Why More Asymmetry Between an Increasing Amount Of Civilisations Really Is A Very Bad Idea (+ a potential solution)


I just created this account because after playing the Beta I’ve been reading this forum basically everyday, so why not jump in? :slight_smile:

I have a bit of a mixed feeling about Age of Empires 4. On the one hand I think it looks incredibly promising and I am absolutely planning to buy the game on Steam on the 28th of October (see you all online! :blush:), but I also think that it won’t take much ‘pressure’ towards a certain direction for the game to start suffering from balance issues that are impossible to resolve.

I see a lot of people on this forum calling for an increase in asymmetry between the different civilisations while simultaneously calling for an increase in the amount of civilisations. This worries me, because I am convinced that this would be an extremely bad idea and in this post I will try to explain you why I think that this is the case.

Balance is not about number of civilisations, but about number of possible matchups in combination with the degree of asymmetry

Let’s say that there are two types of RTS games.

1.RTS games that have only a few factions, but which are very asymmetrical, such as Command and Conquer Red Alert 2 and Command and Conquer Generals.

These games only have a few different factions, which leads to the number of possible matchups to be very small. Two factions, for example Allied and Soviets, lead to only one possible matchup (excluding mirrored matches): Allied vs Soviets. Three factions, for example USA, China and GLA, lead to only three different matchups (excl. mirrored): USA vs China; USA vs GLA; and China vs GLA.
In the case of Command and Conquer Red Alert 2, the developers only had to balance one single matchup. This was incredibly easy to do! For example, if the Soviets had an early tank that was too strong and couldn’t be countered by the Allied, they could simply nerf this specific tank and solve the problem. Really quite simple.
However, in the case of Command and Conquer Generals, this was already slightly more complicated, because this game had three different matchups (excl. mirrored). In this case, if China had an early tank that was too strong and couldn’t be countered by the GLA, the developers couldn’t simply decide to nerf this specific Chinese tank. In this case, they also had to take into consideration how this specific change would affect China’s matchup against the USA. Perhaps this one tank was very important for China in order to counter a certain rush strategy by the USA.
However in general, RTS games with only a few factions are relatively easy to balance and can therefore have an extremely high degree of asymmetry.

2.RTS games that have more than 10 factions, but which are very symmetrical, such as AoE2: Age of Kings.

AoE2 got released with 13 factions, leading to 78 different matchups! This is a huge number, because all of these matchups need to be balanced!
This means the following: If the Franks have a horse that is too strong against the Britons, the developers can’t simply decide to nerf the Frankish horse as a result of that, because they would also have to take into consideration how this change would affect the Frank’s matchup against the other 11 (!) factions.
Luckily, AoE2: Age of Kings is a game with only extremely small differences between the factions.

How asymmetry and an increase in the amount of factions ruined a once great game

In the case of Command and Conquer Generals there were only three different matchups at first, meaning that the developers only had to balance three different combinations. This allowed the asymmetry between these factions to be very big. For example; the GLA was the only faction that didn’t have airplanes and its tanks were basically made out of glass, but it still had equal strength against the other two factions.

However, what happened next is also what motivated me to create an account on this forum and to write this post! The creators of the game released an expansion pack called ‘Zero Hour’ in which they decided to add three sub factions (‘Generals’ each of which had his own specialty) for every single main faction, leading to a total of 12 factions! At first this was a very welcome change, because it was very fun to try out all the different ‘Generals’ and see what they could do.

However, from the very start, the online multiplayer was a complete disaster. It just took a while before everyone realised it. Certain matchups were impossible to win or lose. None of the GLA factions had any chance against the USA Airforce General, nobody wanted to play against the USA Superweapon General and the China Nuke General quickly became both ‘Top Tier’ and socially accepted because it could basically handle all the other factions quite well without being too lame itself. The developers of the game at first tried to solve the balance issues, but they never succeeded.

Why did they never succeed? Well, because they had given themselves the impossible task of balancing 66 different matchups within a game with a too high level of faction asymmetry!!
What they had created was simply impossible to fix. They simply couldn’t repair it. The online multiplayer quickly became about countering factions instead of units. The outcomes of games were already decided in the online rooms and playing certain factions was only doable once certain other factions were banned from playing.

