The fine folks at GameCyte did cut a second part of my interview after all. This one touches on the more abstract concepts of balancing games. Topics include symmetric vs. asymmetric games, hardcore vs casual balance, and depth vs. fairness.
Maybe I should have a weekly show or something?
--Sirlin
Related Posts
• My Interview at GameCyte
• Gamasutra Interview about SF HD Remix
• Playing to Win Interviews
• GamesRadar Interview on SFHD Remix
• SF HD Remix in the News A Lot
• Digg this and help spread the word about Sirlin.net
• If you found this article useful and would like to donate to sirlin.net, please proceed to the Donation Page
• Discuss this article in the Sirlin.net forums
This entry was posted
on Thursday, May 15th, 2008 at 4:27 am and is filed under Fighting Games, Videos, Blizzard Stuff.
You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.
May 15th, 2008 at 4:33 am
Weekly show?
YES PLEASE DO :D
May 15th, 2008 at 5:47 am
I’d watch it.
May 15th, 2008 at 8:05 am
I’d pay for it.
May 15th, 2008 at 8:14 pm
You know i would love to watcha weekly show of david sirlin’s viewpoints on fighting game material. Like you would have a break down of different types of fighters each week or something. Just a idea for a show.
May 16th, 2008 at 12:28 am
I like reading text more than I like watching videos. A podcast would be OK though :-P
May 16th, 2008 at 1:48 am
I’d love to have Sirlin remix every popular fighting game, and make a hundred articles on how he did it. This stuff is good reading!
May 16th, 2008 at 7:10 am
I hate watching videos when the same information could just as easily be presented with text. If you do a weekly video series you should do something that takes advantage of the medium, not just talk.
May 16th, 2008 at 1:06 pm
SIRLIN: TEH PODCAST
ZOMG
May 17th, 2008 at 2:43 pm
I agree with Teddust. And this is specially painful when English is not your native language, like in my case. Sure I get everything, but… I have to concentrate! Reading doesn’t require full attention from me, I can always re-read something or look up a word if I’m not sure of its meaning.
Regarding the interview, it’s great ;)
May 19th, 2008 at 11:13 pm
I thought the interview was quite good, and certainly informative.
I like visual aids and cues with subjects I’m not very familiar with, the accompanying imagery within these articles will typically suffice, but I would not be against some form of video based podcast if it ever becomes feasible.
That said, I’m quite capable of reading, too :)
May 21st, 2008 at 8:09 am
No offense Sirlin, but from what little I’ve heard of your voice (a couple SF2 HD interviews and CCC2 tutorials), you’re much more entertaining writer than you are a narrator. You’re excellent at presenting information in text in a humourous, interesting, and entertaining way. Speaking, your straight forward and very informative, but lack the same hook, line, and sinker you have with writing. You’re a writer, not a comedian.
But I never seen Bang The Machine, maybe you were trying to be straight-forward and informative in those, but have proved you can be entertaining there too.
But whatever you feel you want to do.
May 22nd, 2008 at 4:54 am
Why do you consistently misuse “symmetry” when discussing game theory? It’s not about whether player 1 has access to different options from player 2, it’s about whether the payoff matrix is symmetrical.
One could easily imagine games that are asymmetrical in your use of the word, but not in a game theory sense (player 1 has an additional move, but it’s dominated). Likewise, even in mirror match fighting games, the payoff matrix for a particular situation is asymmetrical as soon as one player has a positional or frame advantage over the other.
May 22nd, 2008 at 5:42 am
cody: Nice tone, maybe work on that. Anyway, I need a word to talk about what I’m talking about. When both sides have the same options, the word “symmetric” is pretty descriptive, therefore I use it. Apparently I don’t need a word to talk about what you’re talking about. The concept of whether both sides have the same pieces/moves/units comes up as often as twice per sentence, so I certainly need a word for it.
If I wanted to refer to whether or not the payoff matrix is symmetrical (which I have so far never needed to even mention), I guess I would say “games where the payoff matrix is symmetrical.”
Also, I don’t like the wording of your fighting game example at all. If it’s a mirror match, then it’s a symmetric game. A symmetric game has “local asymmetries,” sure. If you take a snapshot and I’m knocked down and you’re hovering over me as I get up, the particular situation has a different payoff matrix for each of us. But the game itself of Ryu vs Ryu is symmetric in that we both start with the same options and there is no need to worry about whether Ryu1’s moves are too powerful versus Ryu2’s moves. There is need to worry about whether the various situations that occur along the way are interesting and deep, but I tend to call that “depth.” Certainly, I need a term that calls Ryu vs. Ryu symmetric, because otherwise there is no value to the term. It should be understood that a symmetric game does not have every single situation that occurs along the way be symmetric, too. Quite the contrary, actually.
May 22nd, 2008 at 10:39 am
People are *still* arguing over what “symmetric” means? It’s like going to a class on colour theory and spending all course trying to debate the proper meaning of the word blue.
Anyhow, nice interview. I guess I prefer reading to watching as well but I’m not picky.
May 22nd, 2008 at 11:52 am
Heh. I wonder if these people would yell at a poker author for saying AK “dominates” AJ in Texas hold-’em even though that’s not remotely close to the game theorists’ definition of the word. Even the ones like Chen and Ankenman whose book specifically employs in-depth game theoretic analysis.
May 22nd, 2008 at 8:15 pm
Yeah, the fact that a symmetrical game can still have uneven payoff matrices at any given point during the game seems so obvious that it’s almost a truism. Games where this is not the case, such as RPS, are not particularly interesting. Hell, even in asymmetrical games, if the payoff matrices for any given situation remain largely static, rather than being dynamically altered by player action (e.g. SF3: Third Strike, where it’s extremely difficult to exert an advantageous situation over an opponent thanks to parrying), it’s not very interesting either.
Coming back to ST, I think that’s why I’ve always found mirror matches between characters whose optimal range is point-blank so dull. In a game where the character’s payoff matrices at any given moment are largely influenced by range, and only secondarily by frame advantage, it’s extremely difficult for, say, Zangief to assert a meaningful advantage over another Zangief outside of obvious situations like a knockdown.
I sometimes wonder if a game that combined the complex space control of ST with the complex frame advantage control of VF would be interesting or just too much.
May 23rd, 2008 at 12:36 am
Polarity: What would that game be like? VF includes a lot of space control due to the strength of wall combos and ring outs. And ST certainly has important frame-based advantage situations… reversals, for one. It’s not like a black and white difference… just about every fighting game falls into some shade of grey betweens positions and frames, and a lot of them are already pretty close to the center, it seems.
May 23rd, 2008 at 1:03 am
Polarity - I don’t think that both is simultaneously possible. A particular situation will either be dominated by space advantage, or by time advantage. Consider, for example, that knockdowns are really a frame advantage situation and are AFAIK a pretty big deal in Street Fighter.
In a game like chess people talk about controlling the center - which is position - and tempo - which is time advantage, and consider the two to be aspects of winning play. The same is true in fighting games - positional advantage and time advantage are typically tied together.
May 23rd, 2008 at 1:25 am
Yeah, my point wasn’t that both would be simultaneously possible, but rather that there would be two main games - up close, the frame advantage game a la VF, and from further away the space control game a la ST.
It’s true that there are features of both elements in both games, but there’s no denying that ST’s space control is hugely more complex than VF’s, and VF’s frame control hugely more complex than ST’s.
May 23rd, 2008 at 3:05 am
Up close, VF and further away, ST… That’s the game I’ve been dreaming of for years. Guilty Gear and Street Fighter 3 has a little bit of the frame games, but neither have the same feel as VF/SC/Tekken. Maybe it has to do with the “rubber band” (i.e. moves push characters toward and moving backwards is slower) that 3D fighting games have.
May 23rd, 2008 at 7:55 am
This isn’t about tone, or arguing.
Sure, you need a word, but “symmetry” is already in use in the context you’re speaking in, and it means something different from what you intend.
http://en.wikipedia.org/wiki/Symmetric_games
Most fighting game players are going to understand what you mean if you were to say “mirror match” instead (arguably, moreso than would understand “symmetrical”). Furthermore, the first three google hits for “mirror match” are an accurate definition of what you’re describing.
May 23rd, 2008 at 8:05 am
cody, can you honestly say that his use of the term “symmetry” confused you? Because, if it didn’t, then it is a good use of the word. Words can have multiple meanings, you know. Also, I think that there are many people who would find “mirror match” a confusing term with which to describe chess, far more than are confused by “symmetrical game”.
