Balancing with Tiers

Melf_Himself and I have been having a discussion in the comments of one of my DotA MMO entries.

I've not played DotA, but it's a bad idea to include a large percentage of classes that would never be chosen by competitive players. 10% is a very low number and is much lower than what you'll see in well balanced fighting games. It really makes things hard for new players, making the amount of information they have to possess before they can start playing the actual game a lot higher. Ideally any choice that you make before a match starts should not affect the likelihood that you will win the match, only how much fun you will have.

I'm not sure what you mean by "painstakingly difficult to balance because there are so few". The more classes there are, the harder it is to balance them all.

I draw a lot of my design and game-playing principles from David Sirlin. Not only his Playing to Win articles, but also his Multiplayer Game sections.

DotA has over 90 heroes. I would say about 20 (22%) of them are viable for competitive play. My favorite fighting game of all time is Naruto GNT 4 and only 9 out of 25 (36%) characters are viable competitively. A more Western SSB:Melee has 5 characters out of 26 (19%) at the top tier.

DotA isn't a very newbie friendly game, partially because of the reason you described with tiered heroes. That doesn't mean DotA isn't fun or isn't competitive--quite the contrary.

If DotA had fewer heroes, say 10, then all of them need to be viable. That is much more difficult than making 50 heroes and letting the players figure out which ones are the best. The designer obviously needs to make sure that none are dominant and none are dominated, and over time he can try to pull their differences in power closer together. But with "tiered-balancing" the success of the game isn't dependent on the viability of every character.

You are quite right when you say that choices made before the match shouldn't affect your chances to win, which is why MMOs are very difficult to balance. Every class needs to be viable because players have invested resources into the character and are locked in to that choice. Compare this to DotA or a fighting game where players can see what character their opponent (or team) has chosen and react appropriately. I love drafting in DotA and MTG for this very reason. There is a game that takes place before the game starts.

Many moons ago when Blizzard announced that switching gear in Arena was going to be disallowed, I was heartbroken. I have never played Arena, but here was a counter to team composition, and they completely removed it! They had valid reasons for doing that, but it removed an entire level of depth to Arena PvP. Players were not locked into their stat choices and could react appropriately to the opposition.

So for the DotA MMO, we can have hundreds of abilities and scores of ability groups. Allow the player to react to his opposition (either through the meta-game or with a pre-match draft system). Abilities and builds will fall into tiers (WoW talent builds are tiered, by the way). As long as no one combination is dominant, and players have counters to abilities, builds, and team compositions, then the game will be fun and competitive.

The Role of Chance

“Cutting down the role of chance” is one of the more overlooked of my 10 Points for the MMORPG Revolution, so I will tackle it first.

The first step towards cutting down the role of chance is to understand the role of chance. Randomness is a root preconception that players have in MMORPGs, and, to seasoned RPG players, it seems absolutely necessary. In actuality, this assumption need not hold. In this post, I will review the roles of chance and why we want to reduce its role in upcoming MMORPGs. In a later post I will offer suggestions as to how to cut down the role of chance in specific instances of each of these roles.

(As a thought experiment, I encourage you to think about how you’d design an MMORPG that does not use a random number generator. We’ll see if our ideas match up when I get to the second part of this discussion later on.)

Here are the three roles chance plays in MMORPGs. (If you can think of more, please comment! I want to make this article as complete as possible without getting overly concrete and nitpicky.)

Random Numbers Simulate Character Ability

In role playing games of all kinds, random numbers are used to determine the outcome of events because they can resolve actions independently of player skill. If a character needs to roll a 15 or more on a 20-sided die to successfully perform an action, that character has less skill than the character who can roll a 10 or more and succeed. No level of player planning can overcome his characters limitations without some luck—this is appropriate, because ability gating and character growth are important, fundamental parts of RPGs.

Random Numbers Model Unforeseen Factors

Random numbers model the effects of actions that are outside of our scope of observation. In real life, we notice events that seem to occur randomly, but this is just a symptom of our subjective viewpoint. Because we don’t see the causes of certain events, we assume that they’re random. Under further analysis, we find that human-level events have complex cause-and-effect relationships that, if followed prior to the “random” event, make the event an obvious effect. As individuals, we can perceive so few of these strands of cause-and-effect that it is impossible to get our heads around most of the complex activities that we are subject to in a given day. Hence, the idea of "randomness" or "chance" in human-level interactions and one of the roles that chance plays in RPG systems.

