Even more on balance
So, the wielder of the Nerfbat and I got into an email discussion about balance. Ryan posted his side of the conversation, so now it's my turn to post my side.
We've come to the agreement that "balance" is important to evaluate because that's what players ask for. The question now becomes: what is a useful measurement for balance?
Now, let me repeat one more time for those that are a bit below average in reading comprehension: I'm not talking about statistical balance here. I'm not talking about making everything perfectly equal only with different names. (Mages throw magic missiles! Fighter shoot arrows! Thieves throw knives! Clerics, uh ...throw crosses? ...shoot holy laser beams from their eyes?) Instead, my definition of balance is: an option is balanced when that choice is just as appealing as other options. This does not mean that damage output is exactly the same, or anything else people have tried to bring up. It means that if you were looking at two different classes (or skill sets, or pieces of equipment, etc), they would be "balanced" if both were equally appealing overall.
The really hard part here is that we're talking about a very large issue. You can't just look at one aspect and try to "balance" that alone. For example, if fighters and mages did the same amount of damage over a long period of time, that wouldn't necessarily be balanced. What about the fighter's armor protection? What about the fighter's ability to survive surprise attacks easier? What about the mage's ability cast area of effect damage spells? What about the mage's utility spells, such as teleportation? As you can see, this gets really complicated very quickly. Now consider that you have to do this with 6-23 more classes in a typical game.
So, how does this relate to our discussion?
Ryan wrote:
Ive long believed that skill systems aren't that much more difficult to balance than class systems, they just require an additional element of creativity. In fact, it may be easier to make adjustments to specific skills because you aren't nerfing an entire class (and therefore a person's game persona) permanently by changing something. They may have to find a new skill, but you haven't made them reroll.
I agreed with this. The reason why classes are "easier" to balance is because players generally have to suck up and deal with changes. In most skill-based systems, people could ditch the nerfed abilities and pick the new overwhelmingly powerful group of skills. In addition, if something is overpowered players can still take that skill (or skill set) and still take other skills as well to still make unique type characters. On the other hand, if you fuck up balance in skills you generally fuck up big time, because then everyone's playing cookie-cutter characters, the bane of skill-based systems.
Ryan also wrote:
The problem with the "it depends on your playstyle" reply to most of the build questions (even though that is the best answer) is that very few players actually know their playstyle.
This is really why you want to allow players to have as much flexibility as possible. Raph pointed out in a blog post or interview somewhere that what he should have done in SWG was allow people to master everything, but make each skill group dependent on some certain item of equipment. You could only equip a limited number of these items at a time, limiting what you could do at once. That way people could shift between skill sets to find their ideal playstyle. This also allows the player to feel "uber" because they can learn everything even if they can't do everything at once. More options instead of more raw power. This is similar to how Final Fantasy XI does their class system.
So, that's the continuing saga of balance in online RPGs. What can we do to help players feel that things are balanced? Is allowing more character flexibility the answer? What about allowing players to experiment to find their ideal playstyle? Or, are these related?
RSS feed for comments on this post.
TrackBack URI: http://psychochild.org/wp-trackback.php?p=215
Are you sure that was Koster? I did several searches and couldn't come up with anything similar on his site.
Comment by Wizzel Cogcarrier Wizzleton IV — 28 September, 2006 @ 6:30 PM
I'd prefer to say something's balanced when it's proportionately as appealing as the game calls for, rather than equally appealing as other options. It seems a mistake to believe all classes/skillsets should roughly share a particular ratio of design investement to player preference (# of players preferring that option). There doesn't need to be roughly as many players playing each class.
A better way of looking at it would be design investment to overall return. This acknowledges (1) that a player's gameplay experience may be significantly improved by the presence of other players who have chosen other, less-popular options, and (2) that the production of options for less common playstyles can, at times, be fiscally justified by the number of players those options attract, despite their being less commonly appealing.
Or maybe what you meant by "as appealing as other options" was more about a sense of fairness, that the lesser-played class is as appealing to its players as the more frequently-played classes are to theirs. To try to match those separate appeals quantitatively is like trying to discern if my friend likes fajitas as much as I like steak...hell, I couldn't even tell you which one I like better! Only a very general measurement could approach a confident accuracy. And then you run into the wall of people often not really knowing what they want, despite what they say (many, if not most, people are not terribly intrigued by self-analysis).
If an equivalent sense of fairness among players of different options is what you're after, it seems more efficient to simply concentrate on the numbers of players interested in each class...ensuring that they all justify their design expense, rather than trying to match the numbers.
