(Upcoming) Ant Analysis

#1
Hello everyone! I'm working on a testing-based analysis of the Army Ant Media, the first of a series of "reports" I intend to make about the balance of various units currently in the game. The reason I'm posting now instead of when it's finished is that I need to make sure I'm not missing any criteria before I collect too much data. Besides, if you have any interest in the results, it would be better if I asked now rather than later. Here are the general assumptions and rules for these reports:
  1. The baseline unit is the Black Ant Soldier, which is assumed to be the 'average' ant. All units will be compared it when determining how powerful any given unit is.
  2. All units will be scored (with a percentage relative to the standard unit) based on their cost-efficiency in three catergories: Combat, Utility, and Synergy. Each category's score will be relative to the Black Ant Soldier (which has a score of '100' or '100%' due to it being the base unit). For example, a unit that does twice as well as a Black Ant Soldier at fighting will receive a score of 200.
    1. "Combat" refers to how well a unit performs in combat. Units with a high score in this department will do well in Extra Levels, the Formicarium, and other game modes where utility functions are of minimal importance. It will be determined by ant vs ant combat testing (Levels 1-3 vs Black Ant Soldiers of the same level, graded based on the number of survivors on the winning side relative to the losing), ant vs creature combat testing (equivalent forces of Level 1-3 Tested ants and Level 1-3 Black Ant Soldiers will both face off against a group of creatures of equivalent strength; they will be graded on how fast the creatures win or lose and how many survivors are left after the encounter), and freeplay combat testing (which ant type wins more games 1v1 against a freeplay AI, with a total of 10 such 1v1s each). I may also analyze them for different subcategories of the three at a later date, mostly since combat with Ants and combat with Creatures are two very different things.
    2. "Utility" refers to how well a unit can be used for non-combat tasks, such as harvesting and construction. Things such as speed and carrying capacity will be considered in this department.
    3. "Synergy" refers to how well a unit does in each of the previous two categories with ally support. To determine this, a force consisting of a 50-50 ratio of the unit in question and standard Black Ant Soldiers will face against an 100% Black Ant Soldier force, both in the "Combat" and "Utility" departments.
  3. After all categories have been scored, the unit's overall score is determined by averaging the three values, and then a new Effective Cost is suggested by getting the unit's score divided by 100 (to make it a percentage-based multiplier) and multiplying it by the unit's previous Effective Cost. The result is a new cost that is scaled to be equivalent to a Black Ant in terms of power level.

Terms Used:
  • "Effective Cost": The overall cost of a unit, with hatching costs considered. Can be calculated by taking the average of the unit's build cost and ten times its respawn cost.
  • "Expected Cost Ratio": Refers to the LCM (Least Common Multiple) of the unit's Effective Cost and the Black Ant Soldier's Effective Cost (which is a flat 50), a value that represents how many each of the chosen unit and Black Ant Soldiers would be needed to face each other in a equal-cost battle.
Long story short, this is a LOT of math. However, you guys won't have to deal with that! :D
As long as the tests are accurate, all you'd need to know is the new suggested cost, since you can compare it to the current cost to see how good a unit actually is. But I won't stop you from reading the extra data, either, so yeah. Anyhow, thanks for reading!


Disclaimer: The game would be really boring if every ant did the same thing at the same level of performance. Since some ants have special effects and/or more specialized stat spreads, however, such a bland situation is rather unlikely to occur. Thus, this shouldn't be that much of a concern.
 
Top