Civ Stats for Autumn 2016 - RO128/64

[Mitoe]

I seem to remember that someone compiled a list of tournament results for civs throughout ESOC's first tournament, the Spring Championship.

I thought it'd be fun to see these stats evolve throughout this tournament as the meta seems to be changing quite a bit on the new patch!



Raw numbers:
[spoiler]Aztec
Picked: 5
Won: 2
Mirrors:

British
Picked: 23
Won: 9

Mirrors: 6

China
Picked: 3
Won: 2
Mirrors:

Dutch
Picked: 17
Won: 6

Mirrors: 2

French
Picked: 20
Won: 6

Mirrors: 5

German
Picked: 7
Won: 6

Mirrors: 1

India
Picked: 12
Won: 9

Mirrors: 1

Iroquois
Picked:
Won:

Mirrors:

Japan
Picked: 10
Won: 4

Mirrors: 1

Ottoman
Picked: 8
Won: 3

Mirrors:

Portuguese
Picked: 15
Won: 10

Mirrors: 1

Russia
Picked: 23
Won: 16

Mirrors: 2

Sioux
Picked: 1
Won:

Mirrors:

Spanish
Picked: 7
Won: 4

Mirrors:[/spoiler]

I think I managed to record all of the games played in the RO128/64 (so far), but there may be a couple matches that haven't been played yet. Mirrors are recorded independently of pick/win rate.

So it would seem that players' top 4 favoured civs for this round were:

1) British
2) French
3) Russia
4) Dutch

Each being played over 15 times each, with a fair number of mirrors in there as well, especially on French and Brit.

Interestingly, despite their play-rates, France, British, and Dutch have incredibly poor winrates ranging between 30-40%!

Russia in contrast, has a staggering winrate of 70% for this round.


Some other interesting things to note:
- Germany's winrate, losing only 1 of 7 games played, despite falling out of favour with many players.
- Otto's playrate seems to be going up compared to previous tournaments. I'm not even sure Otto was played 8 times in the entirety of the Winter tournament.
- China seems to have fallen out of the meta, on these maps, at least. Although they are reasonably successful in the few games they were played.
- RIP Iro/Sioux


Anyway I thought it'd be fun to take a look at this sort of thing every couple of rounds and see how the meta progresses as the stronger players start facing each other.

Important to note that the map pools also play an important role in picking/favouring certain civs. Note the winrate on India/Russia (two pretty controversial civs at the moment), probably due mostly to Cascade Range.

[Garja]

So far it seems it reflects the skill of players (not very high atm) combined with the considered best civs. Russia//germans are relatively intuitive while Brits/Dutch are reactionary civs and thus a bit trickier to use properly for the average player. Put in other words, offensive civs are easier to use at lower levels it seems. I expected somewhat higher winrate for otto but I guess now you really need proper counterbuilds to prove them strong. Should also consider them case by case probably since I imagine many of the otto games being of the "good player vs bad player" type.
Iroquouis and sioux are probably just totally unknown civs for the average player before anything else. It is sort of a popularity contest afterall, just look at Spain, Aztecs and China usage rates for example.

[iNcog]

Yeah Russia was already a good pick before and with recent buffs it's a very, very solid pick, I think.

Shame I can't play EP.

[Mitoe]

[spoiler]So far it seems it reflects the skill of players (not very high atm) combined with the considered best civs. Russia//germans are relatively intuitive while Brits/Dutch are reactionary civs and thus a bit trickier to use properly for the average player. Put in other words, offensive civs are easier to use at lower levels it seems. I expected somewhat higher winrate for otto but I guess now you really need proper counterbuilds to prove them strong. Should also consider them case by case probably since I imagine many of the otto games being of the "good player vs bad player" type.
Iroquouis and sioux are probably just totally unknown civs for the average player before anything else. It is sort of a popularity contest afterall, just look at Spain, Aztecs and China usage rates for example.[/spoiler]

Agree 100%

[momuuu]

The data with regards to winrates is probably insignificant. It is sad that iro and sioux are yet again not played at all in the tournament and not really played in casual games either.

[BrookG]

I believe Russia was always underestimated. Now with their recently eco buff, even with those early calculations they turn to be quite strong. Interesting indeed, the fact that mostly picked civs on the lower PRs, maybe due to the rumor they are strong, tend to number a significant amount of losses. Warchief's civs were always harder to learn. Their rebalance makes the players incapable of laming, and they actually need to play seriously the civ.

[deleted_user0]

well on lower levels russia has always been strong, so has india btw. so im not surprised at these winrates, because as garja said, on lower levels agressive all in builds are usually very effective. hence every pr20-25 will say janrush is op, while every pr40 will say jan rush sucks donkeyballs. because by that pr you will have mastered skills that render such playstyles ineffective. and ofcourse, garja made sure with his maps that any advantage such styles did still have, were completely taken away from them

[lemmings121]

besides the fact pointed by garja that those games where mostly played by us, the mortals sub 30 pr, its also relevant that they were played in a static as small map pool.

apparently everyone likes russia in adirondacks and india on cascade..

[Goodspeed]

Yeah Russia was already a good pick before and with recent buffs it's a very, very solid pick, I think.

Shame I can't play EP.

Why can't you?

[iNcog]

Yeah Russia was already a good pick before and with recent buffs it's a very, very solid pick, I think.

Shame I can't play EP.

