It was a great honor to have the opportunity to design the balance changes and help oversee the new art for the High Definition version of Capcom’s Puzzle Fighter. Game developers do not often talk about why specific changes were made, so I thought the fans (and other game developers) might be interested in the reasoning behind these balance changes.

My high level goal was to change as little as possible because the original game is very well-designed and fun. I would even go as far as saying it’s the single best competitive puzzle game. That said, it does have a problem: only 2 of the 11 characters are playable in a serious match. I did not want to get into any deep changes with the underlying formulas or rules because—character balance aside—the game is very good already. So, I restricted myself to changing only three variables: 1) drop patterns, 2) damage scaling per character, and 3) the “diamond trick.”

Drop Patterns from the original Puzzle Fighter
		Akuma (deals 70%)
Ryu	Chun Li	Dan	Sakura	Ken
Morrigan	Hsien-Ko	Devilot (deals 70%)	Donovan	Felicia

New, Rebalanced Puzzle Fighter Drop Patterns
		Akuma (deals 100% takes 120%)
Ryu	Chun Li (deals 120%)	Dan	Sakura	Ken
Morrigan	Hsien-Ko	Devilot (deals 85%)	Donovan	Felicia

Drop Patterns

An early drop pattern

A drop pattern is the pattern of gems you send to the opponent's side after you break gems on your side. Drop patterns help differentiate the characters, but they also serve a useful overall purpose: they allow for defense. Because you know the pattern of colors that the opponent is capable of dropping on your side, it’s possible to build your side so that enemy attacks sometimes help you. If all characters sent random patterns of blocks, it would be very difficult to build up large Power Gems on your side without having them covered up all the time. So we want the drop patterns to be predictable and not too mixed-up.

In the original game, Ken and Donovan were the best because they had the “least bad” drop patterns. Actually, Akuma and Devilot had the best drop patterns, but they also only dealt 70% of normal damage, a handicap that made them the worst characters. Because I knew the game was already fun when playing Ken vs. Donovan (that’s all anyone ever played), I thought it would be good to balance the game around their power-level. For this reason, Ken and Donovan are exactly the same in the rebalanced version.

	► ►
Old Ken		New Ken (Unchanged)
	► ►
Old Donovan		New Donovan (Unchanged)

I also wanted to keep Dan unchanged. He’s a joke character who can only send red gems, so he’s supposed to be the worst. I liked the idea that Akuma and Devilot have the best drop patterns in the game but with a drawback. It’s just that the drawback of dealing only 70% damage was too severe. The new Devilot deals 85% damage (better than 70%!) while the new Akuma’s damage went all the way up to 100% (normal damage). I thought that it would be a more fitting drawback (based on his Street Fighter appearances) if he takes 20% more damage than the other characters.

	► ►
Old Dan		New Dan (Unchanged)
	► ►
Old Akuma (Deals 70%)		New Akuma (Unchanged Pattern Deals 100%, Takes 120%)
	► ►
Old Devilot (Deals 70%)		New Devilot (Unchanged Pattern, Deals 85%)

I didn’t want to tinker with damage scaling numbers for very many characters, so there is only one character besides Akuma and Devilot with a damage adjustment: Chun Li. I thought it would be interesting to make one character that’s the reverse of Akuma: instead of having the best pattern and taking more damage, Chun Li has the worst pattern but deals 20% extra damage. I hope this makes Chun Li a tempting character to play because she has the ability to do so much damage, but her terrible pattern can really backfire against you sometimes.

	► ►
Old Chun Li		New Chun Li (Unchanged Pattern, deals 120%)

So far, that’s 6 of the 11 characters, and I haven’t changed a single drop pattern yet! The remaining 5 characters needed new drop patterns, though. Even though these 5 drop patterns needed updates, I wanted to keep the general feel of each one for nostalgia’s sake. Sakura and Felicia had similar patterns, and both suffered from only being able to drop green in column 1 and yellow in column 6. This is a huge disadvantage for both of them, so I mixed up the greens and yellows in each of their column 1 and 6. I mixed up Felicia’s the most (alternating green and yellow each row) because her red/blue middle pattern (power gems!) is worse than Sakura’s red/blue middle pattern (horizontal rows, which we know from Ken are powerful).

Felicia gave me some trouble.

After playtesting a while, I decided that even the improved Felicia was too weak. Sending power gems to the opponent was just too much of a handicap, so I changed the red and blue part to a Tetris configuration with interlocking “L” pieces. There were a few possible ways to arrange the Tetris pieces, and I chose the most powerful one, such that it’s kind of hard to build against it without filling up your own column 4. (And remember kids, don’t ever fill up your column 4 because that’s the only column that can make you lose the game.)

	► ►
Old Sakura		New Sakura
	► ►
Old Felicia		New Felicia

The red Power Gem in the middle of Morrigan’s pattern was very bad for her, so I replaced it with a slightly better set of interlocking “L” pieces. This change makes the bottom two rows of her drop pattern at least close to the power level of Donovan’s bottom two rows, while the upper part of her pattern is certainly better than Donovan’s.

	► ►
Old Morrigan		New Morrigan

Ryu’s pattern was the hardest to decide on. It has a nice flavor in that the all-vertical pattern is simple and opposite of Ken’s all-horizontal pattern. Unfortunately, having an all-vertical pattern is extremely bad in Puzzle Fighter. One good quality Ryu had is that his vertical pattern would not create any Power Gems for you, but it’s very easy to build, say, red in column 2 and let Ryu fill in the red in column 1 for you. Far worse, his inability to drop anything other than yellow in column 4 means that you can place a single yellow crash gem in column 4 (or at the bottom of 3 or 5) and clear out your entire column 4 against Ryu. I stress again that in Puzzle Fighter, column 4 is the only one that matters: when you fill up column 4, you lose.

Although many patterns were tried for Ryu, I ultimately decided to keep his vertical theme, but replace the third row with a jumble of colors. You now at least need one or more green crash gems to clear out your column 4 against him, and the other junk in row 3 somewhat limits the size of the Power Gems his pattern helps you build. When the jumbled row was row 4, Ryu turned out too weak (and identical to his original bad form whenever you sent 18 or fewer gems). When his jumbled row was row 2, he was a bit too strong because his bottom 2 rows were more jumbled than even Akuma and Devilot’s bottom 2 rows. The jumbled 3rd row was about right.

	► ►
Old Ryu		New Ryu

This leaves only Hsien-Ko, whose drop pattern in the original game was even worse than it looked at first glance. Although her diagonal-themed pattern sounded good in theory, it was too easy to chain together very tall blue and green power gems against her in practice. Other than Dan, no other character in the rebalanced mode ended up with any columns that only sent a single color (they were eliminated from Ryu, Sakura, and Felicia). Hsien-Ko’s new pattern explores the trade-off of having solid colors in columns 1 and 6 (known to be very bad from the original Sakrua and Felicia), but with columns 2, 3, 4, and 5 that rival Akuma and Devilot’s powerful patterns. Furthermore, I kept the diagonal theme Hsien-Ko originally had, and I also kept the “build blue on left, green on right” counter-measure against her. Her new pattern at least makes it difficult to chain together your huge blue and green towers, and the middle portion is actually very powerful.

