Designing Yomi
SirlinIn 1824, Sadi Carnot developed the theory of the reversible heat engine, the most efficient engine nature will allow. In 2005, I set out to develop the best game of paper, rock, scissors that nature will allow. That is, a competitive game based on the simple concept of paper, rock, scissors, and with the minimum amount of extra trappings needed to make the game deep and skill-testing. Furthermore, it had to test only the skills that make competitive games interesting. That means no joystick dexterity required, no aiming of a cross-hair, no physical reflexes, no memorization of book-openings, and no leveling up. As I covered in my book, Playing to Win, I think the two most interesting skills—the skills the lie at the heart of competitive gaming—are Yomi and Valuation.
| I give you my latest invention: the Sirlin Reversible Heat...er...I mean Yomi: Fighting Card Game. |
Yomi and Valuation
Yomi is the Japanese word for “reading,” as in reading the mind of the opponent, and also the name of my game. Valuation is my term for your ability to determine the relative values of pieces or moves in a game. Specifically, this means your ability to know the changing values of pieces during a game. For example, a pawn is generally considered to be worth one point in Chess, while a bishop is worth three, but given a particular board situation, these values will change. In so many games, the ability to read the mind of the opponent and the ability to know how much a piece or move is worth right now are the keys to success.
What if a game were designed to test only those two things? Everything else could be discarded to make the game as simple and accessible as possible. Competitive games tend to shut out most of the population because they tack on extra skill tests. Street Fighter, for example, requires so much dexterity with a joystick that most potential players are not able to experience the interesting mental challenges at the core of the game. I want to extract those great mind-games and offer them to everyone, not just the few who can pass the other extraneous tests.
Paper, Rock, Scissors in Other Games
Even these guys play rock, paper, scissors.Every competitive game I can think of has some conceptual tie to paper, rock, scissors. For a game to be interesting, there have to be different options, and the opponent has to be able to counter those options if he knows what you will do. Sometimes this means guessing whether Zerg will ground rush you or go for an air attack in StarCraft. Sometimes it’s guessing whether to go grenade (paper), rifle (rock), or melee (scissors) in Halo. Sometimes it’s guessing whether to attack (paper), throw (rock), or block (scissors) in Street Fighter.
Timing is also a key component here. Street Fighter, for example, appears to be a game of complete information. At any given moment, you can see exactly what both characters are doing and what their resources and options are. And yet if this were actually true, it would not be much of a game. Fighting games only work at all because of their double-blind nature: at the moment when you do a move, you usually do NOT know what the opponent is doing exactly. Human perception is only so fast, so you’re acting on information that’s a few sixtieths of a second old. At the exact moment you jump, you do not know if the opponent threw a fireball or not unless he threw it several frames ago. Most of your moves are essentially made in a double-blind situation, though I don’t think most players really understand that. This is true in other genres as well, though it doesn’t often happen on such a short time-scale as in fighting games.
Paper, rock, scissors already is double-blind, of course. At the moment you choose rock, you do not know what your opponent chose, except from an informed guess based on his history or tendencies. Even though paper, rock, scissors does offer different options (three options to be exact), a way to counter each one, and the nice double-blind structure that makes it all work, it’s still not very interesting to most people. (I should note it has a weirdly large following at worldrps.com, though.) There isn’t enough information to go on to really read someone’s mind. There isn’t any Valuation-testing at all because everything is worth the same amount.
The Two Secret Sauces to Improve Paper, Rock, Scissors
One of the secret sauces I used to spice up RPS.To spice things up, we need to add two varieties of secret sauce. The first one I covered years ago in another article about paper, rock, scissors: different payoffs for each option. If winning with rock is worth 10 points, while winning with scissors is worth 3 and paper only 1, then I have some opportunity to read your mind. Are you type of person who will go for the 10, even though you know I know about it? Are you the type who will sneak in a bunch of single point papers because you know only a fool would ever counter paper? I can at least start to measure you and make more educated guesses about your behavior.
The property of unequal payoffs isn’t enough though. Mathematically, you can’t do better than randomly choosing between rock, scissors, and paper in a 10:3:1 ratio. That said, it’s still a lot better than equal payoffs because at least human nature starts to come into play. But we can do better.
The second secret sauce to making paper, rock, scissors deep is unclear payoffs. What if we all know that rock is worth the most, but it’s not exactly clear whether it’s worth 5 times more than paper or 10 times more? Now following the 10:3:1 randomization strategy mentioned above is not as important as actually understanding the relative value of the pieces and how they change over time—and that’s Valuation. Part of the skill we want to test is the ability to make on-the-fly judgments about what things are really worth. Ideally these judgments are so complicated that you must use your intuition to solve them and knowing the exact answer is virtually impossible. Incidentally, using intuition rather than math to make a decision in a game makes the game more “fun” anyway.