For example, you want to play GLA Demolition General, someone enters your hosted room and then picks USA Airforce General. You then have two choices: you either kick out the player and change the title of the game to “1vs1. No USA Airforce General allowed!” or you instead pick China Tank General for its amazing anti-air, basically leaving the USA Airforce General without any chance to win. If you do the latter, the other player will either leave or change his faction to something else. If he chooses to do the latter, he is no longer USA Airforce General and you can change your faction back to GLA Demolition General. Once you do this, the other player changes his faction back to USA Airforce General. Anyway, you probably get the idea. This is how battles were fought in this game: in the online rooms before the actual gameplay even started. Random matches were often ended by one player immediately leaving the game at the very start after finding out that he didn’t have the right matchup and online tournaments were fought between people mostly playing China Nuke General or perhaps GLA Toxin General.

Anyway, the only way the developers could fix ‘Zero Hour’ was by either dramatically toning down the asymmetry between the factions, for example to a level comparable to the first release of Age of Empires 2 or by reducing the number of factions back to a manageable level. Of course, both of these radical interventions would have been impossible to sell to the public. The game never got fixed, some people went back to the original game consisting of only three factions (including myself), but most people fully disappeared from the online multiplayer. The online forum of the game became a platform for constant verbal attacks against the company and the developers and at some point the company decided to fully remove the forum.

Would they have made a different choice at the very start, for example by following the same path as of one of the best mods that were created much later (Rise of the Reds), which basically went back to the base game and added only 2 factions to it (Russia and the European Union) while expanding the three existing ones as well, leading to only 5 different factions with 10 different matchups, they would have created a more workable situation and the online multiplayer might have still been alive today.

I suspect that Age of Empires 4 might already be in trouble balance-wise and that it will take more time to fix than most people realise

Right now, AoE4 has 8 different factions, leading to 28 different matchups (excluding mirrored). AoE2: Age of Kings launched with 13 different factions (78 matchups, excl. mirrored), but AoE2 only had an extremely small difference between the factions.
AoE4’s factions are less asymmetrical than the three original factions in Command and Conquer Generals (and they should be!), but much more asymmetrical than the 13 factions of AoE2: Age of Kings.
AoE4 has an interesting situation where factions often have very different core mechanics, but still have access to the same basic units; while in addition they also have quite some unique units and upgrades. I think that 8 civilisations with this current level of asymmetry might actually already be a bit too much. 28 matchups really is a lot!
If the Mongol mangudai would be too fast to handle for the Delhi Sultanate, the developers can’t just decrease the movement speed of the mangudai, because they would have to take into consideration how this would affect the other 6 (!) matchups of the Mongols. Perhaps the Mongols wouldn’t stand a chance against the French and the Abbasids if the movement speed of their mangudai were to be decreased.
My prediction is that Age of Empires 4 from the very start will experience moderate to severe balancing issues and that it will actually take more time to fix than most people hope and expect.

Another problem: Adding extra factions while maintaining balance might be impossible

Another issue that I think people are missing are the consequences of adding a single new faction to the game given the current level of asymmetry. To give a quick overview of what only a small increase in the amount of factions would look like:

8 factions → 28 matchups (mirrors excluded)
9 factions → 36 matchups
10 factions → 45 matchups
11 factions → 55 matchups

This means that adding only three new civilizations to AoE4 would already double the amount of matchups in a game that has a quite high degree of faction asymmetry. This harsh mathematical reality might soon clash with what people are expecting from the developers!

AoE2:DE normalized a very large amount of civilizations. What people are expecting now is 20 to 30 civilisations. A reflection of this can be seen in the large amount of posts on this forum trying to predict which civilisations are going to be added next. These are topics that I absolutely love to read, as I am too a big fan of history and of ancient cultures, and the idea of having the Japanese, the Turks, the Spanish and the Italians added to this game also brings to me a feeling of enthusiasm. However, I also think that the developers should be realistic and honest. ‘Fan service’ in this area is something that could quickly turn AoE4 into another Command and Conquer Generals: Zero Hour. Adding three generals per main faction to the game was also amazing at first and there are even fan made mods adding even more generals to the game, all of which are very “fun”; but initial enthusiasm cannot change the hard mathematical reality of a game that is simply too complex to balance. Sadly.

Developers wanting to be ambitious and wanting to make fans happy, although completely understandable, could quickly lead to a product that the developers can no longer maintain. Not because they are “lazy” or “incapable” (you can almost already smell the forum posts coming in…), but because they have created a situation that is mathematically impossible to correct. In this case, the developers would not be “lazy”, but instead lack carefulness.

