Tbh technically the yoloswag clan kind of tried to abused the system and fizzer’s matchmaking system punished them for it. If a new clan were to show up with 0 rating, they would just get free wins on the first day and there wouldn’t be much issues afterwards cause their rating would eventually be adjusted for in the following day. New clans wouldn’t even have 40 new players but maybe around 10ish people who would want to play, so getting free wins on the first day and being able to play games after the first day would be fine. But since yoloswag was created to win clan wars, the system punished them for it by not giving them enough games for the amount of players they had. Of course the matchmaking system sucks but if you try to manipulate it, with both a negative rating as well as 40+ players then obviously this would happen.

But anyways, please add more templates Fizzer. At least you can fulfill this request? The templates we have now are pretty stale.

If only someone had foreseen this and suggested changes to the matchmaking months ago...

This forum is full of yolo fans so i never expected a support grp or anything. And secondly @Tacky, you could stfu as i am a player of this game and have the right to express myself and share my opinions. If you have a problem with that, then either leave the game or just ignore my comments.

https://shorturl.at/rCFW4
this you arrow?

people who invest time to play this game do have a say in what they feel about stuff.

Is this you @Rick

Edited 9/16/2021 15:06:55

https://www.warzone.com/MultiPlayer?GameID=28063840

this you? enjoy your ban. hopefully it'll be permanent this time.

LOL Tacky, this isn't just for the trolls' sake, kind of hoping Fizz takes note too. Not for our benefit, but for the betterment of CW as a whole.

To further exemplify the point . . . (and oddly this will unintentionally include something close to Arrow's heart, lol, but just picking the bottom rated clans)

How are these clans ever expected to get games in CW? Yes, it's their fault for losing games, sure - but it's not their fault for not getting games. In order to improve, you need games assigned. What other matchmaking system in the world punishes low rated players by not even letting them compete at all? Either you're in the competition or you're not, and if you are, you get a game.


(recapping in case the pic gets cut off for some)
#531 Special Forces -70.26
#532 Sons of Ragnarok -73.02
#533 Ruby -73.48
#534 The Fancy Dot ● -92.53
#535 Cacaoçian Empire -108.08
#536 Raging Penguins -232.07

And when did we hear the players from these clans complain in the forums? Right, we haven't ... why is that? B/c they aren't your hardcore dedicated player base that will persevere enough to inspire a change. So are you really fine with those who remain silent to continue to be punished/disadvantaged...?

Btw, also heard from JK that TLMS CW players gave up CW b/c they just never got games. This is a problem! This is an established clan that just can't get games. It's just not fun, and eventually people will give up. How is this okay? No one is asking for more free wins or easier matches, etc - just to get games assigned to be able to compete.

@Fizzer
Choose one:
-> want people to play Clan Wars
-> don't allow people to play Clan Wars

Edited 9/16/2021 16:38:38

arrow having a stupid opinion is like having a penis, you're allowed to have one but don't shove it down our throats pls

Guys, the more toxic and cluttered this thread gets, the more likely you make it that it'll get ignored. Just be clear and concise and keep the drama on Discord.

I'm really surprised people are actually somewhat divided on the subject. Just take your opinion of yolo clan or any other bias you may have from your own clan perspective out of the equation, and it's really obvious. The situation where there are players joining timeslots, and not getting any games or free wins, because of matching algorithm, should never happen.

And no, this won't be solved if we simply wait a bit for the rating to adjust; the lowest rated among most active clans will ALWAYS have a significant chance to be subject to such nonsense, as we have seen in previous seasons (obviously this can also happen to active clans with not-lowest rating, but in a smaller degree).

I always knew Wid was a smart guy.

Wid is wise. Aptly and succinctly stated. This should just never happen. Players investing time and getting rejected by the system is simply not an acceptable outcome.

Alright, so I was looking back at Discord to find some pics of when TLA had problems getting matches.
I've found a few interesting things:

People have been complaining about the matchmaking since Season 0 (the test season)

On 16/03/2021, a new update went live that claimed "– All: Clan war ratings no longer go negative.", which from Muli's script clearly proves to be wrong.

The first screenshot of loads of players not getting games was send on 17/03/2021:
Note that from those 10 people, 4 TLA vs non-TLA games could have been made.

https://www.warzone.com/Forum/536651-matchmaking-clan-wars-bugged is the first forum I could find about clans with lots of players not getting matched up. The final post in the forum mentions Rento and others reporting the bug, but 5 seasons doesnt appear to have been enough to get it fixed. (But maybe this bug was that free wins count as a loss for the rating, instead of a win, in which case it is fixed)

Players investing time and getting rejected by the system is simply not an acceptable outcome

Especially if it happens so repeatedly.

