Team Balance Theory

A thesis of team balance theory with regards to Warcraft 3: DoTA


The genre of Real Time Strategy (hereafter referred to RTS) is currently dominated by the popular game Warcraft 3, the third installment in Blizzards epic story of fantastical warfare. One thing that has encouraged continued play of the trilogy has been the inclusion of a map editor, allowing the player to custom design his/her own levels, game types, and campaigns. Combining the elegance of the map editor with moderate coding skills results in a polished, extremely engaging variant of W3, called Defense of the Ancients.

The premise is thus: two magical forces, the Scourge and the Sentinel, are battling between each other. There are two bases, in opposite corners of the map, consisting of unit producing structures (that constantly churn out new foot soldiers to send against the enemy), Towers, both benign and malevolent (benign towers provide resources in able to produce foot soldiers, malevolent protect the base and the three lanes of attack leading to each base with powerful ranged attacks), and a key structure, which causes the end of the game in favor of whichever side destroys it.

Aside from the average foot-soldiers both factions spit out (commonly called "Creeps"), there are heroes, controlled by the players, who form focal points of combat and interest during the game. Heroes are, by definition, more adept at combat than a lone creep, and progress through a series of levels. At each level you may increase the power of one of the three basic or one advanced skills, or increase by 1 point each the heroes three attributes (strength, agility, and intelligence). The three basic skills may be increased a maximum of four times, and no basic skill can be at a level more than half the hero's overall level, except when beginning the game, when you are allotted one skill point (and are at level 1), and the advanced skill can be upgraded only three times (once at level 6, once at level 11, and finally at level 16).

Each hero reaches a maximum level at level 25, at which point the three basic skills are maxed out at 4 each, the advanced at 3, and the small increases to attributes add up to +10 to all skill levels. The heroes progress through the levels through a system of exponential experience earning: killing other heroes and creeps earns both money and experience, with opposing heroes offering significantly more of it than do the creeps. Killing towers offers little experience, but a large amount of gold, and killing other buildings offers little gold and little or no experience.

Some familiarity with the vernacular is requisite to an understanding of this essay.

"Farming" is a term that means steady killing of creeps, to earn money and experience. This is a common practice early and late in the game, when all the players race to acquire level or item superiority.

"Pushing a Lane" means to aggressively attack along one of the three fortified paths of the forest. Two lanes circle around the sides of the map, with one cutting through the middle. There are minor paths through the woods, and a frozen river that intersects all three lanes across the middle of the map, but the computer-controlled creeps stick to the three main lanes, and the most money and experience is generally to be earned there.

"Healing" or "Buying" is when a hero is either low on health or fat with the spoils of combat, and returns to base to replenish life points or to spend money on items.

"Items" are both complicated and essential to winning the game. They can be bought from two characters and a store in town, or a "special store" each team has in the woods. None of these characters or stores can be destroyed. Items provide opportunity to nullify or reduce the effects of opponents' specific spells or abilities, and also to drastically increase your own power or abilities independent of the level scale.

"Recipes" are a combination of items that result in a super-item. Combining smaller items with a spell bought from one of the characters in town results in an item that combines the aspects of all the involved items, and adds something special. Sometimes the effects of the item are extremely powerful; also, sometimes the effects of the super item are only nearly related to the items used in its construction. There are three levels of reciped items, with the highest level being the most powerful and expensive to build.

"Recharge" rate is the period of time between spell use to spell use, during which you may not cast a spell or use an ability. The recharge rate varies from hero to hero and spell to spell, and can be extremely short (Lycanthropes may use Howl every 1.5 seconds) or very long (the Faceless Void may only Time Stop every 5 minutes).

The main concern of this essay is to determine how to balance teams in a game of DoTA. Currently, my dorm floor has two quite excellent players, who are very familiar with the items, their prices, the possible characters and their classes, many of the recipes, and various tactics for maximizing profit and experience at all periods of the game. Jon and Jack are, far and away, the kick-assingest players we've got.

Myself, Jesse, and Tony are developing, good players. We understand the heroes we are, we understand basic recipes, but our play falls apart at higher levels and our tactics are less than ideal.