This guy got a clear payoff. Will Yomi tournaments be this big?
The Yomi: Fighting Card Game System
Now we need a system that is so complicated that its exact paper-rock-scissors payoffs can’t even be calculated, and yet so simple that anyone could play it. It sounds like a tall order, but here’s how I did it.
Each player starts with one deck that represents his character. These are fixed (non-customizeable) decks with 52 poker-type cards, 2 jokers, and one character card. You could use these decks to play poker, but the cards have extra information on them for the Yomi game. The numbered cards represent normal attacks, the face cards are special attacks, the aces are super attacks, and the jokers are “bursts” used to get out of trouble.
;)
Established Conventions Convey Information Quickly
I deliberately used the conventions of a poker deck in order to instantly convey a lot of information to the player. When you see a queen that is a Dragon Punch card, you know right away that that there are three more of them in your deck. You also immediately know how likely you are to draw a super move (there are four aces in your deck). You probably also have at least some intuition about how often pairs and straights come up if you’ve played any other cards games at all. Because I need to convey so much information and make it seem natural, it was very useful to piggyback onto a structure everyone already knows. (Two of them, in fact! Yomi: Fighting Card Game contains Street Fighter moves/characters as one bundle of information that a lot of people are vaguely familiar with as well as the poker structure that almost everyone is familiar with.)
Double-blind Decisions
The core of the game is that each turn, each player plays one card face down, so that his choice is double-blind with the opponent. Each card has two options on it, so the orientation of the face down card does matter. The side you orient closest to your opponent is the move you chose. Both cards are revealed simultaneously to see who won the exchange. Attacks beat throws; throws beat blocks and dodges; and blocks and dodges beat attacks. Different things happen depending on which option you win with—this is where the unequal and unclear payoffs come in.
Attack, Throw, Dodge, and Block Have Unequal Payoffs
1) Attack
Usually, the best way to deal damage is to win with an attack. (If you both attack at the same time, the faster one wins). When you win with an attack, you can then perform a combo by playing more cards from your hand. There are rules for combos, and part of the fun of the game is saving up for big combos. Normal attacks can be chained together, but only in increasing sequential order like a straight in poker. Normal moves also can also combo into specials or supers. Each character (deck) has a “combo limit” which limits how big a combo can be each turn. The specifics of the combo system aren’t important to go into right now, and they’re written on the cards anyway.
On a side note, there is tension between saving up straights (for chain combos) and saving up pairs. At the end of each turn, you have the option of discarding pairs, three-of-a-kinds, or four-of-a-kinds to search your deck or discard pile for either one, two, or three aces respectively. Is it worth it to break up a pair of fives to use one in a chain combo instead of trading in the pair for an ace? The answer is unclear and depends on a lot of things about the current situation.
2) Throw
Winning with a throw is very similar to winning with an attack, except that your resulting combo is usually not as good because of the way the stats on the throws are designed.
3) Dodge
Winning with a dodge lets you avoid an incoming attack completely, and then lets you hit back with any single attack. You can’t do combos though, so do your strongest attack, hopefully a super.
4) Block
Winning with a block lets you return your block card to your hand and draw an additional card. Note that you get to return the block card to your hand if the opponent attacks, blocks, or dodges, and you only have to discard it if your block gets thrown.
Unclear Payoffs
As I explained earlier, the payoffs are unclear. How good is drawing an extra card when you block an attack? It depends how many cards you have and somewhat on how many cards the opponent has. How good is winning with an attack? It depends on how big of a combo you can do. How bad is it if you try to throw but get hit by an attack? It depends on what kind of combo your opponent has built up, and you only have indirect information about that. Would it be twice as bad to get hit by his combo compared to letting him draw a card? Three times? One half times?
Players Naturally Develop a Pattern
These girls are somewhat related to what we're talking about here.Even though the payoffs are unclear, a definite pattern develops in how people tend to play. This exactly the property we want because it sets up the mind-games perfectly. If you are low on cards, I know you want to block to draw more, and you know I know. If I just traded in several pairs for aces, you know I’m itching to dodge so I can hit back with a super attack, and I know you know. Will you follow “the script” and do the obvious thing? Or will you go to the next Yomi layer and do the counter to the obvious thing? Or to the Yomi layer after that and counter the counter? These types of mind-games occur in just about every worthwhile competitive game, and the Yomi card game lets you practice these situations and practice reading opponents.
Special Abilities Make Payoffs More Unclear
There are more nuances. Each character has a special ability on his character card as well as two more special abilities that appears in the deck. For example, Ken's character ability lets him pay 3 life (starting life: 90) for each attack card he played this turn that he'd like to return to his hand. This is yet another Valuation test, constantly asking you how much you think each card is worth. Also, each of Ken’s 7s can be used for his “crossup” ability that gives him a 50% chance for his attacks to beat a block.