	► ►
Old Hsien-Ko		New Hsien-Ko

The Diamond

This van was hit by the original game's "diamond trick."

That covers all 11 drop patterns, but there is still one last detail: the diamond. In the original game, the diamond was intended to deal only 50% of the damage you’d do by breaking the same pieces without the diamond. However, there was a glitch that allowed you to bypass this and deal 100% of the damage. In order to perform this “diamond trick,” you must first find a place on your playfield where you can rotate your diamond-piece 180 degrees with just one button press (rather than the usual two presses). For example, if you have a lot of blocks in columns 2 and 4, you could put the diamond-piece in the well in column 3, so that it has no room to rotate horizontally. At this point, hold down on the d-pad, then when the piece touches down (with diamond on top), press rotate at the last moment. The diamond will rotate to the bottom position, and it will appear to break blocks as it always does, but it will do enormous damage because this technique avoids the 50% penalty.

The diamond trick is well-known by Puzzle Fighter tournament players and is considered by many to be a part of the game. Puzzle Fighter has a delicate balance where large attacks that almost kill are the most fun because they give the opponent enough ammunition to fire back a large attack of his own. Small attacks, such as the original diamond with its 50% penalty, aren’t as fun because they don’t nudge the game into that state where you are simultaneously almost winning and almost losing. This is why so many people consider the powerful diamond trick more fun than the original weaker diamond.

That said, it’s pretty convoluted to have to explain this technique to new players. I thought it should just become the default behavior of the diamond all the time, for simplicity’s sake. When this change was made, it ended up making the diamond even more powerful than the diamond trick in the original game. By removing the “trick” aspect, it was much easier to use it on exactly the color you wanted all the time, because no setup was needed (you didn’t have to create that narrow well between pieces to set it up).

Another somewhat related factor was the new “fast drop” feature bound to “up” on the d-pad. The arcade, Saturn, and PlayStation versions did not have this feature (it was new to the limited-release Dreamcast version). The ability to fast-drop pieces actually exacerbated the power of the diamond, because it further increased the reward for playing fast. I felt the game was moving too far towards “play fast at all costs, regardless of how many mistakes you make, just so you can get the diamond.” Playing fast is still necessary to win, but all things pointed to a slightly weaker diamond than the always-100%-diamond-trick version. We toned it down to always doing 80% damage (still quite a step up from the original 50%) damage, and I think it turned out well.

I hope that gives some insight into what I changed and why. The new game’s gameplay remains very similar to the original game’s, but hopefully there’s a wider range of reasonable characters to play, and a slightly improved mechanism behind the diamond.

--Sirlin

I'd like to take an in-depth look at an example of designing balance into a game. Although I'm choosing a fighting game (surprise!), the lessons should apply to many types of games. I'll go into some excruciating, genre-heavy details, but I think that's necessary to give the full force of what's really going on here.

Some games end up balanced through sheer coincidence, such as the fighting game Marvel vs. Capcom 2, which is "accidentally a very good game." Somewhere in Japan, there is a very lucky stable of monkeys who managed to type up Hamlet, or perhaps a screenplay to The Seven Samurai. The game featured a huge assortment of varied moves and 56(!) characters inside the most chaotic fighting game engine ever created, and it somehow happened to all work out. I will not be able to help you reproduce that. (Was there brilliant forethought involved, and I'm just unaware?)

Welcome to MvC2. Don't even ask.

The hero of our story, the oddly named Guilty Gear XX (ggxx), had quite a different genesis. It's obvious that a great deal of design went into the creation of the overall game system and into the design of each of the 20 characters. I suspected that a separate designer was assigned to each character, and from the look of the game's credits this is mostly true.

Each character in Guilty Gear XX plays very differently. I think it's easy to just read over that sentence, so I want to make it clear: compared to other fighting games, Guilty Gear XX has significantly more diversity in the gameplay each character allows than any other fighting game I have ever played, and that is saying a lot. So the game has more gameplay diversity than its peers, yet it is also one of the most balanced games in its genre. Diversity and Balance are natural enemies, so how is this possible? The solution is that Guilty Gear XX 1) has a game system that gives all characters equal access to an unusually large number of safeguards and abilities and 2) each character has some set of unique abilities that stray further from the standard template than they would in most other fighting games.

Faust and Dizzy of ggxx. If you'd like to play a Japanese doctor with a huge scalpel who wears a bag over his head with one eye-hole, this is the game you've been waiting for. If that's not for you, try the half-angel, half-devil, half-naked girl who throws bubbles, homing arrows, and laser-beam-shooting chomping mouths.

Basically, there is a "design skeleton" shared by all the characters, with each character having his own unique "meat on the bones." Since the designer can count on all characters having so many ways to get out of trouble, he can then give each character an unusual amount of "unique unfairness."

Let's look at the "skeleton" of features common to all characters. I'll start with boring stuff and work my way up. I will spare you the exact definitions of these first items, but with only a few exceptions, all characters have access to this suite of movement abilities and basic attacks:

• Double Jump
• Ground and Air Dash
• Sweep Attack
• Overhead Attack that Launches
• Ground Throw and Air Throw
• Air Recovery (aka "tech recovery")
• A silly Instant Kill mode

"f+p" invulnerability. Every character has a move performed by holding forward on the joystick and pressing the punch button. For every character except Testament, this move grants some invulnerability to the upper body. This means that if an enemy jumps in at you, a f+p move is very good "anti-air." That is, it works well against attacks coming from above (unless the enemy expected it, and double jumped…).

Super Meter. Lots of fighting games have a meter (other than your health meter) that gives you limited access to certain moves. In this game it's called the "tension meter" but we all know it's a "super meter." In ggxx, the meter charges up whenever you do anything offensive, such as attack, air dash forward, or even walk or run forward! All characters share the same mechanics for charging and expending super meter.

Green Blocking (aka "faultless defense"). While you are blocking, if you hold two buttons down, your character becomes surrounded by green rings. During this time, your super meter depletes, but you will take no "block damage" as you normally would from projectiles and other special attacks. Also, when you green block an attack, your defender will be pushed farther away than normal. This allows you to loosen up your opponent's traps pushing him too far away. You can also use green blocking while you are in the air to block attacks from an opponent on the ground (you can't block those types of attacks without green blocking).

Slayer (left), a vampire with many invulnerable moves(!), has his kick "green blocked" (aka "faultless defense") by Baiken (right), a cheap character who I hate playing against. ;)

Super Moves (aka "overdrive attacks"). Each character has a few big moves that require half of a full super meter to use. Very standard in fighting games.

Bridget (left), the transvestite nurse who fights with a yo-yo (yes, I'm serious), performs a super move on Ino (right), a guitar playing chick who fights with...music.

