Login

russian armor

GCS placement match stats (w/ updates)

16 Apr 2017, 20:20 PM
#81
avatar of Grumpy

Posts: 1958

jump backJump back to quoted post15 Apr 2017, 13:11 PMCruzz
So if you take the games where both players won a game (so as to ignore matchups that might have been determined purely by skill gap), the win rates for factions are, assuming I didn't enter anything wrong:

Code

Ost OKW Soviets USF Brits
Count 15 37 17 25 10
Wins: 10 25 3 11 3
Losses: 5 12 14 14 7
Win % 66,7% 67,6% 17,6% 44,0% 30,0%


Soviets looking pretty good there. Brits not too shabby either.


Korean army doubled the Soviet winrate by wining all four of his matches including both as Soviets, at least until your sorting method is applied. Then Soviets back to being on life support.
16 Apr 2017, 20:34 PM
#82
avatar of Cruzz

Posts: 1221 | Subs: 41

jump backJump back to quoted post16 Apr 2017, 20:20 PMGrumpy


Korean army doubled the Soviet winrate by wining all four of his matches including both as Soviets, at least until your sorting method is applied. Then Soviets back to being on life support.


Him winning as soviets won't matter at all, because he won both games. I was just running the numbers on games where both players won one game, because otherwise I'd have to actually look at every single player involved and decide whether the skill gap was too high to include the games.

That said, the new games did improve stats for allies anyway as every single one of the newest 10 games was won by allies.

Code

Ost OKW Soviets USF Brits
Count 20 42 20 27 15
Wins: 10 25 6 13 8
Losses: 10 17 14 14 7
Win % 50,0% 59,5% 30,0% 48,1% 53,3%

16 Apr 2017, 21:01 PM
#83
avatar of Grumpy

Posts: 1958

jump backJump back to quoted post16 Apr 2017, 20:34 PMCruzz


Him winning as soviets won't matter at all, because he won both games. I was just running the numbers on games where both players won one game, because otherwise I'd have to actually look at every single player involved and decide whether the skill gap was too high to include the games.

That said, the new games did improve stats for allies anyway as every single one of the newest 10 games was won by allies.

Code

Ost OKW Soviets USF Brits
Count 20 42 20 27 15
Wins: 10 25 6 13 8
Losses: 10 17 14 14 7
Win % 50,0% 59,5% 30,0% 48,1% 53,3%



Faymonville Approach and Kholodny really good for UKF (maybe allies in general). I haven't looked at any of the replays but think a mortar pit on Faymonville could cover a huge chunk of the map from just outside of the base.
16 Apr 2017, 21:04 PM
#84
avatar of Mr.Smith

Posts: 2636 | Subs: 17

jump backJump back to quoted post16 Apr 2017, 20:34 PMCruzz


Him winning as soviets won't matter at all, because he won both games. I was just running the numbers on games where both players won one game, because otherwise I'd have to actually look at every single player involved and decide whether the skill gap was too high to include the games.

That said, the new games did improve stats for allies anyway as every single one of the newest 10 games was won by allies.

Code

Ost OKW Soviets USF Brits
Count 20 42 20 27 15
Wins: 10 25 6 13 8
Losses: 10 17 14 14 7
Win % 50,0% 59,5% 30,0% 48,1% 53,3%



I'm a bit lazy >.> Could you provide matchup-oriented stats, thx!
16 Apr 2017, 21:30 PM
#85
avatar of Cruzz

Posts: 1221 | Subs: 41



I'm a bit lazy >.> Could you provide matchup-oriented stats, thx!


Code

Ost Sov Ost USF Ost Brits OKW Sov OKW USF OKW Brits
Wins 3 5 2 11 9 5
Losses 2 5 3 4 8 5
Axis Win % 60,0% 50,0% 40,0% 73,3% 52,9% 50,0%

16 Apr 2017, 22:41 PM
#86
avatar of frostbite

Posts: 593

i got a few wins with soviet and a few with usa. its alot more difficult to win but games are still winable. dunno if game is balanced in this mod tho
18 Apr 2017, 08:28 AM
#87
avatar of Siphon X.
Senior Editor Badge

Posts: 1138 | Subs: 2

Updated, added the last 3 games (up to 113 now).

Also, added the other replays (all 110 that are available) for more commander choices.

Here are some map stats (couldn't add those to the original post as it got too long...)

Crossing in the Woods




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
10
10
16
10
6
won
8
3
10
4
1
won(%)
80
30
63
40
17


Crossroads




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
11
13
27
19
6
won
8
4
12
11
3
won(%)
73
31
44
58
50


Faymonville Approach




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
12
10
21
10
13
won
6
2
12
5
8
won(%)
50
20
57
50
62


Kholodny Ferma (Summer)




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
6
4
8
6
4
won
1
4
3
4
2
won(%)
17
100
38
67
50


That's it for the placement matches, thanks to the players and A_E!