That took me a while to spit out, so hopefully I'm not missing something completely obvious (as often happens). =P
Comment by Aaron — 28 September, 2006 @ 9:07 PM
Wizzel wrote:
Are you sure that was Koster?
Yeah, pretty sure. IIRC, he posted it somewhere back when he did the interview with the Escapist and talked about SWG; I think it was in a comment on someone's blog. Not surprising that other people have had the same idea. :)
Comment by Psychochild — 29 September, 2006 @ 4:35 AM
The only possible issue with the item limitation design, as I posted on Wizzel's blog, is that you are essentially fostering a "bank culture". The game becomes focused on the player's bank and inventory, so much of your player whining will come in the form of how the UI supports bank and inventory management, how much they can carry, how easy is it to get back to town and swap out at the bank, etc. I'm not saying its a bad idea, because few ideas are truly bad (there are a few though that are real stinkers), but rarely do I see anyone confront the impact of their design change on the rest of their theoretical game.
Comment by Jason — 29 September, 2006 @ 5:24 AM
Psychochild said:
and
But aren't most of the players who rant about balance the ones who are big time number-crunchers? I'm with you on the appeal thing, but I'm also not a big stickler on balance. Any time I've read balance discussions the most vocal players tend to pull out their actuarial tables and HP-28C calculators and launch into a pyroclastic tirade about how things don't match.
So, while balance seems like it should be a subjective matter, it seems (to me anyway) that the most vocal about it want very badly for balance to be strictly by-the-numbers.
Comment by chabuhi — 29 September, 2006 @ 7:09 AM
More on Balance
[...] It sounds like Nerf and Psycho came to an agreement. I think the answer to the question, “Is allowing more character flexibility the answer?” is “Yes.” Give players the tools and let them do with them as they please. If they want to go kill things, fine. They’ll find the best way to do so. The best thing the developers can do is give choices. Faster attacks with less damage, or slower attacks with higher damage? That’s where strategy comes in. [...]
Pingback by Sierra Kilo — 29 September, 2006 @ 8:34 AM
The reading comprehension piece is a bit hypocritical don't you think? I mentioned exacting statistical balance for a definition, only so I could quickly discount how pathetically lame that kind of game would be, it's also the only part you bothered reading, and quoting. It could be with my sad level of writing ability it just didn't make sense.
I'll try again.
You cannot reach balance based on appeal either. Everyone has different tastes. I hate playing rogue characters, nothing you can do about it to make it more appealing except for to make it more like the characters I like, in which case it wouldn't be a different class and would just be several classes all the same, with different names. No one would play such a game, becaus there would be no diversity, and most everyone elses favorite abilities wouldn't even exist. You can't balance for my appeal (it would be lame), you can't balance for your appeal (equally lame).
You've already stopped reading, but I'll continue anyway. ;)
So balance for the whole, to have an equal population distribution. Wrong! Next you'll think we could balance public speaking so that an equal number of us introverts are attracted to it.
It would be easier (though probably still impossible) to retrain the players to not expect balance. Competitive game play, compelling gameplay, not balanced game play.
Comment by BugHunter — 29 September, 2006 @ 9:44 AM
BugHunter wrote:
The reading comprehension piece is a bit hypocritical don't you think?
Welcome to my blog! Sometimes I'm an asshole, but I usually mean it in the spirit of fun. Don't take anything too personal, as I don't mean it that way.
You cannot reach balance based on appeal either. Everyone has different tastes.
Yes, it's subjective. So is the definition of what is "fun", but it is a central issue to game design. Even though there is no absolute answer to this question, we still have to find answers and deal with them. Yes, my definition is very subjective, but I also think it's the most accurate and useful.
So balance for the whole, to have an equal population distribution. Wrong!
No, no, no, no, no! This is not what I'm arguing! I'm arguing that the choices have to be equally appealing, not the choices have to be equally selected. This makes no sense in most designs, and it is too restrictive to require that all game designs have an equal distribution.
We've been focusing on only one choice here: initial class selection. But, you can measure class balance in a lot of other ways. Consider these questions: "In a PvP battle, would you rather fight a Mage or a Rogue?" "In a small group (PvE), would you rather invite a Mage or a Rogue?" If the answers to these are about equal between Mage and Rogue, then these two classes are balanced in these situations. This does not mean that the percentage of fights between you vs. a Mage and you vs. a Rogue have to be equal. It does not mean that for every Mage you invite into your group, you have to invite a Rogue.
