I heard you all like stats

I was on reddit and this guy built a comprehensive stat chart for aoe2. Its got a lot of good info and in bar chart form. Thought you all might like it and you can see how popular your favorite civ or map is. The number of people who play franks is ridiculous compared to the other civs.

4 Likes

That was a goodjob, the stats shows accurate results.

For example why keep adding civs if we barely use all the other civs, its amazing to see how 5-7 civs have like 80% of all games played.

An useful stat there are the games for +1850 rank 1x1, the civs and their winning rate are accurate to what needs to be balanced and whats not, interesting how weak and terrible are indians in forums but somehow their winning rate is quite high in good levels, demonstrating that the lack of knowledge is dangerous.

image
Mayans are OP

2 Likes

nerf mayas pls so unfair in ranked

1 Like

What settings did you have the mayans in?

Above of 2K, where the civ balance matters more than ever.

2 Likes

Does the chart show how many singe player games happen and what civs players pick for those, though? (A chart showing MP 1v1 games only (if the description is correct) only represents a small portion. Sizeable, but probably small in relation to the overall population of players)

Regardless, I also think we have enough civs for a while. I think the game has room for improvement and change in many respects other than additional civs

I do agree that even if not used often variety is super nice.

1 Like

I have said this already in multiple threads, but many people seems to not like that opinion. Some people even claimed i cant share the opinion of not wanting more civs. Adding more civs only means more civs arent used frequently.

Many balance suggestions are made by noobs who dont really understand the balance. For that reason i try to stay outside most of these discussions, because most of them are pretty stupid.

There might be a small reason that made sense (to me) why people dont like balance changes. That is if the devs remove the flavor of a civ. That is the most difficult part of balancing. Every civs needs something that stands out for that civ, but that point cant be OP.

1 Like

Yeah I guess that’s mostly because civs like Mayans or Franks are so straight forward in how they play out so it actually doesn’t matter if other civs are well balanced or not as long as it takes more time to figure them out.

Not sure if it’s necessarily a lack of knowledge (might be though). At least it seems to me that a lot of players just copy what pros do and Indians hasn’t been a much played civ for quite some time (despite their great performance in the ladder). And it’s really an insane civs for mid elo players because in castle age your stable units counter everything apart from siege+pike and xbow+pike. If you react with these compositions Indians might struggle and pro know that but when I picked Indians against civ pickers for a couple of days I experienced that most of opponents realized way too late how bad their situations is. To me thats a case for not trying to copy 2k+ strats all the time actually can get you better results.

2 Likes

I also suspect that since DE drew quite a lot of players new to the franchise, there are a lot of people who come from games were you “main” a character or a faction, or where the game is played by working toward getting whatever is stronger, and they apply that to AoE2. So Franks end up being played so much because SotL, AoEstats and complaint threads all agree they are the best civ, Mayans are played so much because everyone repeats el dorado eagles are OP and that they are the best archer civ, and so on.

1 Like

That’s not true at all. The probability of a civ being chosen at random in a particular game is (1/37 + 1/36) ~ 5%. Civs which have more than 5% of the total number of ranked matches are picked/preferred more often. There are about 10 civs that are more preferred and these vary at different elo levels. And even for the same elo level if you exclude Arabia, you’ll see a different set of 10 civs that are played more often. So its not the “same” 5-7 civs that have 80% of games.

Lack of knowledge is dangerous indeed. Just look at the number of matches the civ has in comparison to others. It has one of the lowest play rates at higher elo levels. Clearly people who know the game well barely pick this civ.

1 Like

Who said i was strictly talking about 1x1 or tg, in both of those the number of games with those civs are almost 80% of all games played, which is insane, now go and search how many of those games were mayans and franks, you will be surprised that only 2 civs reign aoe2 in terms of usage.

It doesn’t matter if you pick indians or not, the thing is that they win most of their matches in hands of decent and good players, period.

No again, there are 14 civs with 2.7%+ pick rate 1v1s and TGs all elo combined, which means players pick them more often. And again ranked stats are heavily biased towards Arabia and obviously some civs will be the meta for that map.

It does. When a civ is one of the least picked, it means the good players don’t prefer playing that civ. And you can see that highest win rates are against civs which are not the meta for Arabia - Portugese, Burgundians, Malay, Turks, Cumans.