May 23rd, 2008 at 8:23 am
I can’t play the game worth a whit, but would people consider Virtual On more of a frame advantage or space control game?
May 23rd, 2008 at 10:27 am
I think there is a frame advantage element to using melee attacks in Virtual On… but mostly it’s space control. Timing doesn’t really matter on the ranged attacks as far as I know… it’s just whether they hit or not.
May 24th, 2008 at 11:17 am
Sirlin’s not half-bad on camera, but pretty dry. With a script and devices to take advantage of the medium, I’d think a Sirlin show or podcast could be pretty good. Even it was boring, I’d probably still watch it, but I’m a pretty big nerd.
He probably wouldn’t, but say Sirlin placed a moratorium on the discussion of symmetry in games. Bets on how quickly that thread would turn into an argument about symmetry in games?
May 25th, 2008 at 6:46 am
‘can you honestly say that his use of the term “symmetry” confused you?’
Yeah, it really did confuse me, and it apparently confused a lot of you too, or you wouldn’t have wasted time arguing about whether X is a symmetric game.
The problem is that his use of the term is close to the accepted definition, but just different enough to cause confusion: Symmetric games _are_necessarily perfectly fair; and a game in which player 1 and player 2 have the exact same set of pure strategies with the exact same respective payoffs _is_ necessarily a symmetric game. What confused me is that he’s claiming stuff like chess is symmetric, when it undeniably is not symmetric (hint: is chess perfectly fair to player 1 and 2?). He’s apparently using symmetric to mean “close to being symmetric”, i.e. the two players have the same or close to the same moves, but there may be some differences in payoff. This is apparently justified by comparing these so-called symmetric games to games like ST, which are very far from being symmetric. This makes about as much sense as saying 10 is really far away from 0, so 1 must be equal to 0.
Polarity and sirlin had some valid complaints about my second fighting game example, so maybe this will be clearer: Akira vs akira is a mirror match, right? Same movelist, same damage, yadda yadda. So if mirror match is the same as symmetrical, this game has to be symmetrical, right? Wrong. Player 1 has access to the strategy of avoiding all attacks and throws in order to win, player 2 does not. I don’t know enough about how other fighting games besides VF resolve sudden death rounds, but something tells me this is not unique.
Why does this matter, aside from people being wrong on teh intarnets? Besides, like polarity said, symmetric RPS is boring!
Well, for starters, there are techniques for transforming an asymmetric game like chess into a symmetric game, while preserving the strategies that make chess interesting. If you can’t even properly use the idea of a “symmetric game”, how are you ever going to get to more interesting questions? For instance, can these same techniques be adapted to transform fighting games into symmetric (and thus fair), yet still interesting, games?
May 25th, 2008 at 7:44 am
While I appreciate your insight, you seem to be pedantically belaboring a point that nobody is arguing with. Sirlin’s working definition of ’symmetry’ is concise and thus useful in this context, regardless of whether its an accurate reflection of the game theoretic sense of the word or not.
May 25th, 2008 at 7:46 am
Er, or rather, “The word ’symmetry’ is concise and thus useful to apply to Sirlin’s working definition in this context”
May 25th, 2008 at 11:09 am
Sirlin needed words to describe the difference between “no selecting of characters pre-match” and “players select characters pre-match”. He chose “symmetric” and “asymmetric” because they fit, and he even explained what he meant by those words. There may be some people who posted their confusion about the subject, but I’m willing to bet that the vast majority could read between the lines and understand what he was talking about.
Who cares that those words have other meanings? Those meanings aren’t commonly used for realtime video games anyway. I could see there being objections if he was talking about map symmetry with respect to gameplay options instead of topology, but the objections I’m actually seeing are downright ridiculous.
May 25th, 2008 at 2:09 pm
Akira vs Akira is a mirror matach aka symmetric according to my usage. Saying that one player has skills the other doesn’t is obviously beside the point. If we counted that (and of course we don’t) then all games would end up in the asymmetric category and the terms would have no usefulness left. Cody, if you’re smart enough to even know the game theory definition in the first place, you are surely smart enough to grasp such an obvious point.
May 26th, 2008 at 4:14 am
Polarity, I appreciate conciseness, but sirlin is leaving out non-superflous detail. It’s not enough for the two players to have mostly the same options (or even exactly the same options). He’s using a necessary, but not sufficient, condition of the actual definition as the only component of his definition. As for belaboring a point that no one is arguing . . . to me, not arguing would look more like “yeah, I’m misusing the word, I’ll try to be more precise.” As it stands, the point is definitely still being argued.
PoisonDagger, the problem isn’t that the word has another meaning, it’s that it has an accepted meaning that encompasses what sirlin is trying to talk about. It comes off as misunderstanding the meaning of the word, not using it to talk about something completely different.
Sirlin, reread what I wrote. Akira vs akira being asymmetric has nothing to do with one human having skills the other one doesn’t. The 1p side wins if neither player lands a hit or throw (or ringout, I suppose) before time runs out; this is part of the structure of the game. Just like black going first in go, this is an example of why your habit of arguing “the two sides look mostly the same, must be symmetric” is a problem.
If you’re smart enough to proclaim yourself to be a game design “guru” and the source of “shocking insights” on game design, you’re surely smart enough to use an accepted term correctly, especially since the actual definition is REALLY CLOSE to the way you’re using it.
It’s not like the game theory definition is terribly complex or hard to understand. Anyone who’s taken the time to read ‘The Compleat Strategyst’, which requires at most grade-school arithmetic and about $2, should grasp the concept.
And seriously, I’m not being facetious, you should take a look at how chess can be made symmetric, and how that transformation might apply to fighting games.
May 26th, 2008 at 7:25 am
cody, now that I know what you meant by the Akira example, nothing has really changed. If you put Akira vs Akira in the asymmetric pile because of an outlandish condition that will never actually happen in gameplay, then term has no useful meaning anymore. By that needlessly rigorous standard, pretty much all mirror matches in all fighting games are asymmetric. There are subtle bugs that make 1p and 2p side barely, barely different in ways so small as to be overwhelmed by the weight of everything important to the game being the same.
I’m not really that interested in how chess can be made symmetric by your meaning either. I mean as a laboratory study, sure. But consider the spectrum of difference, say 1 to 10, where 1 is perfect symmetry and 10 is vast difference in characters/moves/races/etc while still maintaining overall fairness. Chess is like 1.1 at most while Starcraft and Guilty Gear are maybe around 9. How to get a 1.1 to a perfect 1.0 is not that pressing of a problem. How to make a 9 that is actually fair is a pressing problem.
Furthermore, if you are looking at a games that did it right, you have to have some standard of what “it” is. Whether Chess is 1.0 or 1.1 or 1.2 or even 2.0 is not really the point here because it’s so far from actual meaningful differences in sides that it’s not part of what I’m looking for or talking about. This reminds me of someone else’s analogy of separating sandwiches into “hot sandwiches and “cold sandwiches.” Even without super precise definitions, we can round up some hot sandwiches while keeping cold sandwiches off our list, even though cold sandwich A might be some 1 degree colder than cold sandwich B. I don’t see why that is so confusing. Getting into mathematical details would bog everything down with little gain when the task is simply, “name things somewhere towards 10 on the scale” or “name some generally hot sandwiches.” So when I say “symmetric game” I mean “game somewhere reasonably close to the 1 on this scale.”
If you use some common sense where Starcraft and Guilty Gear differences in characters/races are a standard to strive for, then CLEARLY Chess is not in the same pile. I don’t see how anyone could even attempt a claim that it is considering each side has 16 pieces identical to the other side while Testament has invisible trees and webs, Venom has pool balls, Baiken has unlimited guard reversals, etc. In fact, of dozens and dozens of games brought up in the thread about this, only Team Fortress 2 caused any real concern about which general pile it belongs in (answer: it’s more similar to 1.0 than it is to 10.0 on the scale).
Final reminder: the scale I mentioned is not a value judgment. A 1.0 game can be superb and deep.
May 27th, 2008 at 11:37 am
‘when I say “symmetric game” I mean “game somewhere reasonably close to (symmetric)”‘
Yes, exactly, glad we’re on the same page. Is saying “almost symmetric” too cumbersome for you or something?