I picked a random collection of 6 consecutive recent time slots (1135-1140, renumbered 1-6 in my rough table below).

Xera signed up in 5 of those 6 slots and got 0 games- not even a free win. 3 other players (krinid, ThanosP*n*s, Tainted Monk) signed up for 4/6 of the time slots to get 0 games. 14 more players signed up for 2+ time slots and still got 0 games.

Of 81 sign-ups, 64 got no games. 38 players signed up for yolo in these time slots; only 8 got games. 26 didn't even get a free win.

(my notes, for which I make no guarantees of correctness):

                          1 2 3 4 5 6  Attempts  Voids  Free  Games
Xera                      v v   v v v         5      5     0      0
CleverTacticButFail       v w     v g         4      2     1      1
krinid                    v v v     v         4      4     0      0
xXxth4n05p3n1sxXx         v     v v v         4      4     0      0
Tainted Monk                v v v   v         4      4     0      0
[Blitz] awesomeusername     v   v   v         3      3     0      0
Princess Jahid Jardin     v v     v           3      3     0      0
SkekSo                        v v v           3      3     0      0
UsualSuspect              v       v v         3      3     0      0
Platinum                  v w       v         3      2     1      0
Stales78                  v w   v             3      2     1      0
John Smith                v       v           2      2     0      0
Quicksilver               v v                 2      2     0      0
Ragnar                    v       v           2      2     0      0
ZelZeiii                  v       v           2      2     0      0
Little_Smurf                v     v           2      2     0      0
Caoimhin                    v       v         2      2     0      0
Zealoustowski               v v               2      2     0      0
Ocean0.04                   v v               2      2     0      0
MenorcanPirate                v     v         2      2     0      0
Cicero_                           v v         2      2     0      0
Krzystof                          v w         2      1     1      0
Tac(ky)tical              w   v               2      1     1      0
Bill Green                v g                 2      1     0      1
Obsidian                        v g           2      1     0      1
Phaeril                   w       g           2      0     1      1
[20]Wini                  g       w           2      0     1      1
ergano                            v           1      1     0      0
Lym                       v                   1      1     0      0
BlaBla                      v                 1      1     0      0
DonkeyTeeth                 v                 1      1     0      0
Dom365                              v         1      1     0      0
Beaumains                         g           1      0     0      1
otter777                      g               1      0     0      1
Z                             g               1      0     0      1
---
TOTAL                                        81     64     7      8
TIME SLOT 1                                  17     14     2      1
TIME SLOT 2                                  16     12     3      1
TIME SLOT 3                                   9      7     0      2
TIME SLOT 4                                   7      7     0      0
TIME SLOT 5                                  17     13     1      3
TIME SLOT 6                                  13     11     1      1

(v/Voids = no game, no free win; w/Free = free win; g/Games = actually got a game, excluding free wins; v/w/g/Attempts = signed up for that time slot)

Edited 9/16/2021 23:26:06

upvoted 52/65 = .8

basically: 80% of our games are voided

Edited 9/16/2021 19:45:41

you would have lost them anyway

[in response to JK's post about the 10 players not getting games]
That is even more perplexing!

I would say skip the current 4 free wins doled out, create 4 games of TLA vs non, and then still 2 TLA players are out of luck (but with a better algorithm on the whole timeslot and not just these 10 that didn't get games, maybe even this could have been avoided, since it is an even # of players still).

I'm not clear on what the point is of foregoing games altogether when you have opportunities like the above. It's one thing like recently with Yolo when all remaining players are from 1 clan, but when there are still people remaining from different clans, just pair them up.

Maybe that is one the buggy parts that got fixed already? Does anyone know if this has happened again since? (multiple clans in same template that could be paired together not getting games)

As for free wins, they only seem to give 1 pt. So yes, technically a "win" and counts towards your clan's territory ranking, it doesn't help much with the rating. You still need legit wins to get the proper sized differential rating change.

Edited 9/16/2021 19:55:44

A free win should not give just 1 point, but a lot more. Maybe the rating will go too high, but at least that clan can get some games next slot. If the rating is too high for the clan, they will simply lose and drop down again, but at least they can play in half of the slots, instead of none.

Anyhow, if Activision Support reads this thread again, here's my request to them:

If this is a time or developer bandwidth or priority issue, I'd be happy to find or make available a well-tested package in the correct language or framework to provide an implementation of one of the optimal matchmaking algorithms I mentioned. Something with similar semantics to pairlib in Python (https://pypi.org/project/pair/).

Just let me know the general interface requirements and what language/framework/dependency requirements you have. I'm convinced this is not a hard problem to solve (at least on the algorithmic side; there's some value to tweaking the graph-generation logic to fine-tune matchmaking behavior) and am willing to bet my time on this.

Edited 9/16/2021 20:27:07