Samples why WINRATE only shouldn't be rely on determine the balance

Because almost every balance topic, some people will mention the winrate is fair for that civ and shouldn’t be adjusted.
So I am going to explain once more in this topic, better than every post.

  1. For A civ:
    1 Pro win 10 lose 0, 9 newbies win 0 lose 10 (for easier calculation)
    Winrate is 10%
    Why I use newbie for example, is because I met many guys even don’t know the normal procedure to play supremacy, (build mill at the beginning, 8 min still age1, their vils keep idling until I raid them all, keep training strelets to face my elephants, etc), even though this is rather extreme case. Moreover when a newbie comes to the game, they are always suggested to play the most powerful civ.

  2. When A civ has more and more people to use, its winrate will close to 50%.
    Because in this case, the chance for A civ to fight A city becomes very large, and when this happens, the winrate is 50% (1 win and 1 lose > 10 win and 10 lose, same). This can explain why some civs has many people complaint, their winrate are still looked fair.

Even winrate is going to be referred, at least each level (Pro/Med/Noob/Newbie) should show the use percentage and their winrate, respectively.

If some people want to talk about other balance issues, the below post has already shown many many detail figures and explained why. (That’s not me)

You can use winrate and pr gap which is what we have done on EP.

1 Like

If you want rating to be used when making any decisions in this game you should prolly argue for replacing the way these trash ratings are calculated first.

Using wr straight to change balance without regard for the data origin is insane, like, nerfing lakota fast ageup is absurd.

I can only guess that its a strategy with high winrate in the lower tiers, because at higher level almost noone used the fast ageup, and absolutely noone ever said it was too strong.

Maybe its just a change with the intention to make newer players feel more welcome? I don’t know, but the patch notes imply that they really think sioux fast ageup was too good.


Using only winrate not right, but winrate is still the most important and representative indicator. Yeah, having winrate close to 50% doesn’t mean that everything is alright, but having winrate higher or lower definitely means that something wrong. And if you offer to nerf civ with a ~50% winrate you should buff something instead to save it.

About your “samples”… The most impactful group of players (not pros but playing a lot of games) have a ~50% winrate anyway, and people usually talking about that group just because they are playing on that level. So your example with pros and newbies is a bit irrelevant. About mirrors, just don’t include them in winrate calculation.


That’s why I mentioned 4 groups, and if one civ that almost not appear/always appear in pro group, means that civ also has problem.

From what I know, Lakota is not the only civ has fast age up for age2. Its siege not that good, Haud also has 4 cav.
Warchief only has melee and when he is facing 2 tigers/lions, he lost most of HP.
For me is also confused for nerfing these 2 points.

Btw this topic is only to show people always mention the winrate (especially summarized winrate), for me that can’t be relied on since original AOE 1.07 version.
(At that time 7 of 10 top rankers were using German and its winrate was only 51~53% what I remember was this range).

1 Like

Looks like is the same thing as AOE2 DE

I wish they would give us access to a this data…

However just bc they said win rate doesn’t mean they just used win rate without second and third level analytics, like the pr gap mentioned or game duration or civ match up and much much much more. Specifically with the Lakota they specifically said the win rate with players using this age up strategy was without a doubt significantly higher across the board. It doesn’t matter whether or not this is across the lower mid or high board of players. Especially if it requires high tier players to defend/defeat this LAME strat.


I do think the nerfs to both lakota and India were needed though
The problem is that these 4 early cav from lakota at 4min mark gets followed by a bunch of bow riders later on and things will just escalate from that.
I am aware that pros know ways to deal with that but the average player doesn’t have all the mechanics and skill a pro has.
Then we have India which aussie_drongo himself said that civ was too strong and can pretty much do anything.
So on that case nerf was more than welcome.

Its siege is not that good but they have the option to increase it by a lot using the community plaza ceremony

Yes, I just meaned age2.
Also may be you are right this Lakota method is strong in med.

However I am also med in causal game but I could say I rarely met another guy uses Lakota and furthermore uses this method.
I think it would be a good study if any videos can share.

1 Like

For me, Pro use percentage and winrate should be considered the most due to they are experts in game.
If some civs use percentage are low and some civs winrate are high in their range, there must be some problems.

We can think if that civ is good, why no Pro is willing to use them for ranking?
Thats easy to understand.
And if that civ has high winrate, sure is some problem.

Med and noob data should also be considered but should be less, they are also normal playing the game and just their skill not good.

However newbie shouldn’t be considered because they don’t know the game

1 Like

I think the lakota nerf is irrelevent at hight level because the fast age up is not use in age 1

1 Like

The players level repartition is like a pyramid.
If they balance according to the winrate of all players then the more you go to the base of this pyramid (=lower players), the more it count in the winrate result.
So if they focus on this overall winrate then yea in the long term that will ■■■■ up the mid to pro level balance and all the future tournaments, hopefully they are wiser than this.For example Sc2 is doing the opposite and balance mainly from the high level.


I feel the arguments here are irrelevant because the changes in the latest patch was more positive than negative at all levels.

Is the balance overall perfect since the patch? No…but it’s a lot better in all level than prior to the patch.


I agree.

Excellent definition!

I’ve already done a topic asking, but no response from the developers.
I would like them to at least send an image of the statistics each time a new patch version is applied. That would be a job of only about 15 minutes.

I agree.

I also want to say that some requests that I see from medium and low players are more at the level of design than really balanced, and many people don’t even know the difference.
Changes in design will not always change the level of balance for professional players, so in those cases, requests from shorter players can be considered.

And talking about balance, I have the opinion that it should always be treated separately for 1v1 and team games.

I also thought about this, but this seems difficult.

The balance in team game is they can boom much easier than 1v1. That makes the boom civs very powerful.