The reason small differences in fairness ( I say “fairness” because I honestly have no idea how you can quantify relative symmetry ) are interesting to me is because if a game sticks around long enough, players are going to push the boundaries: VF matches may not come down to sudden death every time, but games of go are regularly determined by komi. So you need some way to address even small amounts of unfairness, and the exact same techniques that work on nearly fair games also work on very unfair games.. Let’s say a game of chess, on average, is worth .2 to white; i.e. white would have to give black 2 games out of every 10 in order to make the game perfectly fair. As you point out, sudden death is rare, so lets say akira vs akira is only worth .001 to player 1. So akira vs akira is “fairer” than chess; however, this doesnt matter:
You can transform chess into a symmetric, and thus fair game by adding another option to the choice of sides, as follows: Rather than, say, flipping a coin to decide sides, each player chooses black, white, or hedge. If the players choose different colors, play the game of chess as normal. If they choose the same color, or both hedge, its a draw, worth 0 to both. If one player hedges and the other player chooses a color, hedge beats white and black beats hedge. Hopefully, this should be sounding Really Pretty Simple . . .
The payoff matrix is
_ B W H
B 0 -.2 1
W .2 0 -1
H -1 1 0
Clearly, this is a symmetric game with an expected value of 0. Just as clearly, it does not matter how unfair the sub-game is - you could replace the .2 and -.2 with .001 and -.001, or whatever, and it would still be a fair game. However, it does matter, strategically speaking, how much of an advantage white really has, because this informs how often you should hedge. This modification thus rewards appraisal abilities as well as yomi, unlike fixes such as flipping a coin, adding komi, etc. It also doesn’t remove any strategies from the underlying game.
If you think about it a bit, this is also directly applicable to multi-character games e.g. Starcraft. You, as a game designer, don’t even have to know how much better Protoss are than Zerg, you just need to know that they are better. To a certain extent, you can even pawn this off on players by letting each one estimate the tier list that they will be bound by; that way, if the game is unfair, it really is the players fault.
Obviously, there are some issues with this. While this works for reasonably balanced but still unfair games, if the underlying game is seriously broken, then even though this modification will make it fair, it won’t make it significantly more interesting - instead of seeing 95% chun, you’ll be seeing 95% hedging. Also, while it’s clear what hedging is worth when you’re playing chess (you won this game, go on to your next 2 out of 3, etc), it’s less clear what its worth in multi-character games. You can still have simple payoffs: hedge vs protoss = win; hedge vs zerg = lose; hedge vs terran = draw. However, this tends to make mid-tier characters less worthwhile, even though, again, the game is perfectly fair. You could solve this by having fractional payoffs for hedging and letting the players bid on them, but it’s not clear what 2/7th of a game of ST is worth in a tournament.
Fighting game players already do a certain amount of this to balance out games ( while you may think 1p or 2p side is irrelevant, I’ve been in tournaments where the rule was flip a coin, winner gets to either choose stage or choose side ). I suspect this sort of behavior at a meta-game level, including estimating tier lists, choosing matchups etc., is part of why you regard obviously unfair, asymmetrical games as being generally balanced. However, I think it’s worth considering how to integrate these kinds of decisions into the formal game, even if only during playtesting.
May 27th, 2008 at 1:21 pm
Cody: I have one quesiton for you: Who cares? We certainly don’t… Sirlin’s question wasn’t how to make chess symmetric… it was how to make characters balanced in a fighting game or RTS. In fact you’re solution doesn’t even work because you forgot about mirror matches… in chess, if both people choose white, you can safely tell them to “roll again”. In ST, if both people choose chun… well, that’s fine, now they’re both using chun, so there is no “hedge” option.
Regardless, I have no idea why you felt the need to post a rather obvious mathematical proof of how to create game theory based fairness through concepts outside of actual game rules (Your idea basically falls under tournament rules… not the rules of chess or the rules of a fighting games). In fact, your entire idea has already been discussed several times in other threads on this site, and discarded by everyone, as basically irrelevant to the actual discussion we’re trying to have.
May 27th, 2008 at 1:41 pm
I still think you are being intentionally blind to the degree of magnitude of these things. You say you don’t know how I can quantify “how much symmetry” something has, but really you do know. You’re just being difficult. Akira vs. Akira where a draw is a draw would be 100% symmetric. The actual Akira vs. Akira where the only asymmetry is a draw case that will probably never happen is “close to symmetric.” It’s clearly closer than say, Ryu vs. Zangief. Akira vs Akira, both characters have over 200 moves, all exactly identical in every way. Ryu vs. Zangief, the entire moveset is completely different, the hit points are different, the walking speed and jumping speeds are different. One has a fireball and can fight from far, the other must fight from close. The differences are enormous. So if we have a spectrum here, Ryu vs. Zangief is at one end and Akira vs. Akira is so close to symmetric as to be negligible. No one interesting in anything practical would care if we classify it as symmetric.
This complicated hedging system (which I still don’t understand what happens from it, I think you might have left out what to actually do with the results, or maybe I just don’t get it) is not needed in fighting games or Starcraft. Double-blind select works just fine, done.
As for tiny differences of 1p / 2p, they are unfair, yes. But the magnitude of their unfairness is extremely small. So small that they are overwhelmed by character choice and player skill. Flipping a coin is acceptable to me because the difference is splitting hairs that don’t even matter much. If you are able to win on p1 side but not able to win on p2 side, you pretty much deserve to lose anyway. Also, there is no such player as that. Players greatly exaggerate how much that matters and claim it will affect them, in reality, almost the entire fight is unrelated to those differences and they usually do not happen at all. Players can switch side after each game anyway. Now, there could be some future game that’s completely wonky where p1 has an enormous advantage. That sounds like a game that should be abandoned for tournament play though. If I could wave a magic wand to eliminate these (nearly insignificant) unfairnesses, I would. But since I can’t, “just suck it up” is a fine answer. Choi and Tokido or whoever will still win either way.
Again, I’m concerned not with microscopic asymmetry, but rather with enormous asymmetry. And yes, you do know what that means.
May 27th, 2008 at 11:19 pm
Claytus, I didn’t forget about mirror matches. If the mirror match is symmetric, then just play chun vs chun, it’s fair. If it’s not symmetric, it’s because 1p chun is a different character from 2p chun, and could be treated as such (both people keep picking 1p chun? get smart, hedge the next round, and win).
Sirlin, it’s not a complicated hedging system, it’s just balanced RPS with unequal risk-rewards, which you ought to be pretty familiar with. What you actually do with the results is the same as what you do with a given round of chess or match of a fighting game - if they hedged and won, they win that round / match. Win 2 out of 3 (or 3 out of 5 etc), and you’ve beaten your opponent, move on in the tourney. Yes, double-blind select works just fine, but the end result is still an unbalanced game - players will figure out the best character in the game and only play that character (or small group of characters) to the exclusion of others. All I’m suggesting is making the tiering process explicit, to automatically balance the game. Even if you’re just using it for playtesting, you should be able to measure how bad the balance is by how often people hedge.
As for nearly insignificant unfairnesses, I agree that Tokido is just gonna win either way. But at a lower skill level, 1p side matters quite a bit, especially for akira players. You can wave a magic wand to eliminate these small unfairnesses (at worst, just randomize it in-game), so why not do it?
I ‘m also being serious, quantifying (ie a number, not just a gut feel) relative symmetry is not obvious to me at all, I don’t know that much about game theory. Which one of these games is more symmetric:
2 8 6
10 0 4
6 0
4 10
8 2
It’s pretty obvious which one is more unfair, but symmetric? I dunno . . .
If you can’t do this for simple games, how can you, with any kind of precision, say that ST is more or less symmetric than GGXX?
May 28th, 2008 at 12:12 am
“But at a lower skill level, 1p side matters quite a bit, especially for akira players. “
… No, actually at a low skill level it matters even less than for good players. In the few 2D games with side imbalances, the 1P vs. 2P issues are extremely small, such that only extremely good players can grasp the differences and capitalize on them anyway. As far as VF, I’ve only seen one sudden death match occur, ever (a kage did an evade attack while against the ring edge, causing a ring out, but the attack landed, and also KOed his opponent). And the sudden death round to resolve that was concluded with a normal win. I’m fairly certain we can ignore the side imbalance you’re referring to, because it never, ever, actually comes up in a tournament.
As for your final question… again, Who cares? We don’t want to know whether GGXX or ST is more symmetric. We want to know if ST and GGXX are both significantly less symmetric than chess, which they obviously are… they’re near 10 on our scale, while chess is near 1 on our scale. Like Sirlin said before, you clearly understand exactly what we mean in your posts, you’re just bringing up ridiculous nitpicky questions to try and find some case where the definition aren’t exact. Well, it doesn’t matter… the definition are allowed to not be mathematically perfect, because we don’t know that level of accuracy. If we did need that level of accuracy, we’d have dropped the innacuracies of english words altogether, and just talk math.
