount in one stat will somehow be more effective than an amount of equal value in another.

In some games, HP becomes the dominating factor, and many people "build" towards sets with high HP and are able to dominate any other set. Or in other games, everybody will go for speed, and the most builds you will encounter online, would be speed based since it trumps any other kind of build.

Another thing that isn't right is that builds with balanced stats lose easily to anything specialized. Balanced characters should have techniques that would make use of their balanced stats, and allow them the ability to have an equal fight against a specialized character<who wouldn't be able to use such techniques due to their specialization.>

This imbalance of stats I believe is due to many games following the standard number and types of stats found in most RPGs.

If a few new stats are made, it should take away from the centralization that some stats will have.

For example, in the original pokemon, there was HP, Attack, Defense, Special, and Speed.

Those who played competitively, studied the games in depth, and mentioned that most "psychic" type pokemon beat others out. This was because of the special stat which acted as both the attack and defense for special based attacks. It was harder to balance because it was more centralized by doing multiple jobs. This problem was remedied in future games by having the special stat split to mirror its physical counter part, attack. So now there is HP, Attack, Defense, Special Attack, Special Defense, and Speed. Although, now the game isn't perfect, but people still find it fun on the competitive level.

I believe a stat has to have some kind of mirror in order to be balanced. If a stat is doing more jobs than other stats, people will gravitate towards it in competitive gaming in order to win.

Another example is the Armored Core 4 system, which is classified as a simulator<and also happens to be my favorite type, game play wise> The number of stats that manipulate your character is outstanding, and give incredible depth to how it moves and attacks. If some other types of games<non-mecha> played like this, it would be fun, as the ability to go in any direction and actively dodge attacks is rewarding. There is no "auto-tracking" like in other 3D games where your character is always "locked on" and facing the foe. There have been tournaments in the past for this game, and the variety has been staggering. It is probably my favorite thing about this game, as almost nobody has the same machine. To top it off, it is better than a FPS in that everybodies machine, moves, attacks, and defends differently<although the types of defense need as much variety as attacks in this iteration of the game, most forms of defense are speed or armor based.>

I talk about this game because of the notion of speed/dodging having a limit, instead of being infinite like in RPGs, which doesn't make sense as HP has a limit. If dodging had a limit, it would in a way mirror how HP is the limit for defense. Dodging should be an alternative for protection against attacks.

HP/defense or EN/dodge. The more styles of protection, the more interesting I believe. EN would be a bar with a limit similar to HP.

So somebody could have low HP/defense and high EN/dodge, vice versa, or they could have equal HP/DF/EN/Dg.

I also believe attack should have a limit not based on the attack itself like it appears to be in some games.

So in this way, your HP dropping to zero isn't the only way to lose. If your EN<energy> drops to zero, your character becomes tired and because their HP/defense is light as a paper, the next attack defeats them.

What do you think about this? A game like that would greatly intrigue me, although most casual players would be turned off that their character doesn't have an infinite defense stat in dodge. There are players that like these kind of features in games, and there are those that like it simple. Then there are those that adhere to the old ways and methods of RPGs.

Also, here is some gameplay from the Armored Core 4 and FA series. The game engine and gameplay is unlike any other game I've ever played.

Heres a machine made solely for high speed bursts and melee attacks.<For example, other mechs might travel by using a stable, high output, linear boost for their main form of mobility.>


This machine has average agility and speed, but its ability to stay in the air is greater than that of the orange machine, and its style of fighting is mid-range laser assaults.


And here is gameplay against enemies of equal size to your machine.


This simulator game has maybe around 50 stats that affect fighting. Movement alone has things like acceleration, maximum speed, air drag, and many others. It is mostly skill based, as a new player in a high speed machine would have a dodge rate as good as an experienced player in the slowest<yet highest defense> machines.

It might seem that I have gone off-topic, but this game is relevant in that multiple stats enriches the gameplay<at least, if you learn the stats involved.> Somebody new might feel overwhelmed or those not into customization might just want to get into the action already instead of tinkering with some kind of robot.

