Been playing random 2v2 Allies for a week and it has gotten a lot harder. No 2nd IS2 makes running the OST gauntlet harder, and the buffed MG42s gives OST a much better early game tool that when two players build them can lock down 2/3 of most 2v2 maps right at the start.
I don't like Gaurds Motor, but right now it might be the ONLY good doctrine for Soviets.
Balance Data Since The Patch
4 Jul 2015, 15:46 PM
#21
Posts: 824
Youtube Streaming: https://www.youtube.com/@broodwarjc1517/streams
4 Jul 2015, 15:52 PM
#22
1 
Posts: 764
So top 200 2v2/3v3/4v4 random Allies, just have bad luck, and get a harder match more than the Axis random's?
Is that what you're saying?
Not sure what you're saying here either. Large team games don't count as the actual game now?
More people play large team games than 1v1 and 2v2 combined.
Axis have a few more players in large team games. At times when the amount of people searching is low, you will end up having imbalanced teams - ELO (rank) wise. Meaning MM itself can't work with a low player count, and causes biased games.
Goes both ways of course, but...
...some top players on Axis 3v3 / 4v4 will face way lower ranked players more often, than the other way around.
Simple example:
3v3 AT
http://www.coh2.org/ladders/index/1/6/0 < Allies
http://www.coh2.org/ladders/index/0/6/0 < Axis
Looks quite similar, not?
But now let's have a look at 3v3 random mates:
http://www.coh2.org/ladders/index/3/3/0 < USF
http://www.coh2.org/ladders/index/2/3/0 < OKW
Quite a difference.
4v4 is similar, if not even worse.
So in conclusion the perceived balance is way worse than the actual one, and this perception is often biased by match making.
edit: of course real balance issues are a thing as well, like faction scaling, certain offmap synergies for late game etc., but their impact is more neglectable in my opinion, if a lot of games are already "unfair" by skill distribution
4 Jul 2015, 16:01 PM
#23
Posts: 589
Axis have a few more players in large team games. At times when the amount of people searching is low, you will end up having imbalanced teams - ELO (rank) wise. Meaning MM itself can't work with a low player count, and causes biased games.
Goes both ways of course, but...
...some top players on Axis 3v3 / 4v4 will face way lower ranked players more often, than the other way around.
That is actually false. As there are more Axis players in 3v3 and 4v4 searching at any one time, the way the matchmaking works, the higher ranked Allied players would actually play lower ranked Axis players more often, because of cue waiting. But it would be minimal, as it's usually around 60/40 in favour of axis.
Which makes those stats even more peculiar.
4 Jul 2015, 16:03 PM
#24
1 Posts: 764
That is actually false. As there are more Axis players in 3v3 and 4v4 searching at any one time, the way the matchmaking works, the higher ranked Allied players would actually play lower ranked Axis players more often, because of cue waiting. But it would be minimal, as it's usually around 60/40 in favour of axis.
Which makes those stats even more peculiar.
Please read my reply again. I think you missread quite a lot.
4 Jul 2015, 16:56 PM
#25
Posts: 836 | Subs: 5
We already discussed this. Random was used for 2s/3s/4s data rendering them useless. So basically the data for 1s is accurate while any team mode game not measured in AT is useless.
This is it.
The only Stat we should care about for team games is top 200 vs 200 AT ONLY. Anything else is open to many more factors than a balanced AT match.
 | Playercard of Dusty | |||||||||||||||
|
Devils_Brigade: man captain molo why dont you just buy and m3 drive that around instead of playing coh
4 Jul 2015, 16:57 PM
#26
Posts: 692
Really awkward how people can explain away such an obvious advantage in team games, even with a ton of data.
It's like listening to someone try to convince you that the sky is actually green, and you're only perceiving it to be blue.
It's like listening to someone try to convince you that the sky is actually green, and you're only perceiving it to be blue.
4 Jul 2015, 17:01 PM
#27
Posts: 589
This is it.
The only Stat we should care about for team games is top 200 vs 200 AT ONLY. Anything else is open to many more factors than a balanced AT match.
So AT teams don't play random's, is that it?
If they do, the data is still skewed. The only way to have accurate results, would be to combine both sets of data...
4 Jul 2015, 17:04 PM
#28
Posts: 836 | Subs: 5
So AT teams don't play random's, is that it?
If they do, the data is still skewed. The only way to have accurate results, would be to combine both sets of data...
I said AT ONLY, as in AT vs AT. A top 10 allies team vs a top 100 axis team reflects balance better than this current Stat of two teams of some mix of random/at/partially top 200.
| Playercard of Dusty | |||||||||||||||
|
Devils_Brigade: man captain molo why dont you just buy and m3 drive that around instead of playing coh
4 Jul 2015, 17:07 PM
#29
Posts: 589
I said AT ONLY, as in AT vs AT. A top 10 allies team vs a top 100 axis team reflects balance better than this current Stat of two teams of some mix of random/at/partially top 200.