May 28th, 2008 at 12:13 am
er, we don’t *need* that level of accuracy
May 28th, 2008 at 1:33 am
“especially for akira players” == joking about the human element Sirlin originally thought I meant. Most akira players I know can do modified doublepalm at lot easier on 1p side ;)
“Who cares?” As a game designer, if your goal is to create more asymmetry while maintaining fairness, shouldn’t you have some measure of what “more asymmetry” means? Given 2 equally fair games, is the one with more strategies more asymmetric? The one with a larger maximum difference between the lowest and highest payoffs?
“I’m just being nitpicky.” Actually, I honestly was confused the first time I saw sirlin bring up symmetry, and I’m pretty sure he was as well. I think its clear to both of us now that he does actually mean the same thing as game theory symmetry, he’s just being imprecise (or, to be more charitable, close enough) about it. As for merely nitpicky, if you don’t want to clarify your thoughts and make them more precise, why are you bothering to write at all? Why did sirlin bother to distinguish between “balance” and “fairness” if nitpicky distinctions don’t matter?
May 28th, 2008 at 3:26 am
Nitpicky distinctions don’t matter because there aren’t any games who’s rules contain minor distinctions that we have seperate with more precise language. That was the original point of Sirlin’s “most balanced games” thread which is sort of the root of all this… we found that the list of balanced, highly asymetric games basically contains a few fighting games, a few card games, and starcraft, and that’s it. Therefore, we don’t really need to make precise distinctions between things that are on and off the list. The list is so short we can point to it whenever a question arises and say “we mean that”, and everyone nods their head, and says okay.
If you can come up with an actually useful example, we’re all ears… but I think Sirlin himself is ready to move past the definitions phase and start doing some game designing (I think he already has started with some of his card games). Having every single discussion of this topic get derailed by someone telling us the definitions are imprecise is getting really, really old. Using the scale again… we’re talking about all the games that are around 8-10… but you have to realize that there are no games that land between approximately 2 and 8 on the scale. It’s a completely barren wasteland, so just being precise enough to keep everything around about a 2 out of the discussion really is exactly the amount of precision we need.
May 28th, 2008 at 5:20 am
I *still* know what you do with these “hedge” results. Give a concrete example. Someone chooses hedge and someone choses Chun Li. “If they hedged and won, they win the round.” What? If they won the pre-game hedge-off? Or if they won the actual gameplay? I still don’t know what you’re saying.
Side differences in fighting games matter waaaay less at lower levels, as Claytus said. It’s pretty much unthinkable that they could determine a winner between beginners, so it’s negligible.
You claim to not be able to measure differences in asymmetry, but what you mean is you don’t have an exact equation that would satisfy a mathematician. You can, in fact, tell the difference between an Akira vs Akira match with more than 200 identical moves and a Ryu vs. Zangief match with completely different moves, hit points, speeds, and possible strategies. It’s an enormous obvious difference versus a microscopic difference. Just because you don’t have an equation does not mean you need to be intentionally blind to enormous obvious differences. As we’ve said before it’s been very easy to classify whether a given game has design challenges similar to chess (make it deep, don’t need to worry about balancing sides) or Starcraft (make it deep and DO need to worry about balancing sides.)
As Claytus said, this super precision you’re talking about is theoretically interesting but has not been needed in any practical sense due to the barren wasteland he mentioned. My “most balanced games” thread (which sought highly asymmetric games) came up with few games, indeed.
And finally, “fairness” and “depth” are really different concepts. Not different in the way of an Akira vs. Akira match where there’s 200 identical moves and one insignificant difference. Instead, the concepts are extremely difference. Chess has “depth” (lots of interesting decisions) and virtually no need to care about “fairness” because the power level of playing white vs black is so close. Starcraft and Street Fighter need to care and care and care all day about balance due to how diverse the movesets are. Then, they need to care some more and release another patch and another sequel and so on.
It’s yet another case of enormous obvious differences as opposed to microscopic differences. I still maintain that cody can tell huge vs microscopic apart if he stops demanding super precision in cases where it’s just not needed. Actually, he could still demand super precision if he wants, and even then he can tell huge vs. microscopic apart. It’s the overly rigorous attitude that rejects all things that aren’t yet rigorously proven. If I gave you a hot sandwich, so hot that it burned the roof of your mouth, you could still classify it as a “hot sandwich” without needing to know the exact degree Fahrenheit at which it becomes “hot” mathematically.
Yeah some sort of scale of how asymmetric a game is would be interesting, but it’s not very practical to create and not necessary to find the highly asymmetric games that are balanced that I was looking for.
May 28th, 2008 at 8:06 am
What to do with hedge results:
There’s a tournament of SF3 “balanced edition.” Games are decided by best 2 out of 3 matches. Our protagonist, Abe, is playing Bob for his first game in the tournament.
1st match: Abe picks a lowtier char (say Q). Bob picks a toptier char (say Chun). They fight. Abe gets his ass kicked in the fight, so he loses the match. Score is A:0 B:1
2nd match: Abe says to himself, “man, chun is broken, I don’t want to play against that shit”, so he hedges. Bob picks Chun again. Hedge always beats toptier, so Abe wins the match. Score is tied, A:1 B:1
3rd match: Abe says to himself, “wow, this hedging thing is awesome,” and hedges again. Bob picks Twelve. Hedge always loses to lowtier, so Abe loses the match. Score is A:1 B:2
Bob has won the game, so he goes on to winners bracket and Abe goes to losers. Make sense?
Yeah, fairness and depth are different concepts. Fairness and balance, however, are arguably the exact same thing. Trying to distinguish those burns my mouth just as much as trying to be precise about symmetry burns yours.
Why “microscopic” distinctions matter:
Yes, you and I can both tell a hot sandwich (”it burns my mouth, must be hot”) from a cold sandwich (”doesn’t burn my mouth, must be cold”), but you have no idea why it’s hot, or what hot really means. Sure, if all you want to do is eat sandwiches, that’s all you need. If you want to build an oven, it would be useful to know that lighting a fire under your sandwich makes it hotter. If you want to invent a microwave, you damn sure better understand that temperature is the average energy in moving microscopic particles, and that if you make those particles move more, the object gets hotter.
If you do ____ to a game, it makes it more asymmetric.
May 30th, 2008 at 4:58 am
Regarding the hedge concept, ok now I get it and I have to strongly reject it. I didn’t realize you were letting this system actually literally give out game wins. No one wants that. If Chibita shows up to a Virtua Fighter tournament, he will pick Lion, usually considered a low tier character. But Chibita’s amazing skill and knowledge of matchups has lead to him winning major international tournaments. He doesn’t want some out-of-game system to give him wins or take them away. He wants to overcome the handicap with player-skill and knowledge of matchups.
The only people who would favor this hedge system are non-competitive, bad players who see it as a chance to sometimes get a win without gameplay in cases where their real chance at a win would have been around 0%. I think I speak for the entire VF and Street Fighter communities when I say no one wants this system at all. Double blind is much better and uses the game itself to determine wins (and builds in some ability to affect your chances with a counter pick–but you still must use actual gameplay to get the win).
May 30th, 2008 at 6:49 am
Cody: Fairness and balance are the same thing… not sure why you said differentiating between them is a problem, because noone did differentiate between them.
To the rest of it… we’re not trying to make games more or less symmetric, so once again, the precise definition isn’t relevant. We’re trying to make games more balanced, and fortunately, we do we have much more than loose definitions to help us determine balance, and make qualified statements about it. It happens that the balance is more interesting in asymetric games, so the project here was to a grab a “bucket-ful”, to use the technical term, of asymetric games, for the express purpose of examining them for balance. We don’t want or need to examine their precise amount of asymetry, nor compare the amount of asymetry between the games in the bucket.
May 30th, 2008 at 9:11 pm
Sirlin, it’s funny, although not surprising, that you reject that idea out of hand. What are the things you advocate for? Reduced emphasis on execution? (clearly) Depth, aka interesting options? (the balancing I’m talking about only adds options, not removes them) Emphasis on Yomi? (check) Emphasis on assessment? (check) Essentially what you’re saying is that fighting game players want unbalanced games; you’re probably right.