Alpha Counter/Guard Reversal (aka "dead angle" attacks). All those confusing terms mean the same thing: while you are blocking an attack, you can perform this maneuver to cancel your blockstun (the time you are stuck blocking an attack) with an attack of your own, for the cost of half of a full super meter. This means that if you are being overwhelmed by attacks, you can block and use this maneuver to get the opponent off of you.

Roman Cancel (rc). Now we're getting to the wacky stuff. Almost every attack in the game can be "roman cancelled," which means cancelled instantly in a flash of red. This is performed by pressing three buttons during the course of almost any move, and it costs one half of a full super meter. There are numerous uses for this. You can do an "unsafe move" with very bad recovery, then roman cancel the move when the opponent tries to hit you back. You can do a combo, then roman cancel what would normally be the last hit, which usually allows you to continue the combo.

False Roman Cancel (frc). We are deep in the dangerous territory of poorly translated Japanese terms. Perhaps this is intended to be a "force" rather than a "false" cancel. "Roman" seems to be short for "romantic," by the way, which makes no sense either. Anyway, an frc is very similar to an rc. Each character has a few moves that have a small window of frames where an frc is possible. If you attempt to do a normal rc (press 3 buttons) during one of these special windows, you get a blue flash rather than red, and you spend only half as much super meter as a normal roman cancel costs. What's the point? Most frc's are associated with moves that allow you to keep up pressure in your attacks. You might throw a projectile, frc (so you have no recovery at all), dash in and do a few normal attacks, then do the projectile again and frc it, repeat. Without frc's, you have to do the normal red roman cancel which takes half your super meter. That means you'd have only two repetitions of a trap at most, rather than four frc's before your suepr meter runs out.

Infinite Combo Safeguards

These next several features all contribute to preventing "infinite combos." An infinite combo is a situation where once the opponent lands the first hit of the combo (in the right situation), then he can continue the combo forever until he wins. The more complicated a fighting game is, the more likely it is to have unwanted infinite combos. The designers of ggxx put many systems in place to reduce the likelihood that such combos could exist.

Burst. This is the most blatant solution: a move that lets you break out of a combo. This taboo concept is very rare in fighting games, after the semi-fiasco of the "c-c-c-combo breaker!" in the game Killer Instinct. That game allowed the victim of a combo (who usually cannot do anything at all until the combo ends) to input a paper/rock/scissors guess that would allow him to escape the combo. This guess had no cost, so it was extremely common (too common) to break out of combos in that game.

In ggxx, there is an entirely separate meter (the burst meter) that keeps track of how often you are allowed to burst (about once per round). You always start the first round of a game with a full burst meter. Once you use the burst (it's the only move you are allowed to do while you are in hitstun, by the way), you can't use it again until your burst meter refills. The burst meter refills slowly over time, and it also increases the more you get hit. Getting hit by some moves (moves more likely to create infinite combos) fills your burst meter faster than others.

The burst is like a "get out of jail free card." It lets every character get out of trouble once, "for free." Of course, a clever opponent can expect the moment you will burst and simply voluntarily stop attacking right before your burst, then punish the recovery of your burst with an even bigger combo, so you have to be careful.

I should also note that there is an alternate use of the burst, a use often called a "gold burst" or "offensive burst." So far, I have only talked about activating the burst while you are in hitstun (the brief period of time where you are reeling back after getting hit by an attack during which you can perform no moves except a burst). But you are allowed to perform a burst almost any other time as well. You can burst at the first moment of the round if you want to. If you perform a burst when you are NOT in hitstun and the very beginning of the burst actually hits the opponent, then you instantly get a full super meter (aka "tension meter".) So what's the significance of that? If you are getting rushed down and overwhelmed by attacks, you should use the first type of burst to get out of trouble. But if you are the one doing the attacking, you don't need to get out of trouble. You want to cause even more trouble. So you can attack the enemy and use lots of rc's and frc's (those use up your super meter), then you can "gold burst" suddenly as they try to get you off them, and if that burst connects, you will have a newly full super meter to perform even more rc's, frc's and supers in your attack pattern.

Guard Meter. Yes, there are a lot of meters in this game. Right under your health meter is a little red meter called the guard meter. It starts at 50% full, and naturally tends to wander back to 50% over time. The more attacks you block, the higher that meter gets. The more attacks you get hit by, the lower that meter gets. The higher the meter is (the more attacks you recently blocked) the less you benefit from the game's normal system of damage scaling. Usually, when you get hit by a combo, each successive hit is "scaled" down in damage more and more. But when your guard meter is high, even an ordinary combo can do massive damage to you because you are not being protected by the usual damage scaling. This is meant to punish overly defensive players.

On the flipside, the lower your guard meter is (meaning you got hit by a lot of attacks in a short period of time), the more damage scaling you benefit from. A very, very long combo will eventually do only one pixel of damage per hit because of this feature. So even if an infinite combo did exist, it would take an incredibly large number of hits to actually kill you. Furthermore, you receive another even more important protection when your guard meter is low: reduced hitstun. Every time you get hit by a move, you are briefly stuck in a reeling animation where you can't do anything (except burst). This is the basic concept that allows combos to exist at all, since the opponent can often hit you again before your hitstun ends. But in ggxx, the more you get hit, the shorter your guard meter becomes, and the shorter your hitstun becomes. So if there exists a combo that is a "loop" of repeated moves, it may be possible to do 3 or 4 repetitions of the loop, but eventually the opponent's hitstun becomes so short that the combo simply stops working.

Increased Gravity. Another safeguard against infinite combos. The longer your character is being juggled by a combo in the air, the greater the force of gravity on your character becomes. Many infinite combos in fighting games involve "juggling" a character in the air with attacks. Much like with actual juggling, it is a fight against gravity to keep it all going. While it might be possible to do 3 or 4 repetitions of a juggle loop, eventually the victim's body falls so fast to the ground that the juggle is no longer possible.

Potemkin no fast. Potemkin smash.

Phew, some of you might have even made it through reading all that! Let's do a short recap, just the broad strokes of it all. If you want to attack, ggxx gives you super meter, and gives you several options to spend that meter to enhance your attack (rc's, frc's, and super moves). The guard meter system allows you to punish opponents who block too much. Those opponents are being somewhat punished by the game system in general anyway, since blocking doesn't build super meter but attacking does.

More important, I think, are the game systems that help you when you are BEING attacked. Every character has upper body invulnerability with their f+p move (except Testament). That alone is a great help in stopping attackers. If you want to spend super meter as a defender, you can green block or alpha counter. Green blocking will protect you from all damage while you block (except throws) and it will push that nasty attacker away from you. Alpha countering is another method to get the attacker off of you, and it can be performed while you are in blockstun.

That's a lot right there, but there's plenty more. If you actually get hit by an attacker, you have all sorts of things going for you. Your guard meter will eventually reduce the damage you take by the combo and shorten your hitstun allowing you to escape. Increased gravity will also eventually thwart their combo. You can nip it all in the bud, though, by simply bursting right at the very start of their combo, avoiding almost all damage. And don't forget that a great way to nullify attacks is simply not to be in the way of them. Every character can double jump, and every character but one can air dash.