And how do you propose someone differentiates those games, when he only has access to the data provided by this site? AFAIK you cannot separate AT vs AT only...
4 Jul 2015, 17:08 PM
#30
Posts: 836 | Subs: 5
And how do you propose some differentiates those games, when he only has access to the data provided by this site? AFAIK you cannot separate AT vs AT only...
He can't, only Relic can. I'm pointing out that these stats in their current state are a poor metric for people to start discussing balance around because of the multiple flaws.
| Playercard of Dusty | |||||||||||||||
|
Devils_Brigade: man captain molo why dont you just buy and m3 drive that around instead of playing coh
4 Jul 2015, 17:12 PM
#31
Posts: 589
He can't, only Relic can. I'm pointing out that these stats in their current state are a poor metric for people to start discussing balance around because of the multiple flaws.
Unfortunately, we don't have that data yet. Why don't you ask your buddies at Relic to pass it over? It's only data right? It couldn't possibly harm them in any way. Well, unless the data suggests things are skewed in a very clear direction.
DUSTY Stream June 2015:
DUSTY: "Let's play Axis, I'm fed up of losing today"
LEMON: "OK"
4 Jul 2015, 17:14 PM
#32
Posts: 3602 | Subs: 1
In 1vs1, people from top200 can still flank and the better player wins the game. In 2vs2 and above, it is impossible to flank on 80% of the map pool. Axis fanboys in the forum, which are also for a vast majority only playing 2vs2 and above are happy as they just keep telling everyone "Now you must flank ahaha" knowing it is almost impossible early game in those mod.
In 1vs1, it is not the top200 which would be interesting to read but the low levels because here, I can tell you it is a carnage (I'm playing ostheer 1vs1 at low level), this situation is probably going to strongly reduce the pool of fresh USF players, 2xpio + 2xMG42 start is simply ruining the game of many of them. HMG42 doesn't requires you to flank, it requires you to perfectly flank which is not always the case when you are still learning the game.
In 1vs1, it is not the top200 which would be interesting to read but the low levels because here, I can tell you it is a carnage (I'm playing ostheer 1vs1 at low level), this situation is probably going to strongly reduce the pool of fresh USF players, 2xpio + 2xMG42 start is simply ruining the game of many of them. HMG42 doesn't requires you to flank, it requires you to perfectly flank which is not always the case when you are still learning the game.
4 Jul 2015, 17:33 PM
#33
Posts: 60 | Subs: 1
Hi,
I haven't have to time to add AT there yet, but it is coming some day
I like to just say that in my view 1v1 and 2v2 charts shows the reality pretty well. 3v3 and 4v4 charts favors a bit the factions that are more popular.
In 2v2, win-ratios were by and large equal before the newest patch. Now, there is noticeable gap between allies and axis. Soviet was the most popular faction in 2v2, but now Ostheer is. To my mind, the patch did make 2v2 balance worse and also it is pretty easy to see from charts.
I haven't have to time to add AT there yet, but it is coming some day
I like to just say that in my view 1v1 and 2v2 charts shows the reality pretty well. 3v3 and 4v4 charts favors a bit the factions that are more popular.
In 2v2, win-ratios were by and large equal before the newest patch. Now, there is noticeable gap between allies and axis. Soviet was the most popular faction in 2v2, but now Ostheer is. To my mind, the patch did make 2v2 balance worse and also it is pretty easy to see from charts.
4 Jul 2015, 17:39 PM
#34
Posts: 1166 | Subs: 1
Really awkward how people can explain away such an obvious advantage in team games, even with a ton of data.
It's like listening to someone try to convince you that the sky is actually green, and you're only perceiving it to be blue.
If you read any of this thread or the previous threads or even the creator of the data you would understand why. Really its yourself and tatalala calling the sky green...
A scaling effect due to AT pop would yield accurate results.
so...
1v1: top 200
2v2AT: top 150
3v3AT: top 100
4v4AT: top 50
That is the data set that would reveal the most accurate depiction of balance. You scale the pops to cater to smaller competition in large team games.
Taking the top 200 random players (formulated with a random players) in a team games will give you a random data set.
But if you took 2 players who played well together (axis or allies) and continually beat 2 players who played (axis or allies) well you could surmise faction 'x' is stronger.
I can't explain it any further...
4 Jul 2015, 17:49 PM
#35
Posts: 589
If you read any of this thread or the previous threads ... [snip]
So, if for example, the top 5 teams in 4v4 for both factions showed a similar set of results or 'balance', yet the bottom 6000 players showed a swing of 30% or more to a particular faction, that would be fine with you?
