PSA: Before you propose a civ balance change

I’ve seen a lot of balance changes proposed to the civilizations both here, and on Reddit. I think that it’s not only fine, but useful for the community to share and discuss these. It’ll contribute to both the health of the game, and the community over time. However, I find a lot of these discussions severely lacking in many ways. Here are a couple of things I think will greatly contribute to the way these discussions are had. All of these are just my opinions, and I am just a random person on the internet. So, please don’t take it too personally if you have disagreements with anything I say here.

I think that there are a few types of information you should include, whenever you propose a balance change.

  1. Win rates for Elo Categories 0-1200, and 1200+ for relevant categories

People broadly recognize 4 main map categories in AoE2. Open, Closed, Hybrid, and Water Maps. Before proposing a change, please look up the data and include it in the post. You should narrow it down the map category the buff should apply to. Bengalis, for example, have 43% win rate on arabia at 1200+ Elo. Before proposing a buff though, you also need to consider that they are an absolute top tier civ on closed maps. is a fantastic website that gives excellent statistical information for civs. If you want to zoom in on a specific elo range, a specific balance patch, or a specific map, you can use

Without this, we get things like people asking to buff Turks, who are the civ with #2 highest win rate right now. I’ve even seen people proposing buffs for Franks, or saying that eco buff for vietnamese on the most recent patch is “too much”.

Note that there might be cases where these stats are impossible or difficult to obtain. Bohemians are not that strong on closed maps, till they get their deathball rolling. Once they do, it can be pretty hard to stop. However, you cannot find numbers for games after they get to their deathball, and before. The data is available for most cases though, and should be included whenever possible.

2. What you think is the main issue with the civilization is, and support it with evidence if you can

There are a few main types of issues, from what I have observed.

  1. Either the civ is extremely vulnerable to one particular strategy, or is really strong with one strategy
  2. A unit of the civ is too weak/too strong
  3. The civ’s economy is too weak/too strong
  4. The civ has a hole in its tech tree, which leaves a vulnerability during all/many stages of the game

This part is harder to explain purely through evidence. It is also pretty hard to gather evidence. As such, you need to rely on words of pros, comparison with other civs, and appeals to general understanding of how the game works. When you have issues pertaining to a specific unit, you should use scenario editor tests, as we will discuss below.

I will use Dravidians as an example here, because they have 3/4 of these problems. They are extremely vulnerable to knight-skirm-mangonel pushes. I want to post Hera’s gameplay video to demonstrate the issue, but this forum won’t let me. I’ll try posting it at the bottom, and see if it works. They don’t have any land eco bonuses aside from 200 wood, and the barracks discount. They also lack all the late game mining and farm upgrades, meaning, they have one of the worst late game ecos. And, they have a hole in the tech tree in the form of a raiding unit.

Whether you agree with my arguments is not the point here. I am merely trying to explain my point of view, and why I think the way I do. You might think that one or more of these is not an issue, or might point out an issue that I haven’t thought about.

However, what is important is that you draw a line from these arguments to winning a game. Spanish not having crossbow is not an issue unless you can show how it is, and why it should change.

3. Scenario editor tests, and recordings of your games
People often have issues with how strong a unit is, and how strong a composition is. In such cases, you can either do a scenario editor test, or post recordings of your games where you have encountered the unit/composition.

Scenario editor tests need their own explanation, as doing equal resource post-imp tests aren’t usually good enough. What you actually need is an age specific test, with controls on the technologies researched, and a range of comparisons. Crossbows vs Mamelukes can be an interesting test. You need to do it with bodkin and leather archer armour researched/not researched, and same for Mamelukes. Then, you also need to do it roughly resource equal, or slightly more on one side or the other. Yes, doing it this way will take a lot of time. However, there is no other way to gain a comprehensive understanding of the unit, other than just playing a lot.

If you think a unit is too strong/too weak, you need to test it under common scenarios. So, test it against all the trash units, crossbows, knights, camels, and siege. If all these are not necessary, test it for the most relevant cases. There are also a few websites for this. However, I wouldn’t rely on them too much.

4. Finally, your solution (if you have one)
Let me make one thing clear, it is totally fine to point out an issue without proposing a solution. If you can’t think of one, but you feel strongly about a problem, post it. I’m sure we have enough creative minds here who can work on them.

Other than that, I have very little to add here. Your solution should solve the problem, but there are a huge number of solutions to every potential issue. People aren’t going to agree, and there are going to be arguments. Just put forward your best solution, and go from there. Just keep in mind that your solution, if it is a bonus, cannot overlap with another bonus. It can, however, overlap with another UT, if the bonus is weaker.

Lastly, it might be a good idea to put into the context of a 1v1 open map game, a team game, and a 1v1 closed map game.

Closed maps games usually start with FC, then a castle drop/monk-siege push/monk-pike/monk-scout relic hunt. Then, you go for an omega boom or 1 tc ram/mangonel push, and late game trash unit/power unit/ siege comp. If you want to recommend a balance change, you should try to point out what role it will play in this structure of game play. If you don’t know this structure, I’m afraid that you aren’t ready to propose a balance change.

I’m sure that everyone here will have differing opinions of this, but I think that following these guidelines can vastly improve balancing discussions. Please comment your suggestions as well.

Hera’s Gameplay: ######################################## Combat simulator: AoE 2 Database


Great post. Lots of valid points. Many suggestions seem to be posted based on their own experience: they just get beaten a unit or civ and wants a nerf. Instead of asking for help, they are asking for a nerf, without really understanding the civs/balance. As result people make terrible suggestion without any good arguments.


That’s a neat explanation of how someone can propose balance changes while reducing their personal bias.
Personally I’d like to add a few things to this:

  1. When checking for stats, its better to consider not just the current live patch but also previous patches. Two reasons for it is avert small data inconsistencies and a civ underperforming in 1 patch might be more related to its matchup against the civs popular in the current live patch.
  2. Tournament stats being taken into consideration. Obviously this is not easy to comprehend but if a civ that you want buffed has very good “games played” in tournaments over years, it could indicate a steeper learning curve. And a civ you want nerfed having mediocre tournament stats could indicate some obvious weakness that hasn’t trickled down yet to the lower elos yet.
  3. When you do decide to some editor test to prove your point please try to make it more realistic to in-game situations. Like 30 cavalry unique units vs 40 infantry unique unit or 50 ranged foot units instead of 150 vs 200 or 10 vs 40 to even “gold” cost. Reason being when the player has 10 cavalry unique units, they’re not supposed to engage 40 slower units and should use the speed to run away.
1 Like