In general, the number of games played by 1850+ as Indians is quite small just around 1400. The number of games played against another civ shows a huge variance (65 vs mayans to 15 vs Burmese). So the win percentage of a civ that’s almost never picked won’t imply that its a strong or a weak civ. Prior to the November update, Khmer had just a 48% win rate at 1650+ levels but it was still quite a strong civ and had 5+% play rate in TGs. So win rate alone doesn’t determine if a civ is strong or weak.

Do you know that small sample size could give you more extreme results? Check the civs you have used in your ranked games. You might have some civ that you used 3 or 4 times with 100% win rate. But what will happen if you use it every game from now on? The win rate will converge to 50%.

I have a 75% winning rate or 80% in the few nicks i have used to play 1x1, so in general i haven’t really faced enemies around my level, basically from 100 games 70 of them i would have won using any civ, cause the skill difference was more important.

I have even won with civs that i highly dislike like turks on 1x1 non arena maps, you can’t really take your own stats if you are not taking stats from the 48-55%winning rate, so i can conclude that non of my own stats can be used to tell something about civ balance cause i haven’t played enough to be placed properly in the ladder.

However if there are lets say 2000 games with players using indians from 1850 to 2k+ then that is a sample big enough to conclude that indians are a good civ witht hat 56% winning rate, cause the same number of games in the same rank level using a civ like vietnam, koreans or sicilians demonstrates a clear behavior losing way more using those civs.

Also in those levels the players tend to play more random civ encounters rather than lower levels who stick with picking the same civs, that is another factor that matters, if indians would be so bad as highly skilled forum warriors claim, then they would reflect a losing behavior but they haven’t so far, so i don’t know why you keep it going.

If everyone goes random, we would see a more even distribution where each civ has close to 2400 game but Mayans, Chinese, Huns have 4k+ games and Indians have less than 1500 games. When there are 36 possible matchups for Indians, 1500 games is quite low. The mean number of games for a matchup is about 40 but the max and min from that link shows some matchups have less than 20 games and some have 50+. That’s huge variance and this data is insufficient to draw any conclusions.

If you think you’re a genius and want to disrespect others by coining terms, go here and take a look at Hera’s Arabia tier list. (80% of the 1850+ games are Arabia). In case you don’t know, Hera is one of the world’s top 3 players and winner of the recently concluded Hidden Cup 4 tournament 1v1 and Two Pools tournament 2v2. Go to 17:25 - 18:22 for what he says about Indians. He places them as a “D” tier civ and explicitly mentions them as one of the bottom 3.

The average win rate for the sample is 53.54%.

Assuming in each game, one civ wins and one civ loses, this is an impossible result. The winrate aggregate among all civs should be 50%, not 53%, so we can safely assume every data point is inflated in winrate significantly.

When people say “there are lies, damned lies, and statistics,” they don’t usually mean “the statistics are clearly, unmistakably false.” Usually they mean someone’s decided to interpret the statistics in a different way and used the statistics to tell an incomplete story. These statistics are damned lies.

The reason is all these results are counted for players above 2k but not for players who played against 2k + players. So, we’re seeing a general inflation of winrates coming from players at < 2k vs players at > 2k where only the player with the higher rating is being counted.

Thus, 3.54% more civs win than civs lose in the sample, therefore every civ gets a winrate spike. odd that you chose to ignore the Indians and the Chinese, both above 59% winrate in the sample to pick on the Mayans, but again, this data is bunk, so it doesn’t matter in the slightest.

2 Likes

This is a 6 months ago tier list and it has Koreans and Turks in A tier 11

The video was uploaded after the November patch and since then Indians are completely unchanged. Burmese, Cumans, Franks, Saracens were nerfed after that and a couple of civs got added. So maybe Indians are no longer a bottom 3 and maybe they’re a bottom 5 civ but the reason why I included that link stays good - Indians are not a great civ as interpreted based on their win rate at 1850+ elo level by @SouMexican

Joke’s on you dude, the tier list content is from a top-3 player in the world and he definitely knows the game better than you. At high level games, the power of free archer armor bonus of Koreans and the extra p.armor on Scouts for Turks matter a lot.