It’s ok, but they also miss important techs like siege engineers bombard cannons, their monastery is terrible (monk defense is very important on closed maps) and their additional TC doesnt help them in the late eco. Almost every other good civ in closed maps has a way better lategame eco than cumans.
The lack of defensive technologies on makes cumans also quite vulnerable in closed maps, too. You will see on many closed maps cumans are actually still one of the weaker civs and the highest winrates you can achieve with them right now is in the midgame. So they are indeed good in hideout, but not so good in arena or fortress because of these other factors.
It’s never that one-sided as you want to present it here. Cumans lategame isn’t that strong because of many factors. Having aggression potential isn’t everything in the lategame. You must also be able to defent your eco, too. And cumans suck in this regard. All in all, their lategame is below average, at least in closed maps.
Their siege is below average. Infantry has no specific bonus. The high mobility units would be useful in open maps, but there the lack of early eco hinders them to be proactive. Whilst their midgame is indeed insanely strong in closed maps, their lategame is actually quite bad, the only units which are featured are actually no good closed map units.
Of course cumans could be buffed here, but I don’t think it’s necessary. I’d like to have them balanced in most maps rather than op in closed and trash in arabia.
You mean like a siege ram with FU infantry support?
Also i never said anything about closed or open maps. There are a number of civs with suboptimal late game armies for closed maps.
Arabia /open maps are the dominantly played maps. Cumans have a very good selection for the late game on the most dominantly played maps.
Its like saying Bulgarian late game sucks and then referring to islands as proof?
The point being cumans need a leg up. If their early game is buffed like it should be, their open map strength will stand a better chance. But the point is also, they do not need any late game buffs wrt their military… Because it is already good for the MOST DOMINANTLY PLAYED MAPS
Well, I share this information for everybody and wanted to clarify the strenghts and weaknesses of the civ. Because many players just seem to not understand how they work actually.
And I try to read many posts and not only the one from one player. If you haven’t mentioned maps, it’s ok. But there were players mentioning it, so I felt free to go deeper into it. Because it is actually very important for the understanding of the civ right now.
The extra TC right now is not working in open maps because they just can’t afford it without being punished. In closed maps, their lategame is just not working, they miss too many key techs.
So the problem with an early eco boost is very appealing. Once they just get it but their booming potential isn’t reduced there is a high risk they could afford booming in open maps, too (into cav spam in arabia11). And their boom in closed maps could even be more abnoxious than it is right now because they could age up earlier. Besides their lategame isn’t good for closed maps there isn’t much holding them back in the midgame, so it’s high risk in closed maps there will never be a lategame with them then.
So I warn for any attempt to give cumans any early eco bonus right now, it could completely overpower them.
And of course all civs should be balanced for most map types. There can’t be a civ dominating one map with winrates of like 75 %. This would be terrible for the game.
And to add this, I still believe that the low winrate of cumans in arabia is patially because of the misconception. There is no reasoning for cumans to be worse in the early game than for example spanish, portuguese, koreans or sicilians. And Cumans have very good options for arabia in the midgame. The insanely low winrate is easy to explain with some players trying to make the feudal boom with them which is just a bad idea in arabia.
Not that I want to necro this thread, but are there any stat pages for the new patch available?
I would like to have a deeper look into the performance of the new civs, but also the influence of the new arabia on the win rates of all the other civs.
I can provide some insight on civilization performance from the current patch which hopefully can give some food for constructive discussions/civ talk.
RM - All Maps - All Teamsizes - All Ratings
Top 5 civs in descending order (all within 0.5% of 52% winrate):
Franks
Indians
Berbers
Huns
Goths
Bottom 5 civs in ascending order (all within 0.6% of 47% winrate):
Chinese
Byzantines
Vietnamese
Burmese
Saracens
New civilizations from the Dawn of the Dukes expansion:
Bohemians - 48.53% (additional info: top 5 civ in Arena 1v1 specifically, lower than average winrate on Arabia)
Poles - 50.47% (Poles also perform very well on Arena with an average performance on Arabia)
Of course it can also be important to know filtered information related to specific maps/ratings/teamsizes/… but to give a small insight until the community stats website(s) update to the current patch.
Looks like the arabia change hasn’t influenced winrates as much… I’m a bit surprised by this tbh. I thought it would buff the already strong civs.
Would it be possible to get stats based on elo? Cause I think in high elo the changes have much bigger impact.