Is that what you're saying?
Because that isn't true balance. That is balance within a skill ceiling (so not balance at all).
True balance would be equal data results across all game modes, and all players. An ideal that cannot be achieved.
You cannot, however, claim true balance by looking at the top players only. That is a fallacy perpetuated by the CoH elitist community since forever.
4 Jul 2015, 17:58 PM
#36
Posts: 1166 | Subs: 1
So, if for example, the top 5 teams in 4v4 for both factions showed a similar set of results or 'balance', yet the bottom 6000 players showed a swing of 30% or more to a particular faction, that would be fine with you?
Is that what you're saying?
That was directed at Acolyte, but yes snip.
You aren't comparing apples to apples then. The bottom percentile in team games is void because almost all of those team can break the top 200 just by playing 10 games. This is why you use a scale effect in team games to counter smaller AT populations.
"Why the bottom percentile shows a larger swing"
Paidplayer aka Legends just said this was due to popularity over balance. I think hes spot on.
The argument for random teams vs AT is pretty straight forward. If the top 4 players in the game played mirror matches of themselves in a vacuum, this would yield the most accurate result of balance. Obviously this impossible so we use the next best available data, AT.
4 Jul 2015, 18:00 PM
#37
Posts: 589
You didn't answer my question.
I'll ask it again. This is not in regard to AT only, but data spread out across the entire player base. The entire player base is equal in it's requirements for balance. Why? Because they all paid for the game.
A simple yes/no will suffice.
"So, if for example, the top 5 teams in 4v4 for both factions showed a similar set of results or 'balance', yet the bottom 6000 players showed a swing of 30% or more to a particular faction, that would be fine with you?"
I'll ask it again. This is not in regard to AT only, but data spread out across the entire player base. The entire player base is equal in it's requirements for balance. Why? Because they all paid for the game.
A simple yes/no will suffice.
"So, if for example, the top 5 teams in 4v4 for both factions showed a similar set of results or 'balance', yet the bottom 6000 players showed a swing of 30% or more to a particular faction, that would be fine with you?"
4 Jul 2015, 18:01 PM
#38
Posts: 1166 | Subs: 1
You didn't answer my question.
I'll ask it again.
A simple yes/no will suffice.
"So, if for example, the top 5 teams in 4v4 for both factions showed a similar set of results or 'balance', yet the bottom 6000 players showed a swing of 30% or more to a particular faction, that would be fine with you?"
Yes.
4 Jul 2015, 18:06 PM
#39
Posts: 589
Yes.
Well let's hope that Relic don't use your suggestion for data collection then. Afterall, given your previous scale, you'd be ignoring 99% of the CoH 2 player base.
That's not a recipe for good business practise. Might I also add, all those players you'd be fine with ignoring, also paid for the game.
If they are suffering with balance problems, because their skill level doesn't allow them to play a particular faction as well as another, that is a balance issue, whether you like it or not.
4 Jul 2015, 18:10 PM
#40
Posts: 1166 | Subs: 1
Well let's hope that Relic don't use your suggestion for data collection then. Afterall, given your previous scale, you'd be ignoring 99% of the CoH 2 player base.
That's not a recipe for good business practise. Might I also add, all those players you'd be fine with ignoring, also paid for the game.
If they are suffering with balance problems, because their skill level doesn't allow them to play a particular faction as well as another, that is a balance issue, whether you like it or not.
If the focus is actual balance then I would hope they do use this method. Anything else would be foolish.
If you wanted to recommend some sort of handicap mode then I don't really see how you could achieve a balanced match for 100% of the community.
0 user is browsing this thread:
Livestreams
 |
|
|
19 | ||
|
|
|
|
14 | ||
|
|
|
|
3 | ||
|
|
|
|
2 | ||
 |
|
|
1 |
Ladders Top 10
-
#Steam AliasWL%Streak
- 1.831222.789+37
- 2.34957.860+14
- 3.589215.733+4
- 4.1101614.642+2
- 5.305114.728+1
- 6.916405.693-2
- 7.271108.715+22
- 8.721440.621+3
- 9.1041674.607-2
- 10.17146.788+1
Data provided by
Relic Entertainment
Relic Entertainment
Replay highlight
VS
-
cblanco ★ -
보드카 중대
-
VonManteuffel -
Heartless Jäger
Einhoven Country
Honor it
9
Download
1231
Board Info
839 users are online:
839 guests
2 posts in the last 24h
8 posts in the last week
40 posts in the last month
8 posts in the last week
40 posts in the last month
Registered members: 49077
Welcome our newest member, juliavargascom
Most online: 2043 users on 29 Oct 2023, 01:04 AM
Welcome our newest member, juliavargascom
Most online: 2043 users on 29 Oct 2023, 01:04 AM