Now that see how much EVERY character has to work with when it comes to getting out of trouble, it becomes more clear how it's possible that ggxx has more diversity in gameplay amongst its characters than most (if not all) other fighting games. Character designers know that they can go in some really extreme directions with each character design because they know that the game system shared by all characters probably lets just about any character eventually get out of just about any trouble thrown their way.

So we have one character who can alpha counter as much as she wants (not limited by super meter). Another that can create pool balls on the screen, control their formations, and bounce them off each other, allowing seemingly infinite variations of attack patterns and setups. Another character can summon a "shadow" who acts as a completely separate character. The player's joystick inputs and button PRESSES apply to both characters at once, but his button RELEASES apply to the shadow character only. The point is, it's an incredibly different endeavor to play any given character in ggxx, yet the gameplay skeleton (mostly of defensive features) ensures that all this craziness will at least mostly work out in the end.

I think that's the secret of capturing both diversity and balance, or at least one effective method of reaching the holy grail: a robust, shared system of defense with diverse and unique attacks for each character/race/side.

Part 1 | Part 2

Balancing a game is tricky business. Balancing a multiplayer game is nearly impossible. Balancing a competitive multiplayer game that goes on to be very popular--well that actually is impossible...unless the game has no variety.

The first lesson is that "variety" and "balance" are inversely proportional; the more you have of one, the less you're likely to have of the other. Both sides in chess have identical pieces, and the only difference between the black side and white side is that white goes first. Street Fighter would be much more balanced if we just removed all the characters except Ryu. Quake is fairly "balanced" since each player has the same "moves" (or ability to acquire weapons and armor) as any other player. The different spawn points do introduce some "imbalance" though. The trick is to introduce variety, yet keep a reasonable level of balance. And it's quite a trick, indeed.

What I've just described is a very narrow way of thinking about this mysterious concept of "balance." After all, if the only way to dominate a Quake match were to use the rocket launcher, then how balanced a game would that be? Even if all players had equal access to the rocket launcher, it would be hard to call a game so skewed towards one thing "balanced." We need a definition of balance if we are to going to talk about it. Let's give that a shot.

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.

That was quite a mouthful. It basically means that game doesn't degenerate down to a very small number of real options. This is a pretty broad definition since it encompasses the concept of "brokenness" as well. If, in chess, only pawns were used in tournament play (if pawns were so good as to be "broken"), I would say the game lacks balance, even though both players start with the same pieces. Another point of contention might be my use of the phrase "large number" rather than "large percentage" of viable options. Let's use my favorite example of Marvel vs. Capcom 2 to explain that.

Whether MvC2 is anything close to balanced is an incredibly complicated question that we'll have to come back to. For now, let's say that most players agree that there are about 10 "top tier" characters in the game. Certainly no more than 15. Yet the game offers a whopping 54 characters! Although percentage-wise, that's pretty poor, I'd have to say that 10-15 characters that are all extremely viable for tournament play is pretty darned good for a fighting game, even if the batting average of playable characters if low. Others might care more about the percentage, but it's a minor point. Let's move on to a major point.

Multiplayer vs. Single Player

Balancing a competitive multiplayer game is orders of magnitude harder than balancing a single player game. When we try to balance a single player game, we are basically striving to match the "skill" of the computer to the skill of the player. There are many techniques for doing this, and there is a large margin of acceptable error. For example, consider what happens if Joe Hardcore figures out a super sneaky way of beating almost every enemy in a single player game for free. Further suppose that this method is very obscure and discovered by less than 1% of all players. Factoring in strategy guides and the internet, sure, others will figure out this method, but the overall impact will be small. Joe Hardcore feels full of himself, the computer doesn't mind being beaten, and most players will never know about this method at all. It's bad, but it's not that bad.

If the same trick/bug existed in a competitive multiplayer game, the game would be totally ruined. The knowledge would spread like wildfire, as the 1% proceeded to crush the masses. The masses would learn the trick, and soon, there would be no gameplay. There would only be the trick. The game would degenerate to the exploitation of a single bug, and possibly the exploitation of another even more obscure bug as a countermeasure to the first. In short, the multiplayer world doesn't forgive imbalance: it abuses it 100 times more than even many game developers realize.

So how can we make sure that a competitive multiplayer game will be balanced when we release it? Let me tell right now that you can't. You cannot. There is no way. If your game has the complexity needed to be interesting and successful, then it's pretty much guaranteed to be beyond the level of complexity that can be fairly balanced right out of the gate. Anyone who thinks otherwise just hasn't been in the trenches of real competitive gameplay. (Before you send me e-mail, at least read the rest.)

Now that does NOT mean we should give up. And it certainly doesn't mean the problem isn't worthy of analysis. There are things we can do to minimize the chances of insane imbalance, and there are methods to correct balance problems when they are discovered. Before we get to those, let's think about why the problem is so difficult in the first place.

Finding Bugs in Code and Design

Developers often think they are the best players of their own games. They designed and programmed their games, after all, so they have many advantages. They know the actual equations and formulas the game uses. They know the AI routines. They know the nuances, and the little tricks they put in there. But in reality, the gamers outside the company have many, many more advantages than that. Shall we count the ways?

1) Gamers will uncover any and all relevant equations and formulas about a game through trial and error, testing, and hacking.

2) Gamers will find bugs and "features" that the developers never knew about.

3) Gamers will use features in bizarre ways never envisioned by the developers or even the testers.

4) Gamers have far, far more time to play the game than the developers do. The developers are busy making games, but hardcore gamers have seemingly infinite time. Just look at the EverQuest stats. 85 hours a month was the AVERAGE playing time last year, including people who pay but never log in. Just imagine how much the people at the other end of the curve are playing!

5) The gaming public just has far more pairs of eyeballs than the developers and testers. 1,000,000 people will find something that a team of 30 missed.

6) The developers and testers have skewed perceptions on balance since features change often throughout development. A particular move or strategy might be considered weak, but end up strong. The development team might have a bias against this move or strategy, since they remember it being too weak to bother trying.

7) The developers and testers are often playing without all the options available in the final game, or at least without the final tweaking in place. It's very possible for entirely new strategies to become viable after new features are added at the last minute or old values are tweaked. Yet the developers often have little or no chance to test these newly-emerged strategies against other strategies they've deemed to be good.

8) Finally, the gamers have the motivation. If you discover a strategy in Starcraft or a deck in Magic the Gathering that truly breaks the game wide open, you have lots of rewards waiting for you. Fame, glory, tournament victories, even fabulous cash and prizes. You'll have a reputation and in some circles, you'll finally "be someone." That's a lot of incentive for some people. If you're a tester at a game company and you discover the exact same thing before the game is released, then you get to type up another bug report and get you toys taken away from you. Not quite the same level of incentive.

Female fans of top Magic the Gathering player Ben Rubin await his arrival at a Pro Tour tournament in Sydney, Australia.