My humble advice to the developers of AoE4 would be to not fall into this trap. It is better to be criticized for not adding an 11th civilisation to the game than for not being able to balance a game consisting of 11 civilisations. Perhaps it would help to be upfront in advance and to moderate expectations a little bit in the area of the amount of civilisations.

A suggestion to increase player options without breaking the game (from Emperor Battle for Dune)

But to end on a happy note, I do have a suggestion which I believe could really work very well and can also be implemented in a historically accurate way that fits the game.

A way to add content to the game without destroying the balance could be the following.
The game Emperor Battle for Dune (yes, I am thát old) allowed the player to select a main faction or ‘house’ (out of three main ‘houses’) and in addition to that two ‘sub houses’. These ‘sub houses’ simply gave the player access to one extra type of building that could create infantry or vehicles from that specific sub house.
The main houses were: Adreides, Harkonnen and Ordos (each of which completely different from the others)
Sub houses were: Fremen, Sardaukar, Tailexu, Ix, Guild.

For example, you could pick Adreides with Fremen and Tailexu, and your opponent could pick Ordos with Fremen and Ix. Adreides and Ordos were completely different from one another, but in this case both would have access to the exact same Fremen units. Adreides would also have access to the Tailexu insect units and Ordos would have access to the Ix cloaking devices.
The subhouses gave the player access to certain units. This was a way for the game to slightly increase the amount of player options without breaking the game, as these little extras were only very minor. This is something that could also be added to AoE4 in order to include more factions into the game without destroying the balance between the main 8 to 10 civilisations.

There exist a lot of factions in history that weren’t really civilisations by itself, but which were important factors during certain time periods or were simply ‘there’. For example, think about the Teutonic Order or about the Saracens. (The Saracens weren’t really a civilisation, but instead a distinct group of desert dwellers).

But then of course, it would make no historical sense for the Chinese to have access to the ‘sub house’ the Teutonic Order. This can be solved by creating three options per civilisation that make historical sense and then allowing the player to pick one of these three. For example, the Holy Roman Empire could have the option to choose between the Teutonic Order, some Polish kingdom (giving access to winged hussars) and some family from Northern Italy that has a good gunpowder unit. The Abassid Dynasty could have access to Saracen Nomads, some extra Persian siege units (since the House of Wisdom is already more or less based on Persian intellectual heritage) or some group in North Africa that has strong infantry and some transport ship (Moors or something).
There are really many options for every other civilisation in the game, because history is filled with nations, families, orders, tribes, etc.

Anyway, this is just a suggestion.

Thanks for having the patience to read through this topic and I hope the game will turn out to be great!


I’m not sure I’m convinced that balance needs to be done based on matchups, in theory it should be possible to set a standard power level and then bring each civ up to this power level.

It also might be easier to balance aoe4 where asymmetry is mainly economic and militarily its usually minor bonuses + accessibility that is tweaked rather than huge changes to the strengths and weaknesses of a unit. For example longbowmen are stronger archers but they are still weak to fast moving anti-archer units so as long as every civ gets a viable fast moving anti archer unit in the feudal age they are balanced. I’m aware this is a huge oversimplification of balance but the general idea should be viable.

To use your example every civ has access to archers which typically counter horse archers. As long as the mangudai archer matchup is balanced, dehli are able to deal with mangudai.

edit: to be more specific, I think it’s really important to look at very low level balance to get this to work. How does civ-A handle heavy melee infantry in the dark age, feudal age, castle age etc. The goal is to make sure that each civ has enough tools so they aren’t stuck fast castling into a unit that is hard countered by the other civs preferred unit comp.


I totally get the argument however I think AoE 4 will manage fine.

Each civ gets access to the same core set of units which I think will help a lot.

Economies have different mechanics, but they can tweak the balance of those so everyone’s economy is at roughly the same level.

It will just be a trial and error process, small incremental tweaks to try and get each match up close to 50%.


They could have make units different. Like, lancer is the exact same units as knigh.

1 Like

Ofc it should be :smiley: In rock paper scissors every item has 50% winrate but every item has 0,50,100 percent winrate vs any item and this is determinant and unchangable. For rock paper scissors, there is a level of interactivity that is- what item to choose rapidly in the moment it counts. The game is super short and its more about the moment of surprise than actual esport skills. But in a game of aoe that goes 30 minutes you know the opponent civ from the beginning and all these 30 minutes you know already who wins. The whole game and all its decisions would be irrelevant. The lack of civ wins is ofc a very important parameter to judge balance and hence the number of balanced matchups possible is also.

