Login

russian armor

COH2stats.com match and player statistics 2.0

PAGES (13)down
3 Apr 2021, 17:30 PM
#1
avatar of pagep

Posts: 164 | Subs: 1

Hello everyone

We have finally released https://coh2stats.com/

We are not able to get the data from relic easily so the statistics are kind of "crippled" and may not be accurate. The crawler process is described here https://coh2stats.com/about

For the desktop version of the app find the forum discussion here.


Community:

Big thanks to everyone here on the forums who helped with figuring out lot of stuff.
I would like this endeavor to be still community driven. All the source code is open source, and you
can find it here https://github.com/cohstats/coh2stats any contributions are welcomed.

You can find the exported data from the game, for example commanders and bulletins here
https://github.com/cohstats/coh2statstree/master/packages/web/src/coh/data

You can export also these things yourself
https://github.com/cohstats/coh2stats/tree/master/scripts

Next Steps
Please feel free to suggest any features you would like to see.
If you have GitHub account please create a issue for it so we can discuss about that feature in better manner https://github.com/cohstats/coh2stats/issues or you can write it here.

I would like to also know on which part we should mostly focus on.
Some of these things are really big things. Also I have daily job so I am able to squeeze just couple of hours/week, sometimes not even that, so don't except much soon.

General Statistics - DONE
- We could add more graphs, more analysis of the current data
- Be able to select any date range, from the start till the end



Map statistics - DONE
- We could generate stats per map, so we could see win rate for each faction on each map and see if
some maps are unbalanced for some factions

Match analysis - DONE
- So you could open single match and see the enemy stats, their ranks, charts who did most dmg etc.
- For example I often after the match want to see if the enemy was really good or I just played like shit, so I always need to open the player profiles to check this out


Match storage
- Currently we are streaming the matches from the relic API directly.
- Relic keeps only 10 matches / mod / player ( only 10 games 1-1, 10 games 2-2 etc )
- Players could log in and check that they want their matches to be stored "forever" - for some extended period of time

Player profiles - DONE
- We could display the player profiles with more info, for example right now you can see rank, but you have no idea if it's a good rank or not, because each category has different amount of players, we could display percentile aka how good is your rank in compare to all the players

Player statistics
- When we analyze the matches, we could start making analysis for players too, for example their winrate of each map
- Players would need to log in and check that they want this to be included

Ladders - DONE
- I had idea about displaying the ladder history in some cool chart, for example the top 10 spots, but overtime in a chart, you could see how players are climbing the rank, or falling down each day

You can vote on the features here https://linkto.run/p/Z504GGR4
3 Apr 2021, 17:31 PM
#2
avatar of Katitof

Posts: 17914 | Subs: 8

Looks great, that'll be helpful for everyone.
3 Apr 2021, 17:41 PM
#3
avatar of Vipper

Posts: 13496 | Subs: 1

nice work well done
3 Apr 2021, 17:43 PM
#4
avatar of Rosbone

Posts: 2145 | Subs: 2

Looking fantastic!

Just need to make a Battle Planner now :sibHyena:
3 Apr 2021, 18:33 PM
#5
avatar of elchino7
Senior Moderator Badge

Posts: 8154 | Subs: 2

I suggest you also try posting it on the coh2 sub

I suggest at least Arranged teams 2v2 to be added.
3 Apr 2021, 20:57 PM
#6
avatar of pagep

Posts: 164 | Subs: 1

I suggest you also try posting it on the coh2 sub

I suggest at least Arranged teams 2v2 to be added.


Yep it would be good but I don't know how to do it. Logged the issue at https://github.com/petrvecera/coh2ladders/issues/21

There are matchtype_id and statgroup_id based on https://coh2-api.reliclink.com/community/leaderboard/GetAvailableLeaderboards?title=coh2
The matchtype means, 2v2, 3v3 etc and statgroup means random or team.

The problem is that single match object we get from relic doesn't have the statgroup, only matchtype or maybe I don't see it anywhere.

So right now can't tell if the match was played with the team or not :(
https://coh2-api.reliclink.com/community/leaderboard/getRecentMatchHistory?title=coh2&profile_names=%5B%22%2Fsteam%2F76561198034318060%22%5D
3 Apr 2021, 22:01 PM
#7
avatar of OrangePest

Posts: 570 | Subs: 1

Personally interested in player profile stuff
3 Apr 2021, 23:10 PM
#8
avatar of Hannibal
Senior Moderator Badge

Posts: 3114 | Subs: 2

This site looks very nice. Congrats on your good work so far!


One question though: could you elaborate which games you are parsing at the moment? Are these all games played, a random selection or is there some bias in it?

Regarding the random/arranged team: can you check if the players have the same rank? If so, they should be arranged
4 Apr 2021, 11:03 AM
#9
avatar of pagep

Posts: 164 | Subs: 1


One question though: could you elaborate which games you are parsing at the moment? Are these all games played, a random selection or is there some bias in it?

It's described here https://coh2stats.com/about

We are currently crawling top 200 positions from all kinds of COH2 leaderboards (1v1,2v2 etc). This gives us 5200(~3000 unique) top players for the given day. We than proceed with analyzing/saving their matches for the given day. This gives us ~4000 matches / day.
And we are tracking only auto-match games which are against players (filter out vs AI games).
More technical description of this process can be found here

Basically all games which includes any player who is in any top 200 in the leaderboards . So for example player could be top 200 in 1v1, but we proceed all his matches, so if he played some 4v4, those would be proceed too.