Design Balance In, Then Get Good Testers

Ok, so balancing a game is damned hard. But how to we attempt it? The first step is designing checks and balances into a game from the start. Every unit in Starcraft has a strength and a weakness. Even basic strategies have strengths and weaknesses. Attacking early (at the expense of a strong economy) beats expanding early, but loses to defending early (if the defender can maintain a good economy and survive the attack). Defending early loses to expanding early. Checks and balances. You could even say, paper, rock, and scissors.

Another great example of designing (at least some) balance into a game is Magic the Gathering. There are five colors in Magic, and each one has unique strengths. Blue can counter spells, draw cards, and fly, for example. The more colors you want in a deck, the less consistent the deck will be. If your whole deck is blue, you'll pretty much always have blue mana whenever you need it (very consistent), but you will also have the built-in weaknesses of blue: you can't deal with threats once they are in play. You might play blue and white or blue and black to better remove threats (more versatile deck), but now, because of the way mana works in Magic, you are less likely to have the correct colored mana at the right time (less consistent).

Another principle to live by is to try to prevent the situation of a player saying "I knew exactly what the opponent was going to do, but there was absolutely no way to stop it or punish it." That's just another way of saying "checks and balances," really. There are many levels to this concept, though. In a fighting game, you could make sure that whenever someone sits there you can throw them, whenever they attack you can block or parry or do a reversal, and so on. It will feel really scientific as you go through all the different conceptual things a player can do (attack, block, throw, parry, etc.) and make sure each has a counter. But it's a whole new ballgame when you throw all those options together, in combination. Perhaps a pattern of attack (rather than a single move) is too powerful. That can be difficult to uncover during testing.

That just goes to the next point: get good testers. Game testers are usually paid and treated poorly. That's really unfortunate. Testers are usually employed to find software bugs, but in competitive games, they really need to be playing the game constantly in order to break it design-wise as well. That's the only hope of getting anywhere near balanced for launch. I'm a pretty academic guy, but even I would never pretend that someone could take a design document I wrote and turn that into a perfectly balanced game without testing. It's ALL about testing.

One company acutely aware of this is Wizards of the Coast, makers of the collectible card game Magic the Gathering. This game is played by millions around the world, boasts 80,000 sanctioned tournaments a year(!!!), and routinely gives prizes as high as $50,000. They are serious about making a balanced game. What the game industry would call "testers" they call "developers." Their developers are pulled from the top ranks of the "Pro Tour" of Magic. Many of them even made a living as Magic players before joining the dev team. They do extensive play testing of every set in a curiously named internal league called the "Future Future League." Even they aren't perfect, but they go a long way towards balancing a mind-bogglingly complicated game.

Zvi Mowshowitz wins the Pro Tour Tokyo Magic the Gathering tournament, taking home $30,000. He has lots of incentive to discover and exploit balance problems.

Career earnings of Magic the Gathering Players

as of October 24th, 2001

	Jon Finkel	$249,227
	Kai Budde	$144,425
	Ben Rubin	$119,470
	Zvi Mowshowitz	$108,320
	Kyle Rose	$104,225
	Mike Long	$96,202
	Chris Benafel	$93,850
	Tommi Hovi	$93,780
	Darwin Kastle	$90,627
	Bob Maher, Jr.	$82,377

Another company worthy of note is Blizzard. Starcraft is one of the best balanced games ever to be played on the internet, and that's no accident. Blizzard knows that the release date of a game is not the end; online games require support. Thanks to Blizzard's expert team of tester and balancers, including lead balancer Rob Pardo, Starcraft was patched several times for bugs as well as balance issues. Blizzard has a wonderful resource in battle.net data to help in balancing. Battle.net is the only legal matchmaking service for the game, so tons of data can be collected about play patterns. They know the percentage of people who play Zerg on the Hunters map, the average win ratio for those players, the average length of a game, and zillion other stats. It's sure nice to back up anecdotal statements like "Zerg are too good. This one guy can beat me all the time with them" with statements like "Zerg have a 76% win rate in tournament play on Lost Temple."

Starcraft teaches us another useful balance principle as well: purity of purpose. For the most part, units in Starcraft are designed to each have a specific purpose that doesn't obsolete any other unit. The more functionally independent each unit is, the better. The reason is that balancing is made that much easier. Consider this example of the lack of purity of purpose. Mutalisks are the fast flying unit of the Zerg. They have a designated size of "small" meaning they take less damage from the "explosive type" damage dealt by most anti-air units. It also means they take less damage from the explosive type damage used by most other air units. Furthermore, mutalisks can attack air-to-air and air-to-ground and they use the same projectile in both situations. This is NOT a good example of purity of purpose. The mutalisk is suddenly a good air-to-air and air-to-ground unit, and what can you do to tone down one but not the other? Chaning the body size to medium would make the unit worse in both situations, as would worsening the shot. If mutalisks are too good (or not good enough) at air-to-ground, you could change the damage dealt by most ground-to-air units. But what about the Protoss photon cannon that uses the same shot for ground and air? If you make that worse to give mutalisks a chance, you just make the Protoss ground defense worse, too, as a side effect. It's one big mess.

Here's a pack of flying Mutalisks in Starcraft. These suckers were hard to balance, and completely ruled the air in older versions of the game.

In fact, creating functionally independent moves/units/whatever in design is so important, that I'm going to write an article on that, too. There'll be a link right here when it's done.

The Capcom Principle

Capcom has a really interesting take on game balance. "Back in the day" they might have tried a more...traditional method of balancing, but these days I'm convinced they have a rather unique way of doing things. Here's the Capcom Principle of Balance:

Give every character something "so good that it's broken." Include so much variety that by the time anyone ever figures out which broken thing actually does ruin the game...the game will be dead by then anyway.

Clever, really. They understand balance vs. variety well. They create as much variety as possible, making balancing so impossible that even they can't really be sure what's balanced. The saving grace is that the huge gaming audience is faced with a task that takes at least a year to sort out (or maybe many years!).

Capcom will give only this character an instant speed, long-range throw, only this character a ridiculously fast walking speed, only these characters an air-throw. Now hand out the unfair stuff, one or two to each character. Mix it up real good, and bam, now it has so much variety and "imbalance" that in a weird way everything becomes "fair" again. Every character can emphasize their own little unfair thing and considering all the variety thrown in for kicks, there's bound to be some way to stop said unfair thing somehow. This is the essence of Marvel vs. Capcom 2.

Balance this! So broken it's almost fair?

Accept that your game will not be balanced without considerable work. Keep your design functionally independent so you can adjust things without stepping on the toes of other things. Hire excellent testers, treat them well, and motivate them to break the game, design-wise as well as bug-wise. Realize that the more variety you include, the more interesting the game will be, but the more difficult balancing it will be. The other route is to go so overboard on features and variety that it will takes years for players to sort out what's actually broken. If you go that way, be sure to include Capcom in the special thanks section of your end credits.

Part 1 | Part 2

A simple rock, paper, scissors (RPS) system of direct counters is a perfectly solid and legitimate basis for a strategy game provided that the rock, paper, and scissors offer unequal risk/rewards.