Seems like the new civs aren’t outperforming yet. But I would also give them time. They are both very unique and need time for sure to be figured out completely. But atm it looks like they are neither too OP or trash, so seems initially devs found the right balance for their implementation.
And thumbs up for showing up, providing and sharing your inside info with us! Really appreciate this!
In 1600+ RM 1v1s there are 3 civs which remain in the top 5 while two are being swapped out. The deviance in winrates from 50% is also slightly higher compared to taking all ratings into account.
Berbers (out) and Teutons (in) had a small switch with both having decent but not top winrates while Goths (out) fall off quite a bit and instead Mayans (in) swoop in and take the first rank, dethroning Franks.
I think that overall all civs being within the 47% - 52% winrate clearly shows how good the balancing has become.
My thoughts:
The Top 5 civs all have a relatively clear gameplan. Most people who “random” into them will know what to do and how to play them (even if they usually don’t play those civs).
The bottom 5 civs (except Vietnamese) are more gimmicky and difficult to play if you are not used to them. The Chinese with their weird start, Saracens with their market shenanigans, Burmese with their non-forward techtree / boni and Byzentines with their defensive / counter gameplay strength…
So, to me it makes complete sense why those Civs are top tier / bottom tier even though I believe they are all pretty well balanced.
The only outlier might be Vietnamese which seem to have a clear archer based gameplan but still fall short…
The problem of their archer plan is that is not as acute as other archer civs.
The extra HP barely helps in feudal (although it is very good against mangonels) and the wood discount for eco techs helps a lot but there are better eco bonuses. Britons can reach feudal faster, mayans eco is top tier with discounted archers, ethiopiams archers kill faster and the free food and gold is good for fletching, etc.
Vietnamese archer plan is better than italians, though. But italians are known for their flexibility and lack of early bonuses. Actually, vietnamese shine more in arena where their “lack” of feudal military bonus is not important and where their eco bonus helps to boom.
Well, even the korean bonuses make their atchers more useful in feudal than vietnamese ones! And they have the ability to transition to WW easily.
With burmese and vietnamese in bottom tier, I would like to think that it is time to buff their elephants (specially whem those winrates include teamgames and 1v1 all together). But they need something for feudal, or a stronger buff for later stages, or some kind of help for transitions.
makes sense for the ladder at large (china likely swaps out with higher elo)
same, they can still eat some builds simply due to the eagle, but in general struggling
using that age up discount is also a bit of a trick, and combined with their cheaper ballistics in castle, their “early” game can be pretty good simply by surviving into castle age
yeah the buff to OG helps, but i thin the stats are also deceptive… remember china shouldnt be close to bottom 5
If Vietnamese need a buff. How about archers are 5/8/10% faster in feudal, Castle imp, additionally to the hp bonus which isn’t that imesne since archers low hp anyway and something like Britons range or Ethiopians attackspeeds is much more valuable?
Burmese are a very weird civ, my least favourite to play too. How abiit change manipur cavalry that gives bonus damage against buildings for cav. Reduce the bonus dmasge by 1 and replace it by +1 regular damage.
Regarding weak byzantine a. How about a new civ bonus that is, allows them to research supplies twice which fits their imperial well managed system and early roman legions. 30f 20g champs without blast furnace should be about as balanced as goths and as an interesting twist to imperial age byzantines. You can reusde your code for going above and beyond maybe
Sarracen could get as a buff, be allowed to train camel riders is feudal age in a weakened state, that upgrades stats like serjeants do when reaching castle age. Or their UT for monks could get extended to mamelukes to return gold.
Saracens is consistently bottom 5 for over an year. People don’t want a buff for them as they have very regular tournament picks especially in RBW4. As I want more civ become viable in water maps I’d personally replace Market cost -100 wood with start with +100 wood.
Just to double check are these naive win rates i.e. number of wins / number of games played not accounting for difference in elo / any other factors ?
I just posted an Arabia 1v1 >1200 elo update on reddit but am struggling to decide how best to analyses the team games, was wondering if you guys had a specific approach you used internally?
I was trying to avoid using a naive calculation as I think that leads to an underestimation of the confidence intervals and in addition doesn’t allow for accounting differences in elo.
Assuming a team game with civs A and B on team I and civs X and Y on team J, I was thinking of fitting a logistic regression model like: logit(WR) = mean( C_ax, C_az, C_bx, C_by) + mean( elo_p1, elo_p2) - mean(elo_p3, elo_p4)
where C_lk is the coefficient for civ L beating civ K.
but yer I wasn’t sure how robust this was so was wondering what you guys do (if you can share at all)?
