DE 1v1 Winrates

Age of Empires II: DE II - Discussion
62

Ok, I’m sorry but everything you just say it’s incorrect here.
I’m not saying this to offend you, to be clear, but you can’t just say that those percentages are the probability of winning, nor that 53.4% is about 40% just because they seem similare, 13.4% of difference could mean that they are two different numbers as much that they are the same, this isn’t how statistic works, you have to justify trough calculations your hypotesis.

On another discussion which I link below, I already explain why we can’t trust those number too much, those are rough number, you have to analyze them through statistical methods.

We have to be careful on how we interpret those numbers, because they can bring us to incorrect conclusions.

63

There must be something wrong with the Portuguese and Koreans.
Most civs have moderately similar win rates, distributed in a spectrum, but there is a significant gap between the last two and the rest.
The top 3 also seem to have a significant advantage in relation to the rest.

1 Like
64

Portuguese are, by far, the worst civ in the game. they lage behind immensely. Koreans are very easy to counter, and their towers are not abused as much in higher ELOs, because Tower Rushes are proportionally much easier to deal with, the better a player you are.

1 Like
65

How is portuguese the worst civ?

66

There is. You cannot trust those statistics yet.

67

0.87% pick rate and 38.3% winrate in All Elos.
0.52% pick rate and and 38.3% win rate in 1600+.

They have the lowest winrate across all levels of play.

1 Like
68

That’s not a reason for me.

Ok, let me do the calculation on portuguese,

The chance of the winrate being accurate: 9,02%,

As I said, not enough for me.

69

It is for me. Specially from a balance perspective.

If you want the reasons for these results, then you would have to strudy the civs interactions between it’s bonii and Tech Tree, and then it’s interactions with all other civs.

1 Like
70

I agree that portuguese deserve some big buffs, but definitely not the worst civ, and definitely not by far, like you said.

71

the portuguese had very similiar results in aoestats pre de which had huge amount of sample data so the change of an statistical error in these findings seems to be low, since hte findings in this new data just from april coresponds to the prevoius findings.

many civs have received buffs, especially all other bad win rate civs in aoestats data, this allone gives a plausivle explanation while the portugese tanked even a bit further in their winrate (+feitoria buff is mostly useless and organ guns are bugged in this patch)

But you wanted some points why the civs is the worst so i share my opinion.

portuguese have a total of 1 bonus on land maps 1v1 and 15% gold discount is much better on pper than in an actual game, since techs are a major part of spending and gold doesnt matter much until castle age. Also this 15% bonus just allows you to train portuguese all around medicre units, which just doesnt cut it. THe bonus is too unfocused to be of any real use.

Examples:
if berbes go cav, they have a much better bonus,
if mayans go archer, they have a much better bonus

their UU is currently bugged since the lat patch further adding to their bad performance but generallz organ guns just dont perform very well outside of early castle snowball (strong and no upgrades but just too expensive later on and esaily countered by mass trash hussar, mass trash skirms or some siege rams or onagers.)

the civ is on the slow side like teutons and their cav is medioce, their archers are average and their infatrz lacks the speed tech.

yay they have good hand cannons but how many time are these units even used, they are niche, same for bombard cannons, good to snipe enemy siege, but they are also the most expensive unit in the game, and if you go mass gunpowder and lose them, which can happen really fast against onagers or cav, or even archers, it is devastating.

Feitoria is more or less useless.

Ah and lastly on water maps we have a caravel unit that is in any 1v1 setting USELESS since galleons are just better. 60v60 ship battles do not happen and even then the carvel is not even broken as has been show a byz FU navy of 40 galleon 10 fast fire 10 demos beats any portuguese 60 pop FU navy compostion.

3 Likes
72

i got a bit curious. which civ do you consinder worse then?

73

I agree most of what you said, but I think the feitoria has become actually useful after the february buff (not just in no wood, no gold and no stone situations). Building a feitoria is much faster than training 20 villgers, so if you manage to do a fast imperial, you can do a really, really strong boom which non of the other civilizations can do. However, this is very hard to execute, and it is much less useful in the long-run (since 120 villagers are much better than 20 villagers and 5 feitoria, speaking economically), but definitely strong in the short-run. If you manage to defend until you build 5 feitorias like really fast, and then start attacking, there is a very huge potential of an unstoppable snowball. Note that this tactic is only good in maps were you can start attacking fast, and you’re also well defended, so basicly only good in arena and maybe hideout regarding the ranked pool maps. This strategy is very hard to execute, because how much it differs from other strategies, but can be highly rewarded if done properly. If I remember well, Viper made some experiences in arena and in most cases he won. I still think that portuguese is in the top5 worst civs, but I’m very happy that feitoria has actually become useful (not just in the very late game).

Anyways, the answer for the question you asked (which civs are worse in my opinion): Teutons for example. They don’t have any units going on for them, whenever I went teutons I never know what should my tactic be, what should I actually do. Their UU is bad, no raiding unit (since no access to light cav), paladins are too much slow, no bracer means archers are not really an option, so basicly the only thing you can do well is the halbadier siege onager push. But this one is very gold dependent, if you can’t secure golds, this plan wouldn’t work. However, even if you manage to get the gold, your push will be really slow, and if the enemy start continuous raidings, at some point you’ll get so much distracted (because your units move so slow) that you cannot deal with everything, if you focus on your push, your economy falls, if you focus on defending, they’ll be able to hold and push back your attack (siege onagers require lots of micro). All in all, teutons really deserve a buff for their UU, and should definitely get light cavalry atleast. Another example is the goths, whenever I play them I fell like I just cannot defend myself from enemy attacks. It might be just me, but it feels nearly impossible to advance to the imperial age if your enemy is pushing continuously with knights or crossbows.

74

11 at “top5 worst civs”. i like that.

i assume paladins and cheaper farms alone is responsible for teutons win rate above the one of portugal.

1 Like
75

Reddit post from one of the authors of that site

2 Likes
76

Hey all, I’m one of the guys behind AoE-TC. First of all, thanks for noticing and discussing the site! It’s awesome how many discussion had revolved around it.

That being said, we are working on implementing a lot of the feedback posted both here and on Reddit. Better and more filters (map/team stats?), and match-up charts included. There’s room for improvement, and we want to make sure the data is as accurate and meaningful as possible.

Of course, if you have any questions, please ask them here/in person, I’d love to clarify any questions/doubts you’ve been having.

7 Likes
77

Oh, that’s nice. Do you think something as detailed as aoestats is possible for DE?

3 Likes
78

As of right now, no, I don’t think so. Some of the stats shown by aoestats simply aren’t available by the method we’re using (aoe2.net).

Some things are obviously coming, like team stats, map filters and improved elo filters. These are all definitely getting implemented. For other stats, like “Resource X versus game length”, it’s more difficult. There’s no way of determining those stats using our current methods, and we’re not really sure what data is relevant enough to show and what data just clutters up the overview.

We would like to get into contact with other sites and people, though. See what they would like, and what the possibilities are regarding everybodies wishes.

7 Likes
79

Sounds great!

1 Like
80

Addition stats are always nice to have but if these 3 filters will be implement the most important stuff would be there, in my view. Thanks a lot, great work!

2 Likes
81

If you could add in some statistical p level confidence intervals, would be great to have for discussion if the sample size is large enough

1 Like