Regarding the random/arranged team: can you check if the players have the same rank? If so, they should be arranged

Unfortunately the match details does not have this information. It has only player profile ids. So when processing each match I would have to look up each player. And I can't store this information in our own database, because that might not be up to date - for example when you have new team. So as I see it the only way would be to look it it at at the Relic API. But I am afraid that would be too much stress on the Relic API and they wouldn't like it. For example 1k matches 4v4, would require 8000k API calls for the player profiles.
4 Apr 2021, 12:42 PM
#10
avatar of aerafield

Posts: 3032 | Subs: 3

looks like axis are actually OP in teamgames :D
4 Apr 2021, 12:43 PM
#11
avatar of OrangePest

Posts: 570 | Subs: 1

Looking over these stats, holy fuckamoly the okw winrates xD, I think it's because they're really hard to play at lower level, and then completetly flops in higher level games once you start running into good players who can abuse every weakness (though it doesn't help that the entire enemy tank rooster can outsight you and out trade you.)

Also probably due to the extremely punishing tech, since i started playing them in 1v1 again i consistently have to skip p4 because the normal arrival rate of that piece of shit is like 20 min lol.
4 Apr 2021, 12:58 PM
#12
avatar of pagep

Posts: 164 | Subs: 1

looks like axis are actually OP in teamgames :D


Looking over these stats, holy fuckamoly the okw winrates xD, I think it's because they're really hard to play at lower level, and then completetly flops in higher level games once you start running into good players who can abuse every weakness (though it doesn't help that the entire enemy tank rooster can outsight you and out trade you.)


Unfortunately we can't really say if axis has that bigger winrate over the allies in teamgames. What we can say for sure is, that if strong player is involved axis has bigger win rate.

4 Apr 2021, 13:45 PM
#13
avatar of OrangePest

Posts: 570 | Subs: 1

jump backJump back to quoted post4 Apr 2021, 12:58 PMpagep




Unfortunately we can't really say if axis has that bigger winrate over the allies in teamgames. What we can say for sure is, that if strong player is involved axis has bigger win rate.



Im talking about the 39% winrate for 1v1
MMX
4 Apr 2021, 14:06 PM
#14
avatar of MMX

Posts: 999 | Subs: 1



Im talking about the 39% winrate for 1v1


did you check the daily or weekly stats? i've seen quite some fluctuation over different days, on 1. apr, for example, OKW has a ~44% winrate, as do USF and Brits. though on average OKW seems to lag behind the other factions by roughly 5%
4 Apr 2021, 14:32 PM
#15
4 Apr 2021, 16:00 PM
#16
avatar of 1st. Fallschirmjäger

Posts: 67

I love how in 1 vs 1 jaeger armor is like 12th place, but 2 vs 2 and up it jumps to 1st pick.
4 Apr 2021, 16:23 PM
#17
avatar of Descolata

Posts: 486

Holy jesus, you werent kidding. For everyone besides OST, the leading doctrines are 20-30% more played. Jaeger armor is played 100% more often than #2. Seems... problematic.

I'm not suprised by the other 4v4 picks, USF leans on Urban Assault and Infantry, so they appear to always have artillery. 4v4 is all about blob control, so that makes sense.

UKF leans on Ro. Artillery, but has pretty good diversity otherwise. Soviets lean on ISU doctrines and IS-2, but definitely not to the extent of Jaegar Armor (wow, that doctrine is picked a lot).

OKW has a healthy commander pick rate with significant variety, emphasising early game doctrinal infantry. Turns out pros think Volks dont cut it. OST might carry OKW through team games.
4 Apr 2021, 16:43 PM
#18
avatar of Vipper

Posts: 13496 | Subs: 1

Holy jesus, you werent kidding. For everyone besides OST, the leading doctrines are 20-30% more played. Jaeger armor is played 100% more often than #2. Seems... problematic.
...

And the problem is Allied TDs and the fact people keep Ele in the load out just in case goes ISU.
(numbers show commander loadouts not commanders picked)

I guess many people are also testing the AT strafe.

More interesting is how much Osttruppen dropped in preference in 1vs1. Typical Nerf hammer
4 Apr 2021, 17:50 PM
#19
avatar of 1st. Fallschirmjäger

Posts: 67

jump backJump back to quoted post4 Apr 2021, 16:43 PMVipper

And the problem is Allied TDs and the fact people keep Ele in the load out just in case goes ISU.
(numbers show commander loadouts not commanders picked)

I guess many people are also testing the AT strafe.

More interesting is how much Osttruppen dropped in preference in 1vs1. Typical Nerf hammer


Yes it is really hard to counter massed Allies TDs with anything other than a elefant/jagdtiger in 4x4, so most peopl(me included) keep it in the loadout.
4 Apr 2021, 17:58 PM
#20
avatar of Descolata

Posts: 486



Yes it is really hard to counter massed Allies TDs with anything other than a elefant/jagdtiger in 4x4, so most peopl(me included) keep it in the loadout.


Is Fortified Armor such crap that Jaeger Armor just superior? The ISU doctrines are pretty competitive with each other, Jaeger Armor just... isnt.
PAGES (13)down
1 user is browsing this thread: 1 guest

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

775 users are online: 775 guests
0 post in the last 24h
8 posts in the last week
34 posts in the last month
Registered members: 49107
Welcome our newest member, Falac851
Most online: 2043 users on 29 Oct 2023, 01:04 AM