Consider a strictly equal game of RPS. We'll play 10 rounds of the game, with a $1 bet on each round. Which move should you choose? It makes absolutely no difference whether you choose rock, paper, or scissors. You'll be playing a pure guess. Since your move will be a pure guess, I can't incorporate your expected move into my strategy, partly because I have no basis to expect you to play one move or another, and partly because I really can't have any strategy to begin with.

Now consider the same game of RPS with unequal payoffs. If you win with rock, you win $10. If you win with scissors, you win $3. If you win with paper, you win $1. Which move do you play? You clearly want to play rock, since it has the highest payoff. I know you want to play rock. You know I know you know, and so on. Playing rock is such an obvious thing to do, you must realize I'll counter it ever time. But I can't counter it (with paper) EVERY time, since then you could play scissors at will for a free $3. In fact, playing scissors is pretty darn sneaky. It counters paper---the weakest move. Why would you expect me to do the weakest move? Are you expecting me to play paper just to counter your powerful rock? Why wouldn't I just play rock myself and risk the tie? You're expecting me to be sneaky by playing paper, and you're being doubly sneaky by countering with scissors. What you don't realize is that I was triply sneaky and I played the original obvious move of rock to beat you.

That may have all sounded like double-talk, but it's game theory (in the mathematical sense) in action. And it had quite a curious property: playing rock was both the naive, obvious choice AND the triply sneaky choice. For much more on that concept, read my article on Yomi Layer 3.

Fighting games rely heavily on RPS. They have both overall games of RPS going on as well as many rapid fire situations of RPS. Virtua Fighter 3 games can even have 5 sets of RPS take place in a period of 2 seconds! No joke!

Virtua Fighter's overall system of RPS is as follows: attacking beats throwing, throwing beats blocking or reversing, and blocking and reversing beats attacking.

To be clear, let's define terms.

An attack is a move that deals damage. An attack has an initial startup phase where it can't yet do damage (a punch extending), a short phase where it actually can do damage (the sweet spot of the punch), and a recovery phase (the arm retracts). If the defender is blocking correctly, an attack will not damage him, but he can be thrown.

A throw is a special type of move that instantly grabs an opponent whether he's blocking or not and does damage. The catch is, a throw will not grab an opponent who attacking (specifically, a throw will fail if the opponent's move is in startup or hitting phase).

Wolf performs his Twirl and Hurl (one of the biggest throws in Virtua Fighter 3) on Shun.

A reversal is a special type of move that grabs an incoming attack. Reversals usually look like throws, but they work at the exact opposite times. A reversal only works when the opponent's move is in startup or hitting phase, which are, incidentally, the only times a throw would fail.

Even these explanations are simplified, but the RPS system is basically there. Attack the opponent. If they tried to throw you, you'll hit them. If they block or reverse your attack, they nullified your attack. If you expect them to block, you can throw. If they expect you to throw, they can attack.

Dead or Alive 2 basically uses this same system, except that the risk/reward for doing a reversal is much different. Reversals are difficult and relatively rare in Virtua Fighter, but they're incredibly easy and do a ridiculous amount of damage in DOA2. Reversals are so effective, in fact, that they can paralyze the enemy into not attacking for fear of being reversed. Of course, that's when you throw them....

Dead Or Alive 2's system of attacks, throws, and reversals is not pictured here.

Real-time Strategy games are the other kings of the RPS system. Like fighting games there's the concept of RPS on large scale and a small scale. On the small scale, particular units are designed to counter each other in a RPS way. A marine dies to a guardian. A guardian dies to a corsair. A corsair dies to a marine. Abstractly, there are 6 categories of unit. Ground units can either attack 1) other ground units, 2) air units, or 3) both. Air units can attack 4) other air units, 5) ground units, or 6) both. Pure ground-to-ground units usually beat both other types of ground units, yet lose to both types of air units that can attack ground. Similarly, pure air-to-air units usually beat both other types of air units, but loose to both types of ground units than can attack air.

RPS is not limited purely to units countering each other though. Real-time strategy games also have the concept of trading off powerful units now for a strong economy now, which leads to even more powerful units later. So on one extreme, a Zerg player in Starcraft might sacrifice his entire economy to get a quick attack force ("6 pool" is the term). This will likely beat a player who chose the other extreme of playing for pure economy and no immediate attack force (by building double oven triple hatcheries). A moderate build (pool on 9^th peon, one sunken colony) will likely defend against the early attacker's rush, though. Surviving the rush, the moderate build will have a much superior economy and win in the end. However, this moderate build will produce an inferior economy to the player who built 2 or 3 hatcheries and went for pure economy.

This is all very textbook and a number of other factors come into play in practice, but the underlying RPS is there, and it most certainly has unequal payoffs. In Starcraft, the early rush is a very, very risky strategy. It's all or nothing. You'll either win right away off it, or your rush will fail and you'll almost surely lose. Because of this, the early rush isn't all that common (depending on the map), but the very threat that the opponent might play the early rush is enough to stop you from playing for pure economy every time.

RPS Gone Horribly Wrong: Killer Instinct 2

Killer Instinct 2 boasted a rock, paper, and scissors system of moves. Every character had 3 moves assigned RPS designations. It was all rather arbitrary and artificial. Jago's "rock" move would beat any other character's "scissors" move. Jago's "scissors" move beat any other character's "paper" move. The entire system was so homogeneous, that there was little basis from which to choose rock over scissors. The gameplay was based on blind guessing, and felt hollow and devoid of strategy. RPS needs to be a natural part of the game, and it absolutely has to have unequal risk/rewards for each move.

For those interested in reading way too much about RPS strategy, I'll leave you with this link.

"Yomi" is a Japanese word meaning "knowing the mind of the opponent." It comes from the lingo surrounding Virtua Fighter, perhaps the most complex video game ever made. If you can condition your enemy to act in a certain way, you can then use his own instincts against him (like in Judo). Paramount in the design of competitive games is the guarantee to the player that if he knows what his enemy will do, there is some way to counter it.

What happens, though, when your enemy knows that you know what he will do? He needs a way to counter you. He's said to be on another level than you, or another "yomi layer." You knew what he would do (yomi), but he knew that you knew (yomi layer 2). What happens when you know that he knows that you know what he will do (yomi layer 3)? You'll need a way to counter his counter. And what happens when he knows that you know....

Sound like a joke that could never happen in real gameplay of an actual game? Surprise: it's quite common in strategy games. The reason has to do with conditioning the opponent and the inequality of risk/reward in these guessing games (see my article on Rock, Paper, and Scissors in Strategy Games).

Before we get into how ordinary human minds can become entangled in complicated guessing games, let's look at what needs to be there to create these guessing games at all. The designer's tendency might be to create moves and counters. Then create counters to counters, then counters to counters to counters, then counters to those, and so on. Actually, a game need only support counters up to Yomi Layer 3, since Yomi Layer 4 can loop around back to Yomi Layer 0.