Random Numbers Provide Variety

Random Numbers are used in generating loot and monster spawns and can be used for determining the behavior of enemies (if there are multiple equally viable AI strategies). This randomness provides variety in the game world and is partially responsible for how addictive loot-heavy games are, Diablo II especially.

Entire dungeons can be generated on the fly using random numbers. Diablo II did this, but there are more interesting procedurally generated maps in roguelike games. Much replayability in roguelikes is due to the ever-different dungeons and maps that present the play with a seemingly endless number of interesting tactical situations and fiendishly difficult dungeon levels.

There’s a lot of work done in procedurally generated content, as well. Most of this stuff is rather technical, but goes to show that there are a lot of people working on effective ways to use random numbers to generate content in games.

Motivation For Reducing the Role of Chance

Why do we want to get rid of as many random number-determined actions as possible? With few exceptions this mechanic has been the core of RPGs for as long as RPGs have existed.

If we want to give the player the ability to change the world and affect other characters in meaningful ways, we need to make decision-making the central process. We need to allow decision-making to determine the outcome of events, not chance. If we provide the player with appropriate information about the decision they are going to make, there’s no reason to build into the game a chance of that decision failing outright due to a bad die roll (something that’s completely out of any player’s hands). I’m not suggesting that we remove failure from MMOs, but instead that we remove unavoidable failure from individual player’s actions.

In a system that allows player success to the best decision-makers, players can easily see what they have done wrong and accept their losses as lessons for the future. If the dice screw you, you’re going to be upset at the game—nothing is learned. If you plan poorly or make a bad decision, you can identify that without the obscuring power of dice rolls and make intelligent decisions using that input in the future, leading to player skill and player learning being the most important facets of the game, not time invested and luck.

Horizontal Progression and Hybridization

Horizontal and vertical progression seems to mean a lot of different things to a lot of different people. I caught wind of Word of Shadow's definition and was going to write a comment. But looking around the Internet, it felt as if I needed more explanation for my thoughts. Personally I was focusing my definition too narrowly, but I have my reasons, as I will explain. I wasn't thinking about it from a high level point of view, but let's start there.

In the broadest sense, if combat were the only viable path to end-game, then the system would be vertical. Being forced into leveling and combat feels confining for some people, and they'd prefer a "horizontal system". I really don't know what that means. I've seen definitions ranging from a player-skill only system to a system in which there are multiple progression paths to end-game--combat isn't the only means to the ends.

To me, horizontal progression is more of a hybridization technique. If you have a player-skill system, players are still going to min-max implicitly and pick a role. If you have multiple paths to the end game, you are still going to pick one and vertically climb the ladder. A horizontal system means that as you progress, whether through leveling and talents or through some other means, you gain skill equally across all roles.

I was originally thinking of horizontal progression in a very specific realm of play--namely combat. The reason for this is that if you have more horizontally progressed characters, you have more true hybrids. Having the same character being able to fill multiple roles more easily facilitates grouping. If every class is not only a damage dealer, but also a healer or a tank, and they progress horizontally preserving their hybridization, then you don't have to wait around for hours looking for a tank and a healer. You are at worst looking for a single role, which half of the players online can fulfill.

Perhaps an example is necessary. Let's take a look at WoW's Druid. This is supposedly a hybrid class. If you ask your feral Druid friend to shift out of Bear form and heal for this dungeon run, he will laugh at you. "I thought you were a hybrid," you retort. "I'd have to respec," responds your friend. Where is the hybridization? It was lost in the vertical progression scheme of the talent trees. WoW Druids aren't hybrids; very few classes in WoW are hybrids.

Vertical progression is much easier to design than horizontal progression. Make a tanking tree, a healing tree, and a damage tree; done. We're going to switch gears from Druid to Death Knight because Druids are too far gone to be salvaged. Death Knights are "tank and damage dealing hybrids" yet they need to spec for 1 and only 1 role. Their talent trees are labeled Blood, Frost, and Unholy. From a tanking perspective, Blood gives more health and healing powers, Frost gives more armor, and Unholy helps to reduce burst damage.