More kinds of attacks and defense would make the game funner in my opinion, as I illustrated here.
http://s841.photobucket.com/albums/zz337/FragrantX/?action=view&current=OGONITRIS.png

Anyways, what I'm saying is that for games with competitive gameplay, there should be no stat that is greater in value than any others. This might be difficult to achieve in practice, although it shouldn't be impossible. A game that is both fun and balanced should be the goal for any kind of competitive game, whether it is an RPG, simulator, or any other type that allows multi-play.

I've tried to devise many systems on paper for years, trying to solve this problem of imbalance. The results so far have been intriguing, although testing it in my mind and on paper won't cover every possibility. How such a system would turn out will only be revealed once it is played by many people.

Effectively what I get from your post is that the stats are used in such a way that one stat is more useful for "Winning" a game than others, so people gravitate toward beefing up that stat.

A solution I would posit is, should there be other ways of "Winning" in the game that may be more condusive to other "Roles" within the game. Hence benefiting people who are actually playing their respective Role within the game with a Stat that is actually beneficial for them.

I think you cover this aspect from a player "Losing" perspective, but it could also be useful to think of all available outcomes for the player, and for the game in general.

Stats need not be mirrored or have opposites. I see them as ways of distinguishing certain types of skills that may be required in the Role within the game.

Or at least, that's my 5 cents.

Hard one, I dont see any way around it besides aiming for a good balance and then perfecting it through trial and error.


Some facts:

If you design your game to work with very specialized builds, a balanced build will most surely not work. Example: A fighter that gets better hit chance, hp and defence just from str, and has a skill that raises its magic resistance, then there is no point in having the other stats, is there?

Why would he put point in dex to get a better hit chance if str will provide the same?

There is no need for a direct relationship, if you provide a piece of equipment that requires a high str level to be used and it gives hit chance bonus (which should be provided by dex), you basicaly did the same thing in a more subtle way.

I think its very hard to design a RPG with competitive game play, since leveling up is all about overpowering the opponent. Still, a good way to go is to not provide the players ways to compensate lack of stats with items or skills. This way, if the player builds a dumb and clumsy muscle ball, a player with a balanced mage will rip him to shreds easily.

This also opens up many strategic choices for team playing.

Dwarfsoft: Yes, although for competitive games, stats should have a stat that keeps the other in check. For example, the simple accuracy<offensive> vs agility<defensive>. A stat that has no counterpart in the formula of offensive/defensive might be more difficult to balance out, like intellegence. If you wanted to make a counterpart to it, it would be soul if intellegence where useable as an "offensive" stat. Since intellegence is thinking with logic while soul is believing with spirit. For example, some character uses attack "There are no SASQUATCH!", a person with a high soul/faith stat would resist such a statement, and still have enough HP to believe that SASQUATCH exists.

Although, I also believe there are other ways to distinguish certain types of skills required in a role. Say you only had HP, ATT, DEF, INT, and SPD.

If you wanted to be able to use, say a MAGIC attack, then it would use the stats of ATT+INT. If you wanted to use a skill that uses projectiles, the formula would use INT+SPD to determine it's attack power.

Also, I think techniques need to be a more complex in some games. I think "balanced" builds suffer because of how attacks/techniques are formulated. They usually have something like Attack power and accuracy as their attributes. If they had something like Attack power, acc, speed, range, and other variables involved, builds that appear to be average would be able to make use of all their stats so in essence they would be as effective as a build that is highly specialized. By giving attacks more attributes and ways to use multiple stats from the character at the same time, a "balanced" character would be able to have attack of equal power to that of somebody with a high attack, but they would just do it in a different way. What this would do is also increase the variety of attack types, as say a high power attack from a balanced build would have different properties than that from a purely high str character and thus require different defensive techniques to counter it. For example, say a FIREBALL technique vs a technique that is made up of three elements<fire, lightning, wind> each 1/3 the power of the fireball, but all at the same time are equal to it.

Shoyoninja: I'll be back to edit this when I have time.
EDIT: I'm back, but I'll reply in the newer post.