1 Like

I 100% agree with your statement and I couldn’t have said it better.
Age of empires 3 is a clear example of what you just mention where even though most civs are “balanced” in a way where their overall strength is rather equal, there are civs that can pretty much hard counter other civs in certain match ups and they can’t be changed it otherwise it ruins several others, and the game is getting less and less fun to play as most ppl now can climb the ladder easily by abusing certain mechanics of certain civs.
I do think the devs approach to this game has been very good, having a good degree of asymmetry while maintaining the core gameplay mechanics for each civ and with each civ having access to all unit types. And I think that approach can definitely leave room for more civs in the future.
The game at earliers stages already felt way more balanced than aoe3, and at least for now I don’t have any major concerns when it comes to balance on this game as it seems to be going in a good direction.


Aoe3 kinda uses that approach but it doesnt work well at all.
Most civs have a similar power level in general but there is a huge amount of one sided match ups on that game and it’s getting even worse now as they keep adding more civs

1 Like

I agree that is a danger with the approach I outlined but I don’t believe it is a garuanteed outcome.

Could you elaborate on what the specific balance issues are? Is it like Civ A is balanced around strong cavalry and civ B is balanced around strong infantry which innately counters cavalry?

1 Like

Honestly, the add-ons in aoe3 ■■■■■■ the balance up.
Nilla was nicely balanced imo.
In 1v1 the balance is still ok I’d say, but in team matches it’s just chaos. Way too many civs which are all asymmetrical.


I think the current approach from the devs is quite good.

There are so many that any player could write a whole book about it, but I can name a few, for example let’s put spain vs dutch for example where spain has a very strong ff timing with an army composition that directly counters dutch ones and is really hard for dutch to stop it as it hits them when they are at their weakest, making the match up heavily favour the spanish, however you can’t simply nerf that ff because it will make them unplayable in other match ups(legacy spain was unplayable after Asian dynasties expansion) while on the other hand you can’t simply buff dutch for that same reason as it gets rid of a weakness that a lot of civs already have a hard time exploiting.
Brits for example do really well vs spain and vs dutch because of their very strong and faster boom, but if you nerf them it will make them unplayable vs civs like french, germany or india since they already have the means to contain them.
Also most civs will lean into very specific army compositions and tend to do better on different stages of the game due to the way they are designed, causing some match ups to force one civ to kill the other early otherwise they just loose because the other will have an eco that goes out of control(french vs japanese and spanish vs British are clear examples of this).
Them there is the most controversial aspect of the game which is houses that gathering ressources(shrines/torpes) as its pretty hard to balance as it either makes the civ too strong and dominate most match ups or too weak.

1 Like

That seems to be a nice idea. Yet, wouldn’t your last example do the same as C&C Zero Hour?
Is it just, that it would still be far more symmetrical then former?

However, I do not believe, that the civs need to be balanced with every other civ. To be more precise, I believe, each civ to have its one unique counter civ, would still allow for mostly even matchups.

If each faction had only one unique «counter» civ, the 8 civs could be arranged like a circle in which for every chosen civ holds true:
The civ in front has an advantage and the civ behind has a disadvantage.

If you want a perfectly balanced match up with lets say 8 players, then you chould choose from…
(All players had to chooce from the same set)

• two four-party civ sets like
A: Holy Roman Empire, English, Mongols and Chinese
B: Abbasid Dynasty, Delhi Sultanate, French and Rus

• sixteen three-party civ sets like
A: Holy Roman Empire, English, Mongols
B: Holy Roman Empire, English, Chinese

E: Holy Roman Empire, English, Delhi Sultanate

P: French, Delhi Sultanate, Holy Roman Empire

• twenty two-party civ sets

Additionally there are cases like, if you play 4vs4 and one team would choose the entire A or B four-party set,
as long as the other team chooses four different civs (from all 8) as well,
it wouldn’t actually matter too much what they choose.
For every counter to the enemy team,
there would be a counter in the enemy team.

If HRE is good against Abbasid Dynasty but weak against Rus,
and French, Delhi Sultanate and HRE are neutral to each other,
same could go for
(French, Delhi Sultanate, HRE) vs (Abbasid Dynasty, Rus, French)
(French, French, Delhi Sultanate, HRE) vs (Abbasid Dynasty, Rus, French, Delhi Sultanate).

Basically, there were only certain matchups, you really want to avoid. As competitive gamer that is.

1 Like

What the point of not making all mu viable?


For competitive gamers? None.