There are talents in these 3 trees which are specifically for either tanking or for doing damage, but very few for both. If we treat Blood, Frost, and Unholy as different flavors or play styles of the same class, we have an easier time thinking about horizontal progression. Blood will have a vampiric and sacrificial feel about it. Frost will concern itself with slowing and nuke capabilities. And Unholy will deal with damage over time and other debuffs. Thus, as a player spends talent points in the tree (and play style) he chooses, he will gain power equally in both tanking and damage capabilities. He will be unable to min-max into a role.

Now all Death Knight characters, regardless of what specialization the player chooses, will be a tank and damage dealing hybrid. If your group needs a tank, invite a DK and ask him to tank. The player must rely solely on his player-ability to tank, not his gear nor spec. He has progressed his character explicitly down a horizontal path of his choosing and is not gimped in any way because of it.

Cyanide’s Blood Bowl: A Two-Faced Monster

I’ve been playing Blood Bowl--the PC game made by Cyanide--recently. There’s a lot to be learned from both the base Blood Bowl tabletop game and Cyanide’s implementation, so I’m going to make a series of posts discussing both the Cyanide implementation and the tabletop game. This post is an introduction to the game itself (I’ll try to let you know as much as is necessary to read my review of Cyanide’s implementation) as well as a brief review of Cyanide’s PC game.

If you want to skip the intro and the specific issues I have with Cyanide's game, you can move on down to the summary and my rating at the end of this article.

A 1,000-foot view of Blood Bowl (The Tabletop Game)

Cyanide’s game of the same name is a rather faithful adaptation of the tabletop game Blood Bowl. In Blood Bowl, you control a team of fantasy characters, each team being confined to one race (standard fare like Elves, Dwarves, and Humans as well as some Warhammer races like Skaven and Chaos), as they play through single games or seasons of a sport that is vaguely similar to American Football. The sport of Blood Bowl has a small number of rules, you’re only forbidden from stomping on players when they’re down, and the ref doesn’t even eject players reliably for doing this—or the ref may be bribed by either side to look the other way when players on that side do something untoward. Players alse have to stay on the field because if they stray beyond those confines the fans will beat them to near-death, putting them out of commission for as long as a couple games or perhaps even killing them.

Blood Bowl is a turn-based game. Teams alternate taking turns, but turns can end abruptly when certain unfortunately events happen, like when a player fails to pick up the ball, fails to complete an attempted pass, or tries to block an opponent only to get knocked down himself. Players move around the field on a square grid. Only one player can be on a given square at once. Players can perform a few basic actions: move a number of squares up to their movement allowance, pick up the ball, throw the ball (and have someone else catch it on either team), block an opponent, or foul a prone opponent. These actions are resolved by comparing the attributes of the players involved to see what chances of success the action has and rolling dice to see if the action is successful. The game uses two types of six-sided dice, a standard pair of dice you’re familiar with and special “block” dice that have faces representing the different outcomes possible when a block action resolves.

That may sound like a lot, but it’s actually quite simple for a tabletop game. It takes only 15 pages of standard-sized paper for the current rulebook to relate the rules of the tabletop game (the core rules—there are more rules and information after the core rules are enumerated, but you don’t strictly need to read them in order to play), and that is including numerous illustrations, sidebards, bits of world trivia, and examples of play. Compare this to Dungeons and Dragons, which requires a 200+ page rulebook, most of which you’ll be referencing at some point if you want to play the game with a full group of five players and a dungeon master.

The game’s rules have been through five significant iterations; It’s quite mature. Most of the sides are balanced, though some of the sides are designed to be more difficult to win with than others. The strategic depth of the game has impressed me. I’ve read through the current iteration of the rulebook and found myself excited to play the game. I’ll go deeper into the game mechanics as I analyze them in future articles.

The part that Cyanide had no real part in—the tabletop rules—are awesome. These rules are the good face of the game. The other face is everything Cyanide has put around the game to adapt it to the PC.

The Uglier of the Two Faces

Cyanide has implemented the rules as they are written with a few exceptions. The game seems to flow as the tabletop game would. This is great, because the years of players banging on the rules and repeated revision has produced a great, fun-to-play set of rules.

Once you get beyond the rules, things get worse.