More thorough analysis of the meta (much) later...
A_E
18 Apr 2017, 09:12 AM
#88
avatar of A_E
Lead Caster Badge
Donator 11

Posts: 2439 | Subs: 6

Updated, added the last 3 games (up to 113 now).

Also, added the other replays (all 110 that are available) for more commander choices.

Here are some map stats (couldn't add those to the original post as it got too long...)

Crossing in the Woods




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
10
10
16
10
6
won
8
3
10
4
1
won(%)
80
30
63
40
17


Crossroads




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
11
13
27
19
6
won
8
4
12
11
3
won(%)
73
31
44
58
50


Faymonville Approach




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
12
10
21
10
13
won
6
2
12
5
8
won(%)
50
20
57
50
62


Kholodny Ferma (Summer)




ostheerostheer sovietssoviets okwokw usfusf ukfukf
picked
6
4
8
6
4
won
1
4
3
4
2
won(%)
17
100
38
67
50


That's it for the placement matches, thanks to the players and A_E!

More thorough analysis of the meta (much) later...


Fantastic Siphon X great work, I will now be compiling all results, applying modifiers for ELO and points accordingly, this will then work in combination with the captured automatch ranks to produce a tournament seeding. I will be releasing the provisional seeds today.

I however urge all players to ensure passports are sent in, this is a tournament that ends in a live venue final, being a paas port holder is vital. If you don't have one yet email me on redditcoh@gmail.com to say how you plan to get one and how long it will take.
24 Apr 2017, 09:29 AM
#89
avatar of Siphon X.
Senior Editor Badge

Posts: 1138 | Subs: 2

Updated with the final rulings on certain games.

Also, unpinned and decolored...
24 Apr 2017, 10:24 AM
#90
avatar of borobadger

Posts: 184

Has the poor performance of SOV indicated to the patch makers that they need a bit of help?
24 Apr 2017, 12:30 PM
#91
avatar of Vipper

Posts: 13496 | Subs: 1

Has the poor performance of SOV indicated to the patch makers that they need a bit of help?

You mean like the 100% win rate at Kholodny Ferma (Summer)
24 Apr 2017, 22:37 PM
#92
avatar of borobadger

Posts: 184

jump backJump back to quoted post24 Apr 2017, 12:30 PMVipper

You mean like the 100% win rate at Kholodny Ferma (Summer)


Look at the rest of the maps. And 4 games is nowhere near enough of a sample size.
24 Apr 2017, 22:55 PM
#93
avatar of Vipper

Posts: 13496 | Subs: 1



Look at the rest of the maps. And 4 games is nowhere near enough of a sample size.

and you check here where Soviet have the highest winrates 70%...
https://www.coh2.org/topic/60897/gcs-main-round-stats-w-updates

Samples are to small for any safe conclusions.
24 Apr 2017, 23:01 PM
#94
avatar of borobadger

Posts: 184

jump backJump back to quoted post24 Apr 2017, 22:55 PMVipper

and you check here where Soviet have the highest winrates 70%...
https://www.coh2.org/topic/60897/gcs-main-round-stats-w-updates

Samples are to small for any safe conclusions.


Samples are too small you're right. I guess the stats from normal ranked games rather than tournament play would be a better indicator but I'm not sure if those can be seen anywhere?
25 Apr 2017, 07:27 AM
#95
avatar of Siphon X.
Senior Editor Badge

Posts: 1138 | Subs: 2

Has the poor performance of SOV indicated to the patch makers that they need a bit of help?


Well, most if not all games here were done with an earlier version of the GCS patch. For the main event the current v1.4 version is used which indeed brought a few buffs to the Maxim.



Samples are too small you're right. I guess the stats from normal ranked games rather than tournament play would be a better indicator but I'm not sure if those can be seen anywhere?


You can check out coh2chart. But I'd also recommend reading this article which tries to explain why coh2chart might not give you what you want if you are looking for "statistics on the true faction balance". The TL;DR version is: The problem is that the matchmaking algorithm tries to eliminate what you want to observe, so you basically mostly get an impression of how well that is working.

In reality we can't realistically quantify the faction balance. We are pretty much left with looking at different statistics without jumping to conclusions...

That is, if we are interested in this at all. That's probably kind of a controversial statement, but at the end of the day unless it is really bad, faction balance is irrelevant in most cases:
  • When playing automatch games it won't matter as the matchmaking should take care of that (although you could argue that the matchmaker has it easier when the factions are balanced).
  • For tournements it doesn't matter as long as you play at least BO3 rounds (actually, with proper rules you can even have BO1 rounds; ok, viewing might be more entertaining when factions are balanced...)