Ross, Sarah, and Noe are mediocre (to be fair, Noe might not be, but I've only played one game with him). Mediocre might be a very polite way of saying Sarah has no experience with this game and is horrible. I do not mean to slight her character or person, only to say that she is loath to waste as much time on this game as we do.

Now, how to balance the teams? Jon and Jack have lately been quite fond of 4 on 2 games; them versus whoever else wants to play. I originally went along with this, but it readily became apparent that a 2-to-1 numbers advantage did very little to impact the outcome of the game.

I put forth that numbers advantage means very little in DoTA unless the players have all reached a thresshold of skill level, at which point they realize how to attack, defend, buy, and optimize their character in a passably optimal manner.

One major concern in the early game is farming. Early levels are both dangerous for teh heroes and very fruitful, because you earn, relative to the amount of experience you need to advance a level, a large amount of experience for each creep that perishes in your proximity. Good players can stay close to a combat without being involved in it, and know how to position their character to take further advantage of creeps by killing them himself (and therefore earning more money and experience) than merely letting his inferiors do the work. Good players know how to take advantage of the staggered waves of foot soldiers to mount the most effective attacks on towers (juggernauts in a sea of ants, if you will, when compared to creeps), and they also know how to conserve life points to prevent the need to return to base and heal (a waste of time and a loss of opportunity to level and earn money).

Good players also understand player combat, and how to track the location of other heroes and their status. This is of advantage because surprise attacks: cornering a hero between a group of creeps, a tower, or another hero, provide an opportunity to steal money from your opponent, gain large amounts of experience, and halt your opponent's plans, as with every level the amount of time a player must wait until he can respawn after death increases. Player killing of other players (pk'ing), is a major part of DoTA strategy, and common advice given to new players (n00bs) is "Kill shit, don't die."

That such advantage can be garnered from killing other players is integral to my theory that having extra, inexperienced players on a team in a game of DoTA is a disadvantage. Basically, having someone that doesn't know what they're doing on your team is giving your opponent (who will have no problem dispatching them) free experience and gold. In addition, each team has only so much gold to buy their initial items with, and having more players on a team dilutes your starting items. The players on a team four strong cannot each afford Power Treads (a low-level recipe that is excellent for early in the game), where their opponent can probably afford Power Treads AND the beginnings to a Perseverence (a mid-level recipe that is excellent in the early and mid-game, and is included in multiple, powerful recipes that are affordable later in the game.

Not only does a smaller team composed of experienced, better players start out with more gold and therefore a significant individual advantage over the members of the other team, and not only do the players level up more quickly and earn more money due to player kills, but the system for splitting experience is skewed in their direction, also.

When two players from the same team are near an enemy unit when it dies, the experience is split between them. Three major lanes of attack means three players from each team can be active in combat without a dilution of experience points, but having an additional hero active in a lane roughly halves the levelling potential of the individuals in that lane, which in turn allows for easier player kills for the other team, and decreases the ability of those players to threaten the opposing base or push aggressively through towers in the mid- to late-game.

Finally, the issue of items must rear its head. Unless you are familiar with the higher and mid-level recipes (the low-level recipes are almost never worth investing in), your ability to compete mid- and late-game is severely hampered. A level 18 character with good abilities and crappy, cheap items is only just a match for a level 18 character with crappy abilities and good, well thought out item selection.

In conclusion, the only way to balance teams initially, and in so doing develop an adept and dynamic player-base is to, basically, do draft or "playground style" picks. The two best players should be on opposite teams, and they should alternate picking. Ending up with more players is not always a disadvantage, but the only way to get out of the situation of it always being one is to have dynamic games where the best player on each team can guide the worst and explain the events of the game while they experience them. A p/maternalistic style of play and a stratified skill level when organizing teams is the only way to have truly balanced teams now and in the future.

Reviews for Team Balance Theory: A thesis of team balance theory with regards to Warcraft 3: DoTA

"I am awestruck. Possibly flabbergasted."
-Aaron

"I'd like to read it."
-Ross

"You have far too much time on your hands"
-Jesse