Graphics

The 3D graphics are a little bit behind the times, but not terrible. They look good when they are relatively static. Animations are passable. You will watch these animations hundreds of times, though. There are not multiple animations for frequently performed game actions, so they get boring quickly. As I played more of the game, I began to see the graphics as getting in the way of my experience. I wanted to be able to easily differentiate player types and I couldn’t do that relying solely on the graphics. Some characters looked very distinct, while among others it was hard to tell without staring for a couple seconds or longer.

The game features “instant replays” of important plays. This is moronic because you see the one player either bashing another to the ground or running into the endzone the same way you’ve seen them do those things a hundred times before, but from a different camera angle which makes the graphics and animations no less boring.

Summary: The graphics are passable as graphics, but are not as useful as well-illustrated 2D tokens would have been.

Interface and Presentation

The interface is passable. The graphics on the menus are flavorful and well-done, but the text is often poorly localized to English, which can lead to confusion. The campaign screen that allows you to negotiate with sponsors is one of the worst offenders. The interface elements just don’t make sense and there aren’t tooltips or any helpers to assist you in figuring out what the hell the game means by some of the options presented (like a “Rankings” option that allows you to select a number between 1 and 15: It’s completely unclear what it’s supposed to mean).

When you’re in a match, because some characters take up a couple squares from different perspectives it isn’t difficult to mistakenly click on a player instead of the ground behind him. One time I just couldn’t select the ground behind an opposing player, the game made me block the player which failed and caused a costly turn-over.

There’s also a nauseating cinematic camera that swings around ridiculously on your opponents turn to try to “dramatically” show you what’s going on.

The camera pans across the benches of both teams before any kick-off is played. I find this to be a complete waste of time; the game would’ve been much better off if it had simply provided you an interface letting you know that certain players are knocked out or badly injured or in reserve.

Everything Else

Outside of playing individual matches, Cyanide has done a questionable job of implementing the league play that contextualizes the matches. I doubt this works as it should, because there are enormous money imbalances between human and CPU teams. I have 160,000 gold after winning a cup. A human team that finished second in the first cup now has 600,000+ gold. I don’t understand how when I’m frugal I end up with 5x less gold than my opponents. This would be more annoying if the teams actually spend their money wisely, which they do not seem to.

Out of match AI is suspect and doesn’t make sensible decisions with its money, whereas the in-match AI is generally passable except for not understanding urgency when they’re behind or tied late in the game.

Cyanide has not implemented the inducement system correctly, which makes or some whacky team compositions if teams have low overall values. The Inducement system is intended to balance out team values, giving the weaker team at boost in competitiveness so that the lower-value side isn’t always getting creamed. For teams that cannot field a full 11 players, Cyanide has failed to implement “journeyman” basic players that fill out the roster and are counted towards team value before inducement money is handed out. This leads to situations where a team will be able to buy star players and a lot more bonuses than they otherwise should have. Perhaps this is to artificially make the game harder for the human player, but I don’t really see the purpose of making a balancing mechanic unbalanced in the opposite direction. It comes off as being hacky at best. I’d be surprised if this slip-up was intentional.

This game has loading periods that are over a minute in length! This only happens when you go into the match engine. Still, this is pretty much inexcusable for a modern game. Crysis can load in less than a minute on my old system, why can’t a game with comparatively little in the way of graphics load within a few seconds? It’s not like this is a Playstation One game.

Sound? Well, after one match I was thoroughly sick of hearing the sound in this game and a turned it off. There is some commentary voice-over while the game happens, but it gets repetitive fast. I don’t think sound is worth noting in this game. It has no effect on my judgment of the game.

Summary

After 10 hours of playing this game, I feel like the tabletop rules are the strongest part. The implementation of them at the match level is good enough to be very fun and rewarding. Unfortunately, just about everything that Cyanide did themselves outside of match play is either half-assed, poorly implemented, incorrectly implemented, hard to use, or just plain annoying.

I give the Blood Bowl rules a solid 8.8 out of 10.

I give Cyanide’s contribution 5.9 out of 10. They are above a 5 only because they implemented the match play very close to the tabletop rules.

Overall, Cyanide’s Blood Bowl gets a 7.5 out of 10.