As for the Chibita argument, if lion really is a low tier character, then chibita doesn’t have to worry about someone being able to hedge & use some so-called “out of game system” (it’s all a game) to take away his wins. As for giving him unearned wins, if he’s always playing lion, then no one with a brain is going to hedge against him and give him a free win. If the fighting aspect of the game really is balanced, people are just going to pick the characters they want, and play. However, if the fighting aspect of the game is unbalanced, the system I’m talking about is essentially self-healing, allowing the players to balance out the overall game.
Claytus, someone most certainly did try to differentiate between fairness and balance, namely Sirlin (in the interview above):
“maybe within balance we could talk about depth, and also fairness”
If balance is exactly the same thing as fairness (which you think it is . . .) then this statement is even more of an exercise in drawing meaningless distinctions than my suggestions that sirlin be more precise about symmetry.
May 30th, 2008 at 11:32 pm
I think you misunderstood him. He was implying that we should seperate out depth as a completely seperate issue from balance (even though they are related in some ways), and that, as such, I’d take it that the word “balance” should be used to describe just fairness, which I think you and Sirlin would agree with.
As for your hedge system, there’s lots of problems there. I said it before, you’re not altering game rules… you’re changing tournament rules. People who play fighting games *want* to play a fighting game. They don’t want their matches to end before the game ever started because of some ridiculous character selection system. Not to mention that even most bad matchups aren’t guaranteed to go to the stronger character, so a hedge really shouldn’t be worth an entire match. Also, you’re system still falls apart in a lot of places. It’s okay if you’re differentiating between chun as top tier and twelve as bottom… but what about the other ten characters… where is hedge supposed to fall precisely? Also note that tier lists are generally quite mutable, and some of the absolute top players with low tier characters still destroy players using top tier characters. I don’t think anyone facing ST’s T.Hawk in the hands of one of three players in the world with enough dexterity to pull off his safe throw loop really would still consider him low tier in the context of that match, even though he’s generally at the bottom of the list because the loop is so hard to learn.
May 31st, 2008 at 12:37 am
A “hedging system”? LOL! I take it that you don’t actually play fighting games competitively. Nobody wants to roshambo their way to the winners finals. I also hope your not a game designer. No offense, but things like this and your “payoff matrix” terminology seem like they’re destined for a research paper, instead of the ingredients of a successful game.
While Sirlin is doing a fantastic job(from what I can tell) of balancing ST, the truth is that the bulk of the population will never really truly appreciate these changes. He’s only in this position because ST is a fun and appealing title. The nebulous traits of “fun” and “appealing” are well outside of the definition of balanced. You can make the most balanced game ever, but if it’s boring or not aesthetically appealing, then it will be lost in the shuffle. Having a mathematically perfectly balanced game means approximately squat if it’s not appealing. And focusing to this kind of degree about terminology and such will only lead you astray.
Ideas like circumventing entire matches to a RPS-style meta-game in the hopes of adding balance to an existing game is just pure silly. If a game is broken beyond belief then there’s only three sane choices - (1) Play the game competitively using mostly/only the subset of “too good” options (2) Don’t play the game competitively at all (3) Patch the game to re-balance it. Adding meta-game options doesn’t fix the game itself. It just turns the players attention elsewhere. And if your goal was to make a fun and exciting game, then you just failed.
May 31st, 2008 at 1:36 am
Claytus: what’s to misunderstand? Depth is nowhere “within” balance, and fairness is only “within” balance in the sense that it IS balance. Perhaps he misspoke, but that’s not the same thing.
The idea that tournament rules are somehow not game rules seems rather silly to me. Rules are rules. If so-called “tournament” rules have an effect on the game, players are going to learn them and manipulate them. Why not make the meta-game explicit?
As for your so-called problems with hedging, I don’t think you were paying sufficient attention to what I wrote, especially since I already outlined some of the actual difficulties with the idea. “Shouldn’t be worth an entire match” - this actually helps make hedging a less enticing option, because it’s very dangerous; if you really don’t like the value of hedging, it’s possible to make it worth a fraction of a match ( ie reduced health bar for the next fight ), or to decrease the relative value of a match so that random hedging is less of a factor in the long run. “What about the other ten characters” - mathematically speaking, it doesn’t actually matter. Just order the characters, and put the dividing line in the middle, the game will be balanced. “tier lists are generally quite mutable” - I already said you can punt this to the players. Let the individual player decide the tier list that HE will be bound by to determine whether HIS hedge wins. This should definitely be public information, ideally just reflected in the order of the character select screen, and should probably stay constant at least for the duration of the contest.
Sweet Johnny V: I’m not extremely competitive, but I’ve played in a number of VF tournaments. I would have no problem trying this idea out. I think it would be a fun experiment, and would have a number of interesting side benefits (like needing to be able to play more than 1/2 the cast in order to be most competitive). And even though VF is a pretty balanced game, there are certain characters (such as lei) that I would personally rather see much less of, even if I know how to beat their basic tactics.
I don’t see a huge difference between “I know that he knows that knee is my best move, so I’ll do kenka hook to beat his mid reversal attempt” and “I know that he knows that lei is my best character, so I’ll pick lion to beat his hedging attempt.” The only difference is in the the number of times you have to guess correctly (so adjust the value of a match), and in the execution required (so make people input arbitrary just frame commands in order to hedge? c’mon).
May 31st, 2008 at 2:33 am
This hedging system is really terrible, and as I said before every single SF or VF player will tell you that. I’d expect every player of every competitive game to agree. Getting game wins for a paper, rock, scissors game outside of the fighting game is not balancing the fighting game. Fighting game players want to play the fighting game. You say they want an imbalanced game. I would put it differently. 1) They want to play the fighting game. 2) They want the fighting game balanced. You have given an option that sometimes has zero gameplay. It’s really ridiculous actually, kind of outside the realm of what I can even believe someone would argue for. I’d love to see the 100-0 poll results on this on shoryuken.com or virtuafighter.com.
Maybe my Chibita example wasn’t the best, but the point still stands that “winning” a fighting game with no fighting game gameplay is definitely bad. Even starting at half health or something is cumbersome, unworkable, and a bad idea anyway. After 14 years of Super Turbo, people still don’t know if T.Hawk is bottom tier or 2nd tier. So what percentage does he get? Answer: there should be no percentage, pick T.Hawk if you think you can win with him or if you think he counters a specific character somehow. But you will have to use actual gameplay to win.
Regarding those other terms, I meant that people are always asking if a game is “balanced.” Well let’s say the answer is no to some game. Then I’d want to know, “Do they think it’s not balanced because it’s unfair? Maybe there are 20 characters but only 2 are good. Or maybe player 1 has some huge bug that gives him advantage. But maybe it is fair, and they think it’s not balanced because it has no depth. It could be a game where all sides have exactly the same weapons, but there’s only one tactic so the game is stupid.”
People sometimes have a complaint about depth and other times about fairness but in both cases refer to the offending game as “unbalanced.” It seems that the word “balance” should take both into account if we are to match our meanings with how the word is really used.
My definition of balance from 8 years ago was: “A multiplayer game is balanced if a reasonably large number of options available to the player are viable–especially, but not limited to, during high-level play by expert players.” I implied that those options included BOTH “you can choose a lot of the characters” as well as “after character choices are made, there are a lot of options on how to actually play.” But that was so implicit that I think people could overlook it. So now I’m trying to be more clear that “balance” can refer to both the concepts of “here’s 17 characters that are all fair vs each other” and “Ryu vs. Zangief is deep and interesting with many effective options and techniques available to each player.”
May 31st, 2008 at 4:56 am
This “actual gameplay” thing is hilarious . . . let me get this straight: it’s “actual gameplay” if you use a joystick to indicate punch, guard, or throw, but it’s not “actual gameplay” if you use a joystick to indicate leifei, lion, or hedge? Why? Because execution really IS important to you?
Also, thanks for confirming that you were indeed trying to make an (untenable) distinction between balance and fairness. What you’re referring to by using “balanced” there is actually a much vaguer notion like “fun” or “interesting” or “people don’t complain about it”.
May 31st, 2008 at 5:19 am
cody: I guess Sirlin covered it, but now you’re the one being imprecise. Any game forum out there you will discover that people actually do the use the term balance when they mean “depth”. Since this is a term we care greatly about (unlike symmetry), Sirlin was defining more precisely, just like you wanted. The implication was always that balance = depth + fairness in that quote you mentioned. Not that balance is some 3rd value, which seems to be the way you read it.