These abilities make computing the exact payoffs even more difficult because they manipulate so many different kinds of things about the game. Some improve the quality of cards in your hand, others make the opponent discard, others increase the speed of your attacks, others let you teleport out of a bad guess, and so on.
I could have put special abilities on every card, but I don’t want the game to be overwhelming. Even with just one ability on your character card and two others on eight cards in your deck, there is enough going on to make it very unclear what the smartest thing to do at any given time is, while keeping the game easy to play. Testers of the game often argue about what’s smart to do in various situations, and then I settle the debates by beating them without calculating anything at all, and just guessing right every time. Just kidding (sort of).
Conclusion
That’s the overview. It’s a game of paper, rock, scissors with unclear and unequal payoffs that tests Yomi and Valuation skills. Because the game is so boiled down to these essentials, I think it’s an interesting way to learn what these two skills are really all about. These are the two fundamental skills of competitive games, so I hope that whatever you learn from playing Yomi will help you in all your competitive gaming exploits adventures.
NOTE: The Yomi card game is still under development and not commercially available yet.
--Sirlin
Reader Comments (3)
If you have some time, you should try Shazamm (http://www.shazamm.info/). You can read the rules in English. You may try the game against other people online on http://www.boiteajeux.net/ although it's only in French for now so you may have some trouble.
It's a game where Yomi definitely plays a huge part. Valuation too, as the value of cards is unclear, and totally depends on the current situation. It's a bit simple, maybe, but if you want to make the best Yomi+Valuation testing game, you'll want to discover Shazamm :)
I'm really looking forward for Yomi :)
As a competitive rock-paper-scissors player, I agree with this article that unequal pay-offs are what makes the game become really interesting. However I think that 'pure RPS' (i.e. the basic game) can still have unequal pay-offs. It's just that the concept of 'pay-off' needs to be expanded to the physical effort and psychological elements of choosing a particular throw. Or to turn it around, the pay-off for all throws is equal but one throw may be harder or easier than another throw - unequal costs that don't affect the result of a game but by affecting the choices, do allow strategy to come into the game.
Example 1: It's easier to throw rock when you feel under pressure (typically the beginning of a match or when it's a 'next point wins' moment). You're tense, and if you're clenching your fist at all, rock is a very natural throw. Also it involves least technique/effort compared to turning your hand to throw Paper. Of course, in competitive play other players may also know the tendency of beginners to throw Rock at these moments, which is where Yomi comes in.
Example 2: You tell your opponent, "I'm going to throw Paper." Now they may be considering how much you value winning honestly - would you want to win with Paper and keep the moral high ground with an "I told you so"? Or perhaps you're a scoundrel who'd gain more satisfaction from winning with trickery by throwing Rock?
Example 3: In a best-of-five match, your opponent just lost the first two throws to you, both times with Paper to your Scissors. Now, while losing the next throw means they lose the match, for some opponents losing with Paper carries an extra 'sting', because they feel they'd look more stupid losing with the same throw three times in a row. For this type of player, Rock or Scissors are psychologically 'easier' choices than Paper for the next throw.
Example 4: Players who are trying to play 'randomly' often tend to switch throws more than the 33% of the time they should do (on average) if they are truly playing randomly. Again, at the psychological level, switching 'feels' more random than repeating, so repeating the last throw is a 'harder' choice to make since they feel it goes against what they are trying to do.
The fun and skill in RPS comes not just from noticing these subtle patterns in how players value their throws, but also trying to engineer differences in valuation (such as example 2 above), and cause them to misread your own preferences. Certainly, the edge that you can gain from skill is not dramatic (though can still be telling in the long run - one reason why many serious players prefer to play first-to-10 or other long-format matches rather than straight best-of-3 matches), but these differences in valuation do exist, and the search for subtle advantages of this kind - to be able to sense your opponent's fluctuating valuations (and - Yomi again - to sense what their read of your own valuations is) is a large part of what keeps competitive RPS players interested in the sport.
Despite recently hating on Kongai, this game sounds quite fun and I'm itching to play it.
The difference is that I think a game needs to have more fun than "anti-fun"[1] to be an interesting, enjoyable game. After you get that worked out, THEN you work on attaining perfect balance, or the closest thing to it.
Yomi combat options sound similar to the combat options of Lunch Money, except with more careful, deliberate balance informing its design. Awesome!
[1] Anti-fun - "Anti-fun is the negative experience your opponents feel when you do something that prevents them from 'playing their game' or doing activities they consider fun. ... a strong mana burn is NOT desirable -- if you drain someone to 0 you feel kinda good, and they feel TERRIBLE -- so the anti-fun is exceeded by the fun"
http://na.leagueoflegends.com/board/showthread.php?t=293417