Let's say I have a move (we'll call it "m") that's really, really good. I want to do it all the time. (Here's where the inequality of risk/reward comes in. If all my moves are equally good, this whole thing falls apart.) The "level 0" case here is discovering how good that move is and doing it all the time. Then, you will catch on and know that I'm likely to do that move a lot (yomi layer 1), so you'll need a counter move (we'll call it "c1"). You've stopped me from doing m. You've shut me down. I need a way to stop you from doing c1. I need a counter to your counter, or "c2."

Now you don't know what to expect from me anymore. I might do m, or I might do c2. Interestingly, I probably want to do m, but I just do c2 to scare you into not doing c1 anymore. Then I can sneak in more m.

You don't have adequate choices yet. I can alternate between m and c2, but all you have is c1. You need a counter to c2, which we'll call c3. Now we each have two moves.

Me: m, c2 You: c1, c3.

Now I need a counter to c3. The tendency might be to create a c4 move, but it's not necessary. The move m can serve as my c4. Basically, if you expect me to do my counter to your counter (rather than my original good move m), then I don't need a counter that; I can just do go ahead and do the original move...if the game is designed that way. Basically, supporting moves up Yomi Layer 3 is the minimum set of counters needed have a complete set of options, assuming Yomi Layer 4 wraps around back to Layer 0.

This is surely sounding much more confusing than it is, so let's look at an actual example from Virtua Fighter 3 (which will almost certainly confuse you even more).

Example of Yomi Layer 3 from Virtua Fighter 3 Let's say Akira knocks down Pai. As Pai gets up, she can either do a rising attack (these attacks have the absolute highest priority in the game) or she can do nothing. A high rising attack will stop any attack that Akira does as she gets up, but if Akira expects this, he can block and retaliate with a guaranteed throw. Pai does the rising kick and Akira predicts this and blocks. Now the guessing game begins.

Akira would like to do his most damaging throw (that's his m), and be done with it. Even though the throw is guaranteed here, all throws can be escaped for zero damage if the defender expects the throw and enters the throw reverse command. The throw is guaranteed to "start" but Pai might reverse it. In fact, Pai is well aware that a throw is guaranteed here (it's common knowledge), and it's only obvious that Akria will do his most damaging throw. After all, this situation has happened a hundred times before against a hundred Akiras and they all do the same thing. It's really conditioning, not strategy, that tells Pai she needs to do a throw escape here (that's her c1). In fact, it will become her natural, unthinking reaction after a while.

Akira is tired of having his throw escaped again and again. He decides to be tricky by doing one of his very slow, powerful moves such as a double palm, a reverse body check, a two fisted strike, or a shoulder ram (we'll just lump all those into c2). Why does a big, slow move work in this situation? First of all, if Pai does her throw escape and there is no throw to escape, the escape becomes a throw attempt. If her opponent is out of range or otherwise unthrowable for some reason, her throw attempt becomes a throw whiff. She grabs the air and is vulnerable for a moment. One important rule in VF is that you cannot throw an opponent during the startup phase or the hitting phase of a move. So if Akira does a big, powerful move, he is totally unthrowable until after the hitting phase of the move is over and he enters recovery (retracting his arm or leg).

Back to our story. Akira is tired of getting his throw escaped all day, so he does standard counter to any throw: a big, powerful move. This c2 move does a decent amount of damage, by the way. The next time this whole situation arises, Pai doesn't know what to do. Her instincts tell her to reverse the throw, but if she does, she is vulverable to Akira's slow, powerful move. Rather than go for the standard reverse, Pai does her c3 move: she simply blocks. By blocking, she'll take no damage from the Akira's powerful move, and depending on exactly which move it was, she'll probably be able to retaliate.

So what does Akira do if he expects this? In fact, he needs no c4 move since his original throw (m) is the natural counter to a blocking opponent. A throw is a special kind of move that grabs an enemy and does damage regardless of whether they are blocking. It's specifically designed to be used against an opponent in block who is afraid of an attack.

In summary,

Akira has: throw; powerful, slow move Pai has: throw escape; block.

As I tried to show, it's actually pretty reasonable to expect players to be thinking on Yomi Layer 3, 4 or even higher. It's because conditioning makes doing the throw escape (c2) an unthinking, natural reaction. But against a clever opponent, you'll have to think twice about doing a standard throw escape, or blocking. The Akira player will do the occasional powerful, slow move just to put his enemy off balance and abandon his instinct to escape the throw. Then Akira can go back to his original goal: land the throw.

Another very interesting property is "beginner's luck." Notice that a beginner Akira in this situation will go for the throw, since that works on other beginners who haven't learned to throw escape. The beginner Akira will never land the throw on an intermediate player, though, since the intermediate player knows to always throw escape. But strangely, the beginner will sometimes land the throw on the expert, since the expert is aware of the whole guessing game and might block rather than throw escape. Of course, the expert will soon learn that beginner is, in fact, a beginner and then he'll be able to yomi almost every move.

Just as a final note on Virtua Fighter to further demonstrate the complexity of its guessing games, I actually greatly simplified the example above. I left out, for example, that Akira has another c2 move besides a slow, powerful move. He can also do what's called a "kick-guard cancel" or "kg." This means he can press kick, which will make him unthrowable until his kick reaches recovery phase. If Pai tries to throw, she'll whiff. But then Akira can cancel the kick before it even gets to the hitting phase. Now he's free to act and take advantage of Pai's whiffed throw vulnerability. Now, Akira has a guaranteed throw, putting him back in the exact same situation he began in. The catch is that if Akira does kg-cancel and then goes for the throw he originally wanted to do, Pai will probably not have time to react with a throw escape. It's just too fast. She'd have to be on the next yomi layer. She'd have to expect Akira to throw, enter a throw escape, see the kg-cancel, then immediately enter her next guess (probably an attack or throw escape). Any hesitation and she'd be thrown.

Crazy huh?

The point I'm making here is that despite Virtua Fighter's absurd complexity, players really are able to think on the levels I'm hinting at. Playing such a game and successfully landing a move because you knew he knew you knew he would do a particular move is the greatest feeling in the world. So design counters and counters-to-counters, and so on, but know that making Yomi Layer 4 the same as Layer 0 allows you to only design counters up to Yomi Layer 3. It's nerdy, but true.

Even some of the very best strategy games (chess and Starcraft, for example) suffer from slippery slope. That means that once one player begins to lose by a little bit, he's at a disadvantage and likely to fall further and further behind. In this type of game, one player usually loses long before the game is technically over, which isn't exactly fun.

When a player loses a piece in chess, his ability to attack and to defend has been slightly reduced. Sure, there are many other factors in chess---positioning, momentum, pawn structure---that determine if a player is actually "losing," but losing a piece does have an affect. Clearly, losing a lot of pieces, say 8, puts a player at a significant disadvantage. It's pretty hard to make a comeback in a chess game, and a chess game is usually "won" for all intents and purposes many, many moves before the actual checkmate move.