As for your idea in relation to VF… you can’t honestly support a plan that requires everyone to learn half the cast, can you? Do you really have any idea what the relative depth of that game is per character. It’s astoundingly difficult just to learn a single character to a competent level, much less understand all the available matchups, and internalize all ~200 moves, and understand their uses, and be able to switch between them all at will depending on the situation. Having to up your match knowledge from 15 matches to nearly 80, and up your move memorization from ~200 to over 1000 would not only be ridiculously hard, but lower the amount of skill displayed in individual matches, since players would be able to devote less time to learning a single character.
As for “dividing the tiers in the middle”. I already tried to explain why that wouldn’t work well with T.Hawk. But look at ST… the tiers are generally based on the amount of good matchups a character has given your opponent chooses a random character. But most low tier characters do have *some* good matchups. You’re entire hedging system would be incredibly easy to exploit by simply choosing the lowest tier character that I know happens to have a good matchup against your strongest character. Therefore, I beat both your preferred character choice, and I beat your hedge attempt. You’re really putting too much weight on the tiers, they were never intended to fulfill an explicit purpose like your hedging system, and they’re too artificial to really do that well. They aren’t even consistent between games. ST tiers are based on amount of good matchups… VF tiers are generally based on the risk/reward of launchers, MvC2 tiers give you insight into how many bull**** infinites a character can pull off.
May 31st, 2008 at 6:06 am
cody, you are ridiculous. Yes, it’s actual gameplay to pit one player’s character against another using precise joystick motions and making very quick decisions at a fast pace. No, it’s not actual gameplay to choose the wrong character and lose without any fight. Maybe it would be real gameplay in a game that’s designed for that, but people play and watch fighting games for the fighting gameplay, not for character choice RPS.
Sirlin isn’t against execution in video games. Hell, that’s what makes them video games! What he’s against is games where the amount of execution is needlessly high - where the game about reacting quickly and making fast predictions has a needless barrier to entry that requires players to do a full circle with the joystick quickly enough so that his character doesn’t jump. This is an important distinction. Learn it.
June 1st, 2008 at 12:31 pm
This is almost as bad as the guy that was arguing about FPSs.
“This “actual gameplay” thing is hilarious . . . let me get this straight: it’s “actual gameplay” if you use a joystick to indicate punch, guard, or throw, but it’s not “actual gameplay” if you use a joystick to indicate leifei, lion, or hedge? Why? Because execution really IS important to you?”
This is retarded. You’re summerizing a game, round, or quarter of a round as a pure double blind pick, where that same game/round/quater of a round would comprise of many actions/choices/tactics. Atop that, fighting games are more than a series of guessing games, even if that is a very strong aspect.
When I’m at a tournament I want to see who is the best at whatever game we’re playing, not whos best at Game X + Hedge. Even if this was somehow a good idea and that everyone who plays these games actually thought this was an alright idea (no one does), how would it be implimented? Who would be the governing body? How would splits in the community over tier listing/hedge results effect the community? Not only is the idea bad, it’s cumbersome. It’s also not fun.
The hedge also waters down competition by giving the hedge any value whats so ever. I’d rather be able to focus on a a handful of very good options, rather then having to spread over a bunch of less interesting/good options in fear of the hedge. Artificial balance (and not really) at the cost of fun is just a flat out bad idea.
And again, by the fact that no one who plays fighters wants it (Sirlin is right on the ball on this one), the idea has no legs to stand on.
June 3rd, 2008 at 3:53 am
Claytus . . . wait, noone (sic) differentiated between balance and fairness, but now sirlin’s a genius for differentiating between them? Make up your mind. Ditto for making up your mind on VF being generally homogeneous vs VF characters being incredibly different and taking a lifetime to master.
Yes, I have a pretty good idea of what is involved in learning VF characters. The differences aren’t that big in most cases because of the near-universal jab / lowpunch / elbow / throw mechanics. You know that the “200 moves” thing is a crock, 20% of those are rising attacks whose differences have minimal impact on gameplay, and a large portion of the remainder are essentially worthless low-percentage moves. I have plenty of empirical evidence that good players have no problem winning with multiple characters; if you doubt that, go challenge denkai to a random select vs your Aoi.
Matter of fact, everyone had a blast with the random select tourney (which, you will note, is balancing that involves even less “gameplay” than what I’m suggesting) at socal this year, so much so that there have been multiple calls to do it again in NYC. So much for fighting game fans being uninterested in experiments.
As for trying to explain to me why hedging wouldn’t work, how about you do some actual math before jumping to conclusions? Your so-called “exploit” makes about as much sense as saying “I know your best move is knee, so I’ll just do rmid-reversal throw escape to beat your knee AND your throw” OMFG Broken! Nevermind that there’s another option ( pick the highest tier character that has a good matchup against your “low tier” character ) that you are going to have trouble with, just like your RTE will have trouble with a kenka hook to the face. In all balanced / unequal risk RPS, it is not a viable strategy to keep doing the same thing over and over, you have to mix it up. And no, there is no exploit, strictly speaking, when each player gets to control the tierlist his own hedges are decided by, as long as he doesn’t choose the dividing line (everyone except my char is toptier).
PoisionDagger, as far as I can tell, your point boils down to “Reaction time”. Ok, to the extent you can separate reaction time from execution, add that to my list of differences - 1. value of a single choice; 2 execution; 3 reaction time. There’s a pretty obvious solution to that - give people a split second to pick a char. Wow! Reaction time is a factor, now its just like the rest of the decisions in a fighting game.
KayinN, you’re complaining about the value of a single guess vs multiple guesses, which I already mentioned. The rest of your complaints are pretty straightforward to address, e.g. “how would it be impleminted?” Uhm, maybe, you know, PROGRAMMED INTO THE GAME? Or is setting a one-time preference ordering a list of 18 items when you don’t like the default tierlist too complicated for you?
Yes, I’m clearly being slightly silly by suggesting this, but if it really is such a horrible idea, shouldn’t you guys be able to come up with some counter-arguments that aren’t completely full of holes?
June 3rd, 2008 at 9:19 am
cody: I have no idea what your first paragraph is complaining about. They’re the same thing… you agree and we agree, but you’re accusing us of saying something else. Try not making up your arguments next time.
Also, I didn’t say skill doesn’t transfer at all. If you’re good at the game in general, you can pick any character, and rely on similar moves. But you aren’t going to win serious tournaments with any character. Learning stances and sabakis and so on takes some serious time investment. It just doesn’t transfer that well, especially for people who don’t want to learn every character. (Btw, I have played denkai, he’s awesome, and I probably won’t ever beat him. I’m not sure what that has to do with anything, though…) Also, you’re a moron if you think different rising attacks don’t matter. I sure as hell am going to pay attention to whether I’m doing a full circular sidekick or a linear midkick when I rise… because it’s a huge difference and often changes whether I hit someone, or they knock me straight back down again.
And regarding you’re new math about your hedging system… congratulations, you’re now clearly advocating learning every character instead of just half the characters as before. Sorry, but I think I’ll stick with the version of the game where I can play the way I want, and not be forced out of choosing a character I enjoy playing (btw, Aoi isn’t even close to top tier, so don’t bring up any of that shit), and instead have to choose based on some ridiculous system that gives wins without anyone actually playing.
June 4th, 2008 at 2:50 am
Hedging is dumb because it’s not fun. This is -game- design, not sure -math- design. All the math in the world can’t make a concept fun. Like Sirlin said, people want to play a game.
And to go on, this is limiting for people. Some people like playing one character. Some people play several characters. Some might argue this is a good thing, but it’s still going to irrate people. Especially on casual/lower end play. Which is another problem. If it’s built into the game, the hedging results and tier lists will seem very wrong to casual players — and very unfair. If you make it a tournament only mode, no one will use it. And again, all the logisticals/disagreements/split communities if it was taken seriously. This is a problem you can’t shrug off. People don’t agree on tier lists very often.
Hedging also doesn’t benefit balance or anything all that much. Top tier are still top tier… And their flaws and imbalances will be felt for an entire game, not one little double blind situation on the character select screen. Also I might not be getting the hedge system, but people would just keep picking upper tier characters. since they’re not ‘top tier’… And why would anyone bother picking low tier as long as they picked outside of the ‘top tier’? Is hedge neutral against mid range characters? If so, that removes a lot of the risk of hedging.
This all dances around the idea of actual balance. Fighting games are not just a series of guessing games. . If you’re going to demand something that definitely can’t be summed up up as execution, you have knowledge. Fighting games are complex, but a generally consistant entity. People like understanding this system and seeing how things interact. They like having real control, not being rail roaded from one guessing game to another. Even at the highest end of play, no one knows everything and some players can make up for some other failings by knowing more.