[Edited by - Fulgent on June 18, 2010 2:41:35 PM]

The problem with generalists is time. In the same time a specialist can do an action with a power of 100, a generalist could do it with a power of 75. It doesn't matter if the generalist can do a different action because the specialist can redo the same action and still be ahead. You can mix and match any variables, but that fact will remain. In group combat, time is the most important resource because rapidly disabling an enemy turns a 5v5 in a 4v5. A group of specialist will always be more powerful than a group of generalists because they have more power per time and are able to drop an enemy first.

That's for the general RPG where combats are a race to drop the enemy's HP to 0. Combat systems which work on a different base might allow generalists to be balanced because they have more ways to turn the battle in their favor. While they have more opportunities to advance toward victory, each time they do, they advance less than the specialist, so it balances out.

Shoyoninja: About your facts, I've observed such nuances in online games before. That is why I feel that stats that do multiple things destroy the balance like I showed in the example of the old gen pokemon's gameplay.

In the example you provided, the stat STR does aim, HP, and def while dex just does aim? 1 point of STR would then be worth several times more than dex. To solve this, dex should do more than just increase aim. It should also give dodge rate<parry?>, some attack power, and a third thing that dex would apply to in a way that makes sense.

Basically, if a stat is underpowered, then other stats should be brought to its level <no fun> or that stat should be brought up to their level by given it additional jobs in order to be of equal value to the others.

Also, I think its erroneous to think that specialized builds have advantages over balanced builds, unless it is the present RPG systems, then its true. If system is done right, then those with specialized builds should have glaring weaknesses or have a tough time against techniques that balanced builds are able to use.

The weakness of a specialized build would be something like extremely low HP, but high attack, so it becomes something of a glass cannon if you will. Another example is something with high HP but low attack. Theoretically, if a system were done right, something with high HP/low attack vs low HP/high attack hitting each other at the same time in a damage race should end up KOing each other at the same time<if using the same attack.>

Now then, this would show that the stats are balanced, because I've seen games where having a equal values in different stats ended differently. Like the person that had put their higher value in HP wins, or the person with the attack value winning by a miles.

Anyways, about that other example, if it were online, real players wouldn't just sit there and attack each other, they would try to use defense techniques and use offensive techniques of various kinds skillfully and strategically to get the win. They should both have an equal chance of winning if they both say have 10 different techniques to pick from during the battle and their all attacks that play a part in their strategy as well as compliment each other in some way. If their techniques don't have any kind of synergy of course, then they won't be as effective as the player with the better picked and well thought out skill set.

Also, the complexity of skills is very important, in this way, a balanced build could have a skill that somehow diminishes some of the attack power of a high attack build<although the diminishing should be of equal value to whatever stats are involved.> Also speaking of skills that increase/reduce the foes stats, they should active rather than something passive in order to balance out such techniques. In this, I mean, that a balanced build sacrificing a stat of theirs won't hurt them much while the highly specialized build would be made less effective, evening the field.

Tiblanc: This example you have provided is true in your standard RPG, however, it should be worth mentioning that in another system, a generalist would be able to tackle the weaknesses of a specialized build and bring them down to their level while such techniques wouldn't be usable by a specialist due to them lacking in some stat<For example, something specialized in high attack is going to be severely lacking in either HP, AIM, DODGE, or some other stat, or all the other stats would be well below average if they are all equal.

Generalists would be able to make use of techniques that make the most of their average stats. For example, a generalist would be able to set up a "trap" or sealing type technique, then be able to fight the specialist head on. This would be possible because their "INT" and "PSI" stat are good enough to allow the usage of "trap/sealing" type techniques, while a specialist would have a dismal "INT" and "PSI stat since most of the points are going into a single stat.