Starcraft is the same way. Consider two Starcraft players of nearly equal skill. One player rushes the other (sends a small attack force very early in the game). The rush forces the defender to spend some time defending himself, and his worker units are disrupted for just a few seconds. One worker unit is killed, then the defender is barely able to defeat the attacker's units. This conflict was actually very close. It probably didn't feel devastating to the defender, but he's just taken the first step down a very slippery slope. The defender spent time managing that battle that the attacker used to build a stronger economy. The defender lost a few precious seconds of collecting resources, but the attacker did not. The defender must spend both the time and money to replace that lost worker, but the attacker does not. Resource collecting is nearly exponential in Starcraft. A small disadvantage early on becomes more and more magnified as the game goes on.

The outcome of this game of StarCraft was basically decided during this early rush: Zerg will lose the game.

Apart from the exponential nature of resources, there's the same concept of losing pieces as in chess. If an attacker kills some of the defender's units in Starcraft, the defender is that much less able to defend or attack in the future. If a defender barely holds of an attack, but comes out slightly on the losing end, he'll be even less able to defend against the next wave of attacks that are sure to come moments later. He'll then probably fall further behind, and be even less able to stop yet another attack wave. There basically aren't comebacks in Starcraft. And just as in chess, the moment of loss comes long before the actual conditions of the game ending are fulfilled. As fun as Starcraft is, this slippery slope aspect definitely detracts from the experience.

Fighting games typically don't suffer from slippery slope. In Street Fighter, for example, your character still has all his moves even when he's about to lose. While it might be "realistic" for a nearly dead character to limp, move slowly, and have generally less effective moves, it sure wouldn't be fun. Comebacks are frequent in Street Fighter, and games often are "anybody's game" until the last moment. Street Fighter does have some very minimal slippery slope aspects (if you're very near death you have to worry about taking damage from blocked moves which aren't a threat if you have full life), but overall it's pretty "slippery slope neutral."

There is one version of Street Fighter that stands out as an exception: Marvel vs. Capcom 2. In this game, each player chooses 3 characters. At any given time, one character is active and on-screen, and the other two are off-screen, healing back some lost energy. The off-screen characters can be called in to do an assist move, then the jump off screen again. The main character can attack in parallel with the assist character, allowing for a wide variety of tricks and traps. The player can switch the active character at any time, and he loses the game when he loses all three characters. But here, slippery slope rears its bitter head. When one player is down to his last character and the other player has two or even all three of his characters, the first player is at a huge disadvantage. The first player has can no longer attack in parallel with his assists, which often means he has no hope of winning. Comebacks in MvC2 are quite rare and games often "end" before they are technically over.

Fighting games with "ring out" such as Virtua Fighter and Soul Calibur as especially devoid of slippery slope properties. In these games, a player instantly loses if his character is ever pushed out of the ring, no matter how much energy he has. Basically, no matter how far behind you are, no matter how close you are to losing, you always have a 100% damage move: ring out. Long ago, I thought this concept was "cheap" and served only to shorten games while adding little benefit, but actually the threat of ring out adds quite a bit to both these games. Since the threat of ring out is so great, another whole element of positioning is added to the game. A player must fight both to do damage to his opponent, and fight for position to avoid ring out. But back to our story....

Lau (right) has Sarah (left) at the edge of the ring. Another hit could result in him ringing her out.

Perpetual Comeback: Puzzle Fighter's Claim to Fame

Is there an opposite to slippery slope? A game in which losing actually increases your chance to win, rather than decreases it? It's a strange concept, and I've dubbed it "perpetual comeback." In all the world, I'm aware of only one game that truly uses it: Capcom's Super Puzzle Fighter II Turbo, or Puzzle Fighter, as it's commonly referred to.

Puzzle Fighter is, in my opinion, far and away the best puzzle game ever made, and even one of the best games ever made. It looks standard enough; it's one of those games where each player has a basin that pieces fall into. There are four different colors of pieces, and you try to build big, single colored rectangles (power gems). You can then shatter those rectangles with special pieces called crash gems. The more you break, the more junk you drop on the opponent's side. When your side fills to the top, you lose. Sounds pretty standard, right?

Several factors come together to create perpetual comeback in Puzzle Fighter. Firstly, each "character" (there 10 to choose from, including secret characters) has a different "drop pattern." A drop pattern is the pattern of colored blocks that a character will send to his enemy when that character shatters blocks on his own side. For example, Ken's drop pattern is horizontal row of red, followed by a horizontal row of green, then yellow, then blue. Every time Ken sends 6 or fewer blocks to his opponent, he'll send a horizontal row of red. Every time Ken sends 12 blocks, he'll send a row of red, then a row of yellow. Since the enemy knows this, he can plan for it. He can build his blocks such that Ken's attack will actually help rather than hurt. There's one catch: when you send blocks to the opponent, they appear in the form of "counter gems," which can't be broken immediately by normal means, and can't be incorporated into deadly power gems. After about 5 moves, the counter gems change into regular gems.

The other very critical property is that power gems broken higher up on the screen do more much more damage (send many more counter gems) than gems broken at the bottom of the screen. So consider what attacking is actually like in this game. Attacks are really only temporarily damaging, until the counter gems turn into regular gems. At that point, the opponent will probably be able to incorporate the gems into their own plans, since the opponent knows your drop pattern. Even if the opponent isn't able to benefit from your attack in that way, he can still "dig himself out" of trouble by breaking all the stuff you sent him. By filling up his screen most of the way you've basically given him more potential ammunition to fire at you. What's more, as he is nearest to death, his attacks will be the most damaging due to the height bonus. Gems broken at the very top of the screen do significant damage.

Puzzle Fighter has the extremely unusual property that "almost losing" looks exactly like "almost winning." Let's say you break a whole slew of power gems and send a large attack at your opponent. You're screen is now almost empty. You're winning right? His screen is nearly to the top---almost full. He's losing, right? Well, he is on the verge of losing, but he has all the ammunition and he has the height bonus, whereas you have almost nothing left to defend with. In effect, your opponent is both "losing" and "winning" at the same time. Very curious, indeed!

Ken (left) was close to losing, but he got the yellow crash gem he needed just in time. Donovan (right) will lose.

It turns out the best way to play Puzzle Fighter is to very carefully never attack until you can make it count. All those little jabs you make just help the opponent in the long run. You've got to save up for a huge, 1-2 punch. You need to send a big attack that almost kills them, then immediately send another attack that finishes them off. 1, 2! The point is that Puzzle Fighter is a high energy, edge-of-your seat game. Your opponent very often has enough attack to kill you, so you have to have enough defense to stop them. Whenever the scales start to tip in your opponent's favor, they have also, weirdly, tipped in your favor as well, in some sense. A game of Puzzle Fighter is never over until the last moment. Comebacks are the name of the game, and the excitement goes to the very last second almost every time.

If you know of any other games that use perpetual comeback, I'd love to hear about them. It's a powerful and dangerous concept that could very easily ruin a game, even though it shines in Puzzle Fighter. It spurred me to design a puzzle game based on Puzzle Fighter to capture the genius of perpetual comeback. I'd also enjoy designing perpetual comeback into other types of games---a challenging notion indeed! Any thoughts, fellow designers?