Why can't you?

viewtopic.php?f=110&t=8258

viewtopic.php?f=110&t=8272

The EP has resisted every single attempt of mine to get it running, for reasons which make no sense. I'm out of ideas, tried reinstalling, tried deleting old folders, nothing works. RE works perfectly fine, but I have no reason to play RE.

[momuuu]

Bengal, pampas sierras, adirondacks and cascade range is not a poor group of maps for russia at all.

I would say russia wasnt as weak as people thought they were and isnt as strong as people think they are, something that can often be applied to balance. Except when it comes to dutch, dutch is structurally overrated

[Gendarme]

[spoiler]So far it seems it reflects the skill of players (not very high atm) combined with the considered best civs. Russia//germans are relatively intuitive while Brits/Dutch are reactionary civs and thus a bit trickier to use properly for the average player. Put in other words, offensive civs are easier to use at lower levels it seems. I expected somewhat higher winrate for otto but I guess now you really need proper counterbuilds to prove them strong. Should also consider them case by case probably since I imagine many of the otto games being of the "good player vs bad player" type.
Iroquouis and sioux are probably just totally unknown civs for the average player before anything else. It is sort of a popularity contest afterall, just look at Spain, Aztecs and China usage rates for example.[/spoiler]

Agree 100%

Is this a way of suggesting that Garja is always right?

[zoom]

How is the meta changing, do you think?

Also, how is the "mirrors" statistic more independent than the others? As far as I can tell, they are all equally independent of each-other.

[Mr_Bramboy]

Sioux :(

The high amount of Sioux games during the last tournament(s) is most definitely caused by Tibet being in every single round.

[deuxballesman]

High plains (5 TPs) and hudson bay for the awesome treasures helped too

[Mitoe]

How is the meta changing, do you think?

Also, how is the "mirrors" statistic more independent than the others? As far as I can tell, they are all equally independent of each-other.

It's just interesting to see which civs people are playing, because it should be what people think are good.

For example, Germany's playrate is significantly down from last tournaments, despite being (arguably) the most dominant civ in the past. At the same time, Dutch and Otto have significantly higher playrates than before.

I included mirrors because I think mirroring in a sense is like saying they're not comfortable playing anything else against it, so they'd rather take the most fair approach and mirror it.

[deleted_user0]

Bengal, pampas sierras, adirondacks and cascade range is not a poor group of maps for russia at all.

I would say russia wasnt as weak as people thought they were and isnt as strong as people think they are, something that can often be applied to balance. Except when it comes to dutch, dutch is structurally overrated

yet you rekt brits with it today?!

[momuuu]

Bengal, pampas sierras, adirondacks and cascade range is not a poor group of maps for russia at all.

I would say russia wasnt as weak as people thought they were and isnt as strong as people think they are, something that can often be applied to balance. Except when it comes to dutch, dutch is structurally overrated

yet you rekt brits with it today?!

I believe some unknown guy named umeu was rather upset that I picked dutch against brits

[deleted_user0]

well i was, cuz brits reks dutch if it is wp. but you played it quite nice nonetheless.

[momuuu]

Well I played that mu mostly because its one of my more decent mus, dutch one of my better civs while brits one of king's less played civs which seemed like it'd be about as good as I could get.

[Darwin_]

This is really cool, but I dont think we can come to any conclusions until the Tournament is over.

[momuuu]

Honestly I think right now you can kinda make one tier, in which the civs are about as far apart as in some tiers on previous patch), which contains like 9 civs or something all close in skill that all have different counters. Theres maybe 2-3 candidates for top civs but I really dont know what they are. Even some of the possibly weaker civs (india/otto for example) seem to have a very serious spot in the meta too. Iro/sioux is kinda sad and unfortunate though.

I would guess tier one is like all but india, otto, iro and sioux basically, maybe germany too. You can try to split that into two tiers but thats just a sad and painful task because its so close.

Assuming the "Zomg dutch is super ridiculous right now" thingy has died out I think we can agree that balance is really good.

[gibson]

Assuming the "Zomg dutch is super ridiculous right now" thingy has died out I think we can agree that balance is really good.

Don't forget about whole "Russia is op" now, even though the only thing people who say this use to defend it are obscure terms such as "they're smoother" and it "snowballs", although I've yet to see any numbers that demonstrate they're any better then an extra Cossack every 5 minutes..... While the real reason why Russia is better is that 5 civs they struggled with in re have all been nerfed and they weren't that terribly bad to begun with.... At least not re port/Dutch bad

[Darwin_]

As I said before, it is inappropriate for us to make assumptions on civ balance until we get further into the tournament, and get to a round where there are fewer clean sweeps. It us undeniable that Ports, Russia and Dutch are now fullblown OP in teamgames, and have improved a lot in 1v1. What we have yet to see is if these civs are just strong, or if they are actually OP and unbeatable in some matchups with players of even skill level.

I'm gonna make some predictions for what these stats will look like at the end of the Tournament: Ports is gonna have a crazy win rate of probably 70%, Japan will be above 50%, Russia will be almost as good as Ports, Dutch will be crazy on non-tp map; Brits will be pretty good, Azzy, Spain, French, and Otto are going to have about even win rates, and Germany is going to have a win rate of around 35%.

[momuuu]

I'm going to make a prediction: First picking dutch is still literally not viable, and as a counter pick I think 10+ civs have harder counters than dutch.