(Ratings: Below 5 is not worth even thinking about playing. Around 6 is where the game is good enough for hardcore fans to eke out some enjoyment. 7 is average. 8 is worth playing, perhaps justifying a purchase. 9 is a very enjoyable game that has a few minor flaws but should be experienced and probably has at least 50 hours of game play. 10 is practically unreachable, it’s reserved for masterpiece games that can stand hundreds of hours of play without enjoyment fading noticeably.)

Everyone, the Mountain Movers

The problem with MMORPGs is that everyone wants to be the hero impacting the world. It's not too fun if you're just a cog in the machine, told of the glorious deeds of others. You'd rather be the one to swoop in and save the princess, or defeat the invaders, or thwart the plan to conquer the galaxy.

How does a game designer make sure everyone has their moment in the spotlight? How does he make every single player important enough so that they not only enjoy themselves, but also throw money at him? This is pretty easy to do with theme-park MMOs; just build the raid roller coasters. Unfortunately, the world ends up being pretty static as a result. Every update cycle, the players might get a new ride, but they are powerless in influencing the course of events in the world.

How does a game designer make sure players can change the world? They create the tools necessary for players to fight each other for land and resources. One guild becomes the invaders, one becomes the evil empire, and another, the good guys. Unfortunately, only a handful of players are at the helm. The average player isn't the one to save the world; he's just a peon in the plan. The world might change, but the player himself did very little to make it happen.

Designers (and critics) have traditionally seen the player-hero and dynamic-world characteristics as trade offs--as being mutually exclusive. Games are either hero-centric (theme-parks) while giving up world dynamics, or they are dynamic (sandboxes) while giving up hero-centricity. What if this was a false dichotomy, and we can actually have both? Everyone is the hero (or villain) saving (or destroying) the world. A goal should be to make one player the center figure in a story told by another player.

You start by encouraging players to make factions/guilds. The more guilds there are, the fewer players are in any one of them, the more responsibility each player has within that guild. They should also be extremely easy to create. Any group of players from anywhere at anytime should be able to found a guild. No purchase necessary.

Whatever sort of combat is in the game, it needs to be small scale (or to encourage small scale if it's mostly open-world combat), so that each player contributes more and can more easily "save the day". If players die, they should be able to spectate the rest of the battle.

There needs to be NPCs, all of which are completely Common. There can be no desirable attribute about any NPC. No Emperor. No Robin Hood. They are all peasants who need the adventuring players to save them. All of whom grovel at any player who walks by. Whether or not these players actually did any worthy deeds, they are still heroes in the eyes of the NPCs. This illusion is paramount. There should be no "prove yourself to me, private!" or "go talk to the General; he has orders for you".

If guilds can acquire land, then the map of the world should show everyone which land is owned by which guilds. Guilds should be able to designate how their architecture looks in style and color. They should be able to change environmental effects on the lands they hold. E.g. lighting color, weather effects, musical ambiance. Guilds should not only have a political influence on the world, but they should also have a visual influence.

Allow market monopolies. Sure they might be detrimental to the economy, but if players can only get their Zoonie-wigs from Xxpinkponyxx, then that player has become much more prominent in the game. (And if Xxpinkponyxx is charging and arm 'n a leg, perhaps she might get pushed around a bit...) Regional resources are an easy way to accomplish this.

You can give a little guidance to guilds. Perhaps send the guild leader a mission objective once every 3 months. In it, have dynamically generated quests which look similar to Risk's mission system. Make guild SuperMen conquer some land owned by guild Umbrella. Require a guild to get 3 other guilds to declare war against a land-holding guild. Ask a guild to own x square miles in land or a certain region. Amass a certain number of members or resources. Once the guild completes the mission, prominently display the guild and some featured members (those who contributed the most to the mission) where everyone in the game can see. Make those lowly peasants talk about the players and guilds.

These are some very abstracted ideas, and implementation details will vary. Give players the tools to influence the world and some place where they can show off. But make sure there are small pockets of communities. A hero doesn't have to be the one to throw the Ring into the fires of Mount Doom or the one to free everyone from the Matrix. He just has to be the one people tell stories about.

Pay-to-Play becomes Pay-to-be-Disappointed

The goal of this blog is not to discuss the business of MMORPGs, but I think it’s important that we discuss the effect of the pay-to-play model on player expectations.