In a multiplayer game, if it were 5vs5 duel where its 5 specialists vs 5 generalists, 3 of the generalists should use defensive techniques to protect the entire group while 2 use sealing/trap type techniques to severely weaken a specialist or bring 2 from the other side down to a level equal to the techniques power. What will result is that it will be 3 specialists still able to attack while the other 2 would have been turned into characters with low stats<it was already low> and attack power that is average or below. On the other side, because of the defensive technique, the initial onslaught of what would have been a devastating initial attack in a simpler game would end up being that most of the team is somewhat damaged if the damage was spread evenly or if they tried to concentrate their attack on a single player, that player would still have 10-20% HP left due to the defensive technique of 3 players.<If they were defensive type specialists instead of generalists, 3 using a shield technique, 70-80% HP would be left even after 5 high level attacks from specialists with high attack power. They wouldn't be enough to KO a single player in a group of high defense types since they specialize in defensive techniques to the point that their damage output against their enemy is low/equal to attack specialists in terms of the rate of % their draining each others HP.>

In my belief, if a specialist and a generalist fight, their damage output against each other should be equal if the generalist uses techniques that turn their various average stats into something that is equal value to a specialist using an attack of a high base. <The weakness of specialists is that they won't be able to use complex techniques that require various stats to have a value high enough to support it.>

Of course, what I propose is that early on in the game, generalists might seem below par when they have only 1 or 2 attacking techniques<or those techniques are just standard types that could only be effectively used by a specialist.> When that character gets a hold of techniques that make uses of those stats and increases their skill set to around 10 or 20 different techniques, they should be able to defeat a specialist, who even if they had 10 or 20 techniques, they would mostly be linear as a generalist would have the ability to have techniques that vary greatly in how they behave when attack because of their variety of different stats being high enough to support such techniques. It is easier to see in a simulator game although I think it would still be observable in a turn based game that doesn't follow the standard RPG formula of stats and gameplay.

You could always have stats not directly affecting the battle.

For instance instead stats are used for storing abilities(for use). So that would mean that X move that is powerful but slow uses XX Strength and X intelligence, while X move that gives you a 10% chance to dodge while active uses XX agility and X stamina.

In other words rather than stats affecting the battle directly they affect how many and/or how powerful of skills of certain types you can use during an actual battle. Which means that when balancing you primarily need to worry about ability(yours) * ability(target) interaction rather than ability * ability * stat(yours) * stat(target) interaction.

lithos: I say that all the stats involved in an attack should affect all the mirror defense stats of the target. I'm against multiple stats going against a single stat in a formula. For example, that is why I say a generalist would be able to defeat a specialist 50/50, since an attack launched with properties based off multiple stats would attack all the stats involved in the calculation, so a offensive specialist has many of their stats low, they would get damage proportional to what they would deal to the generalist with their linear attack based off a single stat.

So an attack would only include in its damage calculation, all the stats involved, so if its just Att vs Def, then it would be a linear attack, but if its ATT-DEF, Special ATT vs Special DEF, MIND vs INT, and so on from the generalist, and because those three offensive stats would be equal in total to that single attack stat from the specialist, the damage would be about equal.

In an attack formula, there must be mirrors to the stats involved, if they all converge on a single defensive stat, that is when an imbalance is born. Which is the essence of the early gen Pokemon example in my initial post. The special stat was too centralized, doing the job of both attack and defense, so it was worth double the value of other stats.

I'm actually working on a single-player RPG design and I was just working on figuring out how to balance the stats. The game is based off of the Soul Eater manga with a good few twists. Here is what I came up with for stat sets:

Main Stats:
MGC - Magic(defines Magic power and resistance)
STR - Strength(Defines Attack power and defense)
AGL - Agility(Defines Speed and Dodge chance)
ACR - Accuracy(Defines Hit chance)
SLW - Soul Wavelength(Defines Special Attack strength)
SPC - Soul Perception(Defines SCAN effectiveness)

If you don't understand the last two, go to http://souleater.wikia.com/wiki/Soul_Eater_Wiki

You just need to keep in mind a balance concept like A beats B, B beats C, and C beats A.

Then you can do things like have an attack that debuffs you but hurts an opponent significantly - thus opening you up to a strong counter attack. Risk vs reward.

You just need to make sure your abilities and weapons offer trade offs that won't overwhelm other abilities and weapons.

You don't necessarily need X/2 stats so that each has an opposite.

Another idea is to vary the value that matters in gaining HP for each class. For fighter class it would be strength and dexterity while mage type classes would be intelligence and dexterity perhaps. Making hp (and other things) based on balancing two stats will help convince players to diversify at the cost of hyper specialization. They won't be able to milk just one stat and be effective.