The question is, if it will be possible for more civs to come.
I wanted to make an example to show, that (theoreticly) the civs don’t need to be perfectly balanced to each other, to still allow more than enough fair match ups.

If it only was like Mongols are good against Delhi Sultanate it would be even less of a problem, even though you wouldn’t get a fair 8-player game if everyone chooses a different civ.

1 Like

Any skill players can win any game, whit any civilisation, even if his civilisation is at disavantage, because of asymetry.

More asymetrique civilisation only give more thinking and strategy making.

People do like to learn and overcome a hard figth, changing the way to play whit the same civilisation and knowing more about units on the batlefield.

Yes, the education video look cool, but what is the point to have them, if it do not affect units mecanic in game and don’t let players learn from them.

For the time being, i feal the 8 civilisation alot symetric, except for the mongols.


If anyone here remembers it, I wanna bring in some typical autowin combo from aoe3 TAD.

Ottoman + Sioux (now changed to Lakota).
In that combination all you had to produce was Otto abus guns and Sioux bow riders and you could fight 2v3 in a 3v3 match and WIN the fights while your 3rd mate was booming.
These 2 units together were super broken.
The thing is, each civ was alright in a 1v1 MU because they had weakness that made up for the unique strengths of these special units.
In a team game, you could just exploit this by covering the weakness by each other and let the OP part shine.

Same goes for Inca+Otto in 3DE, Inca pike + abus age2 timed push. There was basically nothing that could stop this in 2v2 because both units alone were broken in a way but the civ design kept their strengths under control in a 1v1 MU.

These kind of things happen when there are super asymmetrical civs with highly unique units that have an out of the box op unique strength.
Like abus being light inf that hard counters not only heavy inf like it’s supposed to be but also ignores all type of armor and hence shreds OTHER light inf at range brutally, which usually is evenly matched against other light inf and has a high ranged resist together with low hp.
The only thing that can take the abus down is melee cav.
Here the Lakota bow riders come into play, which are the strongest ranged anti cav unit the world has ever seen with the most ridiculous attackspeed + damage Vs cav, super fast movement speed and insane raid (dmg Vs villagers) potential.
Not to mention the abus can one shot snipe falconets which are the only way to hardcounter light inf on range. Also the bow riders shred falcs.

Please Relic whatever you do, however asymmetrical the civs will be, don’t do the mistake and introduce a unit like abus.
With such a unit in game, you can basically forget about competitive and even just casual but skilled team games.
Same goes for Inca pike (pre patch):
Huge (cost effectiveness-wise) overproportional anti cav damage, siege damage, hp and movement speed.
Yes, they get hardcountered by cannons and light inf, but in team games you can easily cover this fact because they outperform (-ed) other civ’s generic pikes in their role by far.
Leave some abus behind them and nothing can come close to this deathball while they take down your buildings, walls, towers and gates like butter.
My elo back then when I had the pleasure to play versus these combos in 3DE was around 1800+, so not exactly super Low.
If you end up against that in 2v2 quicksearch multiple times in a row against the same lame players, you just wanna shut down the pc and go outside to enjoy the weather and wait until next day to play this game without meeting these guys again.


Abus are not that good in 3DE ^^

1 Like

Probably not, right.
Though in certain combinations like mixed with Inca likes, they are insane. The Inca pikes got a mild siege damage nerf around the time my mate (we played lots of 2v2 ranked qs) and me got lamed and probably the problem was lowered a little after that.

Also, in Nilla abus were hardcountered by light cav.
I remember playing Dutch Vs Otto and once reaching fortress age, you could at least massacre abus with ruyts, especially in melee mode.
Sadly abus are considered real light inf in aoe3DE and light cav should be running away scared from them.
Plus they still do siege type damage, which makes them pulverize both heavy (due to multiplier) and light inf (due to ignoring ranged resist), which naturally has low hp and low base damage.


Agree. Also please no houses that gather ressources like shrines, torpes and even kamcha houses as these are hard to balance and will just lead to a kill them early or loose type of game.


Very very very big agree.
Forgot to mention this point.


Abus gus sux vs light cav in aoe3DE, that’s why german hard counter them, just because of WW. I left aoe3DE in Mars 2021 so maybe otto receive some buff but it was not a great civ both 1v1 and team at this time (i was top 20 by the way so i am talking about hight level game play).

And if i rermember correctly, abus gun was absolutly nuts on vanilla, they litteraly countered heavy cav when massed enough ^^

1 Like