People also undervalue execution too much. Yeah, too much execution is bad, but being able to execute stratagy isn’t. My self? I’m a player who tries to guess as little as possible and with as little risk as possible when I do. Poking and spacing I suppose would get clumped under execution. Theres some guessing going on, but in reality I’m not going “I think he’s going to do this and that!” .. I’m saying “I think I can keep in this range better than you and know what moves counter yours better then you know what counters mine.”
I’m not some alien, a lot of people play like this. I like executing stratagy, as do a lot of people.
TO SUMMERIZE/In Addition
* Hedging doesn’t benefit balance. Either it’s too risky and thusly rarely actually stops top tier characters or it’s too safe and most character choices would generally stay within a safe range. This limits the amount of matchups and character interactions you see during the part of the game that actually matters — you know, the part that you wanna play.
* Also if we’re going to gather all this tier data to allow the game to have an updated hedging system, couldn’t we like use it to UPDATE AND BALANCE THE FUCKING GAME OH MY GOD EWTSUFT+%@)@#+%$@#U+%#@UW+U)T+)U@T+)UT@+U#TU+T#@+UT#@R
* Hedging isn’t fun…. Unless you’re some mutant. Also reduces the amount of actual game play time.
* Fighting Game players seem to naturally hate it. So either it has to be in the game at all times (in which case casual players get boned), or it has to be in a tournament mode (in which case no one uses it and all these features to allow and modify tier listings are money you basically set on fire and shit on.
* A lot of people don’t want to be forced into a high risk guess without their opponent earning it.
* A lot of people don’t want to be afraid to pick whatever character they actually enjoy playin due to some stupid hedge system.
* So really, all hedging is, is some external annoyance outside the realm of the part of the game people actually want to play that doesn’t benefit anything besides adding some silly forced guess that I guess, theoretically, contributes to ‘mind games’ in an insignificient way at the expense of pissing people off.
So before going “lol holey argument blahblahblah”, why don’t you try to explain why anyone would fucking want this abortion of a thought when it has little game play merit and it’ll just piss people off?
tl;dr
Hedge system is bad and you’re a bad person for suggesting it. :(
June 4th, 2008 at 9:56 pm
Claytus, I’m not making up anything.
First you said:
“Fairness and balance are the same thing… not sure why you said differentiating between them is a problem, because noone did differentiate between them.”
then you said
“Sirlin was defining more precisely, just like you wanted. The implication was always that balance = depth + fairness”
If fairness is EXACTLY THE SAME THING as balance, the only way this second statement could be true is if depth is COMPLETELY VOID OF VALUE. In other words, stop changing your mind.
The point of playing denkai is that he _can_ play a lot of characters at a competitive level (he’s not alone in that, he was just the example that came to mind).
I’ve played you, and I have serious trouble believing that you know all rising attack properties by heart. If you do, you probably should have spent your time learning something else.
Regarding your new math crack, I originally said “play more than 1/2 the cast”. This is accurate, because it ensures that no matter what tier list your opponent is using, you can play at least one low tier character competently in order to beat his hedging attempts and still have a chance if he picks a character. The dividing line is in the middle, remember?
KayinN , it’s pretty obvious that this is all you had to say:
“I might not be getting the hedge system”
Hint: you don’t have to gather data to have an updated hedging system if the players are in control of the tierlist . . .
June 4th, 2008 at 11:38 pm
Even if KayiN missed the fact that hedging works fine with personal tier lists (ie, each person makes their own tier list), everything else he said is true.
Hedging is a terrible system that nobody wants. Just because something is mathematically correct, doesn’t mean it’s good or fun. Hell, “players switch characters for half the game” works just as well as hedging, yet it’s also a stupid idea for the same reasons.
June 5th, 2008 at 12:03 am
Cody: It’s a disambiguation. We use balance for fairness, however some people use balance when talking about depth. It’s an english word, not a math formula… they don’t always have exact and/or singular meanings, deal with it.
And again, Denkai is not a fair example. You’re argument seems to be “Well, I can prove one of the top players already learned enough characters for hedging to work, so it’s a valid rule”. Well, no, it isn’t… the system only works if you can get every player to learn that many characters. You’re creating an additional barrier to entry to “fix” a game that isn’t even really unbalanced in the first place. The fact that all the average players posting here are rejecting your idea should be incredibly worrisome to you. Also, I never claimed that I know all rising attack properties by heart. I do, however, know almost all of Aoi’s, and a great deal of the 4-5 other characters that are actually used by local players. Out of curiosity, who are you on XBL?
June 5th, 2008 at 2:58 am
PoisionDagger, not everything else kayinN said is true, most notably “Hedging doesn’t benefit balance”, seeing as how it is by definition balanced.
Claytus, some people think mexico is in south america, too. Other people think nuke-u-lar is a word. Are they just as right as sirlin or others that think depth has anything to do with balance? If so, have fun with that. Might as well just say that ST is a perfectly balanced game. Oh wait . . . .
As for the barrier to entry, what I’m talking about actually lowers it - if you’re just playing casually, pick a character and mash. You don’t have to worry as much about choosing the “wrong” character. If your friend is doing some BS with a toptier character that you don’t know how to counter with your character . . . . make him switch.
Does it change the landscape for high-level play? Yes. I’d personally rather watch looser play by people who understand nearly 100% of the game than tighter play by someone who has mastered the dominant tactic for the dominant character.
As for denkai being an anomaly, of course he’s good, but there are other players with less skill that still use multiple characters effectively. I’m a shitty player, but I’ll be happy to play you with a character of your choice; if you think about who else has been patient enough to respond to 20+ messages of your hardheadedness, it should be pretty obvious who I am.
June 5th, 2008 at 3:02 am
…. Personal tier lists? So we can atleast get rid of tier list arguments and replace it with everyone bringing a memory card or something (or wasting time to input their hedge list — whatever, these problems pale in comparison to the ass that is hedging). Now you can do stuff like pick the top local players top 3 characters or just the most popular characters in your area (regardless of tiers) or various stuff. Is that broken? Possibly, but not certainly. But it’s disruptive. There would be no safe characters, no matter how fair they are.
Even then, THIS DOESN’T BALANCE THE GAME. Playing RPS before a game of GO does not balance Black and White. You have a risky magic bullet. If you don’t use it, you’re still in a bullshit matchup for the actual part of game play. The fact that it can be pointed at any character makes it even worse and forces us to know as many characters as possible. And what about the Japanese, who play exclusively with one character?
No one wants this. People want to play the actual game. As much as you’d like to argue that ‘this is okay it’s part of the game’ no one will see it as such.
Also, you have to realize the burden of proof with with you, not us. You are suggesting idea that, to put lightly, has caused some people to vomit in their mouth. The idea is somewhat convoluted and barely contributes to balance (I could argue it raises the chances for low vs high tier matches, making it more imbalanced). You’re suggesting something that isn’t fun. When games are concerned, I think it’s logical, that the person suggesting the most potentially harmful method has to try and win people over, not vice versa. There are no argumentive holes when the fact “It’s not fun and no one wants it” is among the arguments. You have to convince people hedging is worth the cost of fun.
Good luck there. This idea is stillborne at best and had been argued to death already days ago. The fact you see ‘holes’ still is just a sign you’ve lost sight of whats important in game design.
June 5th, 2008 at 3:10 am
Dead Cody,
Alright what is this about a hedging system? You really want to play rock paper scissors before a match and let something so arbitrary determine a win… in a FIGHTING GAME TOURNAMENT?
Are you really thinking this through? Why would you force someone to play some other game before they get to play the game they want? Going by your very flawed logic it would be perfectly fine to give a football team the first touch down at the start of the game simply because the other team is better, you chose rock and beat their scissor.
Instead of trying to prove your point mathematically why don’t you prove it with real life applications. Tell me one competitive fighting game player who would like to forgo playing a match and determine a winner through some outside method.
June 5th, 2008 at 4:20 am
Cody: Your ‘hedge’ system is based on rock paper scissors, where rock beats scissors, scissors beats paper, and paper beats rock. No matter how I may try, if I always pick rock, I will always lose to paper. I will also never lose to scissors.
There is a problem in using this as a base for your hedge system.
Hedge beats top tier 100% of the time. Have a nice day wasting time on practicing your crossup EX air tatsu, ken fans.
Low Tier beats hedge 100% of the time. “You thought I was some yun lamer, huh? Sean all the way!”
Top Tier beats Low Tier ???% “But I picked Chun li! Your hugo is bullcrap!”