To further illustrate the point, consider if things were unbalanced. For example, let's say that in response to the question, "In a PvP battle, would you rather fight a Mage or a Rogue?" the overwhelming majority of people respond, "A Mage, duh!" Why would that be? Probably because the classes are not balanced. Perhaps the Mage is underpowered, or there is a widely-known strategy to reliably defeat Mages. Or, perhaps Rogues are overpowered, and considered practically undefeatable in PvP. So, there needs to be some balance work done in this case, and the game designer now has to figure out which of these scenarios is more applicable. Then the designer has to determine what's the root cause of the scenario. Are Mage spells easily resistible with common gear? Is stealth undefeatable? Or any of a million other issues.
Now, keep in mind that balance is often larger than any specific evaluation. In the case above, maybe mages are designed (or simply implemented) to suck in PvP but they rule in PvE. So, even if people answer the two questions above out of proportion, the classes might still be considered "balanced" overall. (Although, personally, I wouldn't consider them balanced in this case; most people want to hold their own in both situations.)
Now, don't think the answer to all these balance questions has to be exactly 50/50. 51/49 is still fairly balanced (and allows for statistical oddities); hell, I'd be happy with 60/40 given how subjective these questions are. You also need to watch out for the weirdos that throw off the your observations just because they don't like to play Rogues. (See? I did read your whole post. And I did last time, too.) And, once again, realize that balance isn't a magic bullet to get people to stop complaining. People will always want a bit of an advantage, a bit more power for their own character. Balance is desirable because it's the most fair, in general.
Yes, balance is hard. Yes, it is subjective. But, you can still achieve it in a game. Shunning an aspect of game development because it is hard is fruitless: all of game development is hard, except for the armchair game design that happens after the fact. So, if you don't want to have to deal with the hard issues, do not get into game development.
My further thoughts.
Comment by Psychochild — 29 September, 2006 @ 1:44 PM
Skills and Skill Caps
[...] During the recent debate on whether balance in MMORPGs is possible, Psychochild raised an idea (attributed originally to Raph Koster). Raph pointed out in a blog post or interview somewhere that what he should have done in SWG was allow people to master everything, but make each skill group dependent on some certain item of equipment. You could only equip a limited number of these items at a time, limiting what you could do at once. That way people could shift between skill sets to find their ideal play style. This also allows the player to feel “uber” because they can learn everything even if they can’t do everything at once. More options instead of more raw power. [...]
Pingback by WorldIV — 2 October, 2006 @ 8:23 AM
I like where you're going with the whole idea of "Who should we invite to the group, the mage or the rogue?", and the answer being "It doesn't matter, either one would be a benefit." Still not possible though when you ask "Who should we invite to the group, the mage or the cleric?" Maybe it's not even entirely the designers fault. The players themselves can't think past "Needing" a healer to do anything. The typical Tank/healer/dps makeup is BS, but most players won't hear otherwise. So you can't balance appeal from this perspective either, while even if you get it right technically, the players won't figure it out, and they won't listen if it's explained. To get to where you're talking about it will take a very painful process of retraining us.
Comment by BugHunter — 2 October, 2006 @ 1:15 PM
BugHunter wrote:
The typical Tank/healer/dps makeup is BS, but most players won't hear otherwise.
And the games aren't designed otherwise, so "balance" becomes a chimera from the start. We're stuck in the current DIKU MUD design parameters, and we are hesitant to break out of it. I was going to say that we're in the "D&D" mindset, but even D&D doesn't require each tank to have a team of Clerics behind him or her casting powerful healing spells every round. And, even D&D Online changed the system to use spell points so that clerics could sling healing spells frequently as per the DIKU MUD model.
To get to where you're talking about it will take a very painful process of retraining us.
This is why I'm a big supporter of indie games. I think we'll need a small-scale game (or something beyond the high fantasy DIKU clones out there) to lead the way in this "retraining". In geeky math terms, we've reached a local maximum and think we're king of the hill, but I have the feeling that behind that hill is a range of mountains we should be climbing. To get there we have to tread across some lowland.
Okay, enough metaphors for now. You should get the picture.
My further thoughts,
Comment by Psychochild — 2 October, 2006 @ 1:42 PM
Balanced but not equal
[...] don't agree. In my previous posting on balance, I said: An option is balanced when that choice is just as appealing as other [...]
Pingback by Psychochild’s Blog — 11 April, 2009 @ 10:08 AM