All statistics have strength and weaknesses. The good thing about this data set (unlike the current main round statistics) is that players are probably more on an equal skill level. The number of samples isn't great, yes. It's not totally horrible either, so at least for certain matchups and maps the numbers could actually indicate some trends (notice the vague wording :) ).

The issues, apart from the less than stellar sample size are the map choices which will certainly skew the numbers (while it is a bit too early for definite statements, it is noteworthy that the current main event stats include several top players that feel more comfortable playing SOV so that Soviets were picked more often, as was Kholodny Ferma). If you are interested in the effect of maps you can check out earlier breakdowns like this one for ESL.

What I personally suspect to be a very relevant factor is also that the meta hasn't settled. This tournament happens shortly after the WBP so the meta changes coming out of that patch are not clear. On top of that, it's not even WBP but the GSC which likely most players haven't tested much if at all.

So, basically the game dynamics have been shaken up and have not settled yet. It might very well be that a faction might seem very weak now, but that eventually people will figure out ways to play in a different way so that it might end up being considered OP...

25 Apr 2017, 08:38 AM
#96
avatar of borobadger

Posts: 184

Thanks for the detailed reply. I do main Soviets and they don't feel too bad to me to be honest, I'm just wary of how the Maxim changes will feel but to be honest a wider arc machine gun might actually be quite nice on some occasions. They seem to have had nothing but nerfs for months certain builds work exceptionally well against them in their current state.
1 user is browsing this thread: 1 guest

SHOUT IT OUT!

No ProfanityNumber of ShoutsRefresh Shout Box
Lady Xenarra: @Willy Pete The lack of April Fools this year is odd lol
Last Wednesday, 01:34 AM
Willy Pete: @Rosbone not dead yet. when that happens the font will switch to Papyrus :*(
Last Wednesday, 00:16 AM
dasheepeh: it was an honor guys :guyokay:
Last Tuesday, 20:34 PM
aerafield: yeah I already prepared my "Can't believe there's comic mode for the 10 daily visitors even on this April 1st" :guyokay:
Last Tuesday, 20:29 PM
Rosbone: @dasheepeh I guess that means this site is officially dead :guyokay:
Last Tuesday, 20:19 PM
dasheepeh: no comic sans font for april 1st this year?
Last Tuesday, 19:56 PM
Willy Pete: @Lady Xenarra this you? https://i.imgflip.com/3e4thi.jpg
Last Tuesday, 02:53 AM
Lady Xenarra: Does anyone else think that USF needs buffs? It feels like they’re on life support sometimes
Last Tuesday, 02:36 AM
Willy Pete: @Rosbone Ahh I missed that memo. I still think its a bad decision though. Adds frustration for players and isnt gonna make them that much money
27 Mar 2025, 15:46 PM
Rosbone: It is also good they left it free until after the free to play weekend. Points for that.
27 Mar 2025, 09:34 AM
Rosbone: But I agree, the cost to get a full decent Coh game pushing $115 US is not the best idea. Especially when it needs so much more work for casuals.
27 Mar 2025, 09:32 AM
Rosbone: To be fair, it was a thank you to early fans right? They said it was not free for long and it would become a pay DLC at some point.
27 Mar 2025, 09:30 AM
Willy Pete: Re-releasing free DLC so they can charge new players money for it. Brilliant marketing strategy :clap:
27 Mar 2025, 04:31 AM
Soheil: Coh2 still broken server ?
25 Mar 2025, 18:27 PM
Rosbone: Congrats to Relic. Looks like Coh3 has finally usurped Coh2 s the popular Coh. You smell terrific. :snfQuinn:.
24 Mar 2025, 02:46 AM
Nickbn: and again someone else replies. I mean come on guys. Give @adamírcz a chance
22 Mar 2025, 14:00 PM
Willy Pete: @Nickbn you didn't ask a question, and this is a chat box...
20 Mar 2025, 13:11 PM
Nickbn: @Rosbone it's incredibly rude to speak on someone elses behalf, especially when a question is directly adressed to them. I understand your passion for the subject at hand but I want to hear from him.
20 Mar 2025, 10:16 AM
Rosbone: @Nickbn No, I am just saying people should not be using any Relic owned forum since they have proven they ban anyone who says true things about Coh3.
18 Mar 2025, 19:01 PM

Ladders Top 10

  • #
    Steam Alias
    W
    L
    %
    Streak
Data provided by Relic Relic Entertainment

Replay highlight

VS
  • U.S. Forces flag cblanco ★
  • The British Forces flag 보드카 중대
  • Oberkommando West flag VonManteuffel
  • Ostheer flag Heartless Jäger
uploaded by XXxxHeartlessxxXX

Board Info

568 users are online: 568 guests
3 posts in the last 24h
14 posts in the last week
74 posts in the last month
Registered members: 53281
Welcome our newest member, ScarlettdfPhelps
Most online: 2043 users on 29 Oct 2023, 01:04 AM