When you pay every month to play a game, you’re doing two things:

1. You’re validating the enormous amount of time you spend playing the game by “making the game worth the money”.
   
   
2. You’re investing money in the game that you cannot get back. This sum of money only grows.

As long as I pay the monthly fee, I want my money’s worth. As long as I get my money’s worth, I’ll pay the monthly fee. This is an endless cycle! And, in the current forms MMOs are manifest, the cycle is bound to be broken. At some point, the game is no longer going to be worth the money. Everyone gets bored of a game eventually—some people become bored of whole genres of games. But when you regularly pay for a game, at some point you are going to suffer the disappointment of having wasted your money. That experience leaves a bitter taste in your mouth. The taste is made more bitter by the perhaps hundreds of dollars you’ve invested in the game already and how that all has amounted to this disappointment with the game. It’s a completely natural and reasonable process for the player to go through.

In MMOs, You quit a game because you either no longer have the time to play it or because you no longer think it’s worth playing. The result: all former players are either too busy to care any more (or have moved on quietly) or they are bitter or otherwise unhappy. An unhappy former-player is bound to spread his unhappiness to others. Negative word-of-mouth certainly damages games that don’t have multi-million dollar ad budgets; negative word-of-mouth is necessitated by the current business model.

When you pay to play MMOs, you may enjoy them now and continue to enjoy them for years, but you will ultimately leave disappointed. Perhaps Guild Wars’ success can be attributed in part to not generating a wake of disappointed and bitter players. I don’t know if free-to-play games solve this issue, but I do know that it should be addressed if we want the market for MMOs to keep growing and not be eventually subverted by an ever-growing bitter and aggravated populace of former-players.

The MMORPG Revolution (The 10 Points)

I think Syncaine’s got it right in his recent post. MMOs suck as games. As I’ve pointed out before, their mechanics don’t work together with their stories to produce a complete, self-sustaining package. The mechanics lead to the most efficient styles of gameplay being tedious time sinks. Social pressure can keep people playing MMOs for only so long before they completely burn out on the barely-sufficient mechanics and leave. Even the alternatives to theme-park MMOs are weak because they still require from the player a lot of tedious, boring work.

I don’t think these obstacles can be surmounted by evolution—we need revolution. We need games to try whole new styles of immersion, character advancement, combat, crafting, harvesting, and just about everything else.

Below I present you my 10 points for the re-envisaging of MMORPG. I will spend more time going over each of these points and suggesting ways to implement them. A few games have accomplished one or two of these, but I’ve yet to see a game include more than half of them.

The Ten Points:

1. Severely limit vertical character progression. A character should not be inherently more powerful the longer he spends playing the game. A character should be more powerful if the player plans its ability use better or plays the game better.
   
   
2. Focus on horizontal character progression. More abilities that are of relatively equal power become available as characters progress.
   
   
3. Foster non-combat professions and give players meaningful content that doesn’t involve fighting.
   
   
4. Do not force players into PvP. But reward players for doing it—if it’s a higher risk activity, it should be more rewarding.
   
   
5. Death needs to have meaning. The obsession with single-character play has to end. It yields too much investment in one character which leads to severe risk aversion.
   
   
6. Cut down on the role of chance. We can create sophisticated, innovative, strategically deep combat mechanics, we don’t have to rely on random number generators to provide spice in MMORPGs.
   
   
7. When a player is in-game, he should be doing something meaningful. Gameplay has to have effects, even if they are small, on both other characters and the player’s character. These effects build a story and a living world which reward intelligent action and planning.
   
   
8. Let the players make the story. The world can have a rich history that rewards thorough reading, but the present needs to be in control of the players. They need to have the power to make game-level stories happen and to record them in a way that is publicly viewable.
   
   
9. Encourage community-building behavior. Reward players for being in groups, guilds, and factions. Give groups experience bonuses, better item drops, and other perks for participating in the social part of the game. If a game encourages socialization, it more quickly engenders social responsibility in its players—this binds players and keeps them playing the game.
   
   
10. Focus on the player’s in-game experience. If you give the player a rich world, rewarding actions to undertake, and well-constructed game mechanics, your game can still fail. It’s crucial that you present the game to the player with the utmost care. The game has to allow the player to plan and make intelligent decisions: the player must be presented with appropriate information, well laid-out and easy to read and understand.