Essentially, using your hedge system makes it so there isn’t any reason to pick or even learn high tiers. If I pick High Tier and you pick hedge, I lose, and if I pick High and you pick low I still have a 20-45% chance of losing.
To further elaborate, if every player picked each option evenly (1/3 of players picking each hedge, low, and top), The players that pick hedge would win 50% (0% vs Low, 100% vs High), Players that pick low would win 60-72.5% (based on a 20-45% matchup), and players that pick high would only win 40% at best (0% vs hedge, and 80% vs low tiers at best).
All the Hedge System does is make the low tier characters the new top tier. I guess alex just made a comback, huh?
June 5th, 2008 at 7:11 am
cody: (or should I say KoD?) Let’s try this definition thing again, without you inventing examples that are literally false. Pretend you have some friend who just flew in the country, and has never spoken any English before. You’re walking down the street of a major city, and walk past some real gangster-types. Just as you walk past some mustang with custom parts and a great paint job comes around the corner and drives past you, the kind of car everyone wishes they had, but most people never actually get. Right as it drives past all these gangsters point and stare at the gear and go “Wow, dude, that’s bad!” (a wonderfully informative phrase that can be heard in any high school in america). You’re friend turns to you and says “What does the word “bad” mean?”. What are you going to tell him? It’s the same principle at work here.
June 5th, 2008 at 7:12 am
*er, car, not gear… not sure where that typo came from
June 5th, 2008 at 10:48 am
Um, I think someone mentioned this before, but isn’t a hedge system entirely unnecessary in a game that allows mirror matches? The double blind system for choosing characters is already fair. Both sides have access to the same options, and each option has the same reward for both sides.
June 5th, 2008 at 3:51 pm
Well yes. Apparently though, thats not good enough. We have to basically stick in some 10/0 and 0/10 matchups to ACTUALLY be balanced. Also we gotta force people to play more characters instead of reaching mastery and pushing the boundries of one. (And also piss all the players off in general)
June 5th, 2008 at 10:03 pm
Cody, no one has to poke holes in your argument when it’s ridiculous to start with. If I tell you I can make myself invisible but only if no one’s looking and you call me on the obvious bullshit it’s not very credible for me to say “YEAH WELL YOU CAN’T PROVE IT!”
You’re suggesting a pretty extreme solution to a problem that only arguably exists, and the worst part is it doesn’t even sound fun.
June 6th, 2008 at 2:37 am
You want a balanced hedge option? An easy thing to do, if we treat ‘hedge’ like a character.
If we were to design a character (i.e., hedge) so that it won vs the top tier just as much as it lost to the low tier, it would completely balance the game.
I’ll use guilty gear as an example:
Potemkin wins vs Slayer 60% of the time
Slayer wins vs Jam 60% of the time
Jam wins vs Potemkin 60% of the time
Essentially speaking, the only way to ensure properly equal payoff matrices for each and every player, regardless of the character they choose would be to ensure that each character has just as many bad matchups as they have good matchup, and by roughly the wame win/loss ratio (for every 60% favorable matchup a char has, they should have a 40% unfavorable matchup).
As a minor quirk, the only way to ensure this (aside from the obvious patching and re-releasing of various versions of the same game) would be to make certain that the availiable number of characters in the game are odd, not even. This makes it easier to balance a character in such a way that half of it’s matchups are favorable and half arent.
Of course, purposely designing a character to have specific matchups favorable and unfavorable is a difficult feat in itself.
June 6th, 2008 at 11:46 am
Counter matchups are already a bad way to balance a game. It would in no way ‘completely balances the game’.
First turn, you lose to top tier. you pick hedge and win. They pick low tier and win. Huzzah, balance! Only not.
Counter picking is not a great way to balance a game, especially polarized counter picks. They can make stuff interesting (honda), but a game designed on counter matches is not an ideally balanced game. You can’t think of fighting game balance in such macro terms, especially since you get only one double blind pick at the beginning of the game.
This sort of balance works in team construction games, like pokemon or star control, but not in fighters.
Not only with the above problem — not everyone counter picks. Going by the Japanese school of thought, it can be better to know your matchups very well rather then counter pick with a less familar character and then get re-counter picked by your opponent into a matchup you don’t know. A high counter picking game forces a player to know and play more characters (as opposed to making it a choice of taste). It also degrades the character selection screen into a game of RPS with more match influence then it should.
The other huge and gaping issue, is that designers do not possess the ability to create a game with such precise counter character. Designers don’t even KNOW who the top and bottom tier characters are going to be. If they could manage this they could instead make a superior game with all matchups with in the 4/6 - 5/5 range.
Balance has to be achieved in game, not on the character select screens. Characters need ot have sufficient options to survive and project them selves while sitll generating interesting move interactions. You need less math in your designs. It doesn’t work that way. Theres too many variables and emergant factors to handle. This is why fighting games are balanced through testing and tweaking, not mathmatics.
Sirlin himself has said he designs with very little math. Not because it’s not his style (He was an MIT Math major), but because it doesn’t work that way. He can also attest to the blood sweat and tears required in the laborous balancing process.
June 6th, 2008 at 4:27 pm
KayinN is 100% right on all counts here. Well done.
June 7th, 2008 at 1:31 am
Cody is a world class scrub. I refuse to believe he is anything more than a troll. The hedge system is an insult to fighting games and every competitive endeavor ever conceived in the history of man. I’ve never seen anyone put so many fancy words into describing something so incredibly stupid. Cody, I honestly hope you skate into an aids tree and get beaten by a cancer stick. You are a very annoying and arrogant little man. I cannot properly put into words how much I feel the need to bring a swift vengeance upon your face with a nail gun for wasting my time with your posts. I will seriously need a minute before I can continue to write any further.
*deep breath*
*exhale*
You’re like Bruno Manheim at a nuclear plant asking dark seid to make him king. So you are, the king of fools. As you climb onto your little motor boat of ignorance you will never escape the atomic boom of your stupidity. A tremendous wave that represents the inanity of your arguement will collapse above your head under the weight of its own fallacy. The result will leave you washed out in the radioactive shores of your mind as you receive sweet release from your twisted, corporal form.
So yeah, eat shit and die.
June 7th, 2008 at 2:09 am
I disagree entirely with Cody, but that was just unnecessary, Higher-Jin.
June 7th, 2008 at 2:38 am
It was completely neccesary.
June 8th, 2008 at 5:16 am
Sweet, do I get a sticker? :)
Ah, now if only I could write better.
June 15th, 2008 at 2:15 pm
I like Sirlin on camera. :D
I was playing some “Super Turbo” as I’ve heard you hardcores call it, and as far as your “donut” casual vs hardcore thing, I could never figure out the special moves in fighting games (much less the “super moves”). I’ll play a hour or more of Samurai Showdown or something and have a full super bar but never know how to use it.
Are the special inputs for fighting games (a genre I don’t particularly enjoy) supposed to be some kind of secret?
For the Super Smash Brothers series (my favorite of the genre!) all the characters have (basically) the same special inputs, and there’s a little “How To Play” demo during the attract mode. What keeps my friends from playing the game is actually all the clutter and difficulty of tracking yourself on the screen.
For the triumph of game design that is Devil May Cry 4 (the video game I’ve spent the most hours playing), there’s the opening tutorial as well as forcing me to choose the special moves I want. What’s left up for me to explore on my own is the insane depth of the combo system.
Am I missing out on some fun when a game like Tekken # shows me an in-game command list? Because otherwise I sorta stand in the corner with a character and try random things to see if something “special” happens. Perhaps it’s more than adequate for novices to button-mash each other to death over and over, and I’m at some step above at since I’m wanting to learn a character’s full moveset.
June 16th, 2008 at 5:06 am
I think an ingame move list should be a requisite to modern fighting games. It was cute when special moves were “secret” in sf1, but really I just feel that if you want people to play the game competitively against each other then they should know all their options.
I also dislike tekken’s 1 million moves per character ratio. You have to try each one of them out and test out their properties to see which 3-4 is are the best and then just use those, when they really could’ve just had those 3-4 moves to begin with and left the rest of the crap out. The casual gamers usually have the complaint that playing a fighting game feels like taking a test, and with tekken’s mile long move lists I can definitely sympathize.
I really like Sirlin’s approach to balancing games. Easy to learn, hard to master. Just like chess, a game that’s been played and respected since the middle ages. It’s insane to think that the very people the game pieces were based upon actually played the game themselves. Now that’s longevity. If street fighter can attain even a fraction of that lasting power then I’d say we are a part of something very special indeed.