More evidence for broken Alliance War Score algorithm ... post-update


#21

Just seen Guardians Ascending’s latest "match"making against another Russian alliance. Same old, same old.

Their 14th ranked player Ban(backwards)ëk has a 3415 defence team. Our number 14 has one of 2860. Their number 11 is level 44. Ours is 25.

Guess how many +3000 players they have compared to us. Guess what the outcome of tomorrow’s war will be.

Guess how motivated our newer players are going to be to participate in future wars. We have already lost one.

Oh, and they are also not “perhaps” a new league - that level 11 has been around for 160d.

DEVs ARE YOU THERE? Can someone take a look at the components of the war score of Guardian’s Ascending’s current opponent to try to find the source of the anomaly. They MUST have hugely robust rosters.

I retract the title in my OP somewhat. Perhaps some alliances have found a way to “hack” the matchmaking algorithm, because sometimes the war score is super fair and leads to exciting contests. In the past 6 wars, 2 have been like that. The other 4 have been a sad joke.


#22

@Kerridoc

It would be really interesting to know how many alliances are "cup droping*.

Our last war we lost against an alliance that was 8 days old. They were a tad strong for their war score, at least compared to us. :slight_smile:

So I am wondering is there any advantage to form a new alliance after a winning streak when you just filled your war chest.

And of course there could be the old switcherooo……

Take x people with 2 accounts each.

They found 2 alliances A and B.

Strong accounts in A play to win until 1 or 2 war chests are full.

Weak accounts in B loose.

Then Change:

Strong accounts to B weak accounts to A. Rinse repeat for constant easyer match up’s.

You see there are quite some possibilities to “game” the system which defenitely could make it harder for the system to produce fair pairings.


#23

Starting a new alliance has a huge cost—lousy titans for a few weeks, and matching lousy loot. There’s no way the war chest gain offsets the Titan loot loss from the strategy you outline.


#24

Which is not to say that people aren’t doing it. Only that any who are, are shooting themselves in the head loot-wise for the pleasure of a few easier wars.


#25

What do you think of the “switcheroo” Concept.

Pro: Easyer wars.

Cons: Worse Titans…….but not that much depending on B-Team.


#26

I think it sounds like too much work for too,little upside—and maybe a net downside. You’ve got to time your leaving careful to manage both chests.


#27

Depends. If 30 players do it with alt accounts then not much work and rather easy to do(although time consuming unless you find 30 players with already progressed alts that can take down 8-9* titans). Already explained this in detail before though. With the right group of players i believe i could make it happen, but i just feel better about earning the rewards and progressing with the group of players I’m already involved with


#28

And worse War loot chest, since you lose participation during the period while your alts are tanking things.

I guess the interesting question to me is not whether the exploit is possible, but how probable is it? The game is big enough that someone is probably trying it But if there are 6 alliances doing it, total, I’d be shocked.

It certainly is not the probable explanation for the OP’s experience.


#29

And the same type of mismatch continues for Guardians Ascending (GA)… now 5 out of 7 of all recent wars
Latest opponent: Swiss solothurn army (SSA)
HIGHEST DEFENCE POWER player: GA 3507 ; SSA 3802
Power of no. 15 player: GA 2720 ; SSA 3156
Bizarrely the SSA trophy score is significantly lower than that of GA.
And no, they are not a new guild. Their leader joined 365 days ago and their weakest player 64 days ago.


#30

Please post the War Scores of each team (visible in the detail pop-up under ? By the alliance score) and, when revealed, the number of participants.


#31

@bedu. i guess these are the 2 alliances involved. If it’s correct, your opponent has a lower war score than yours


#32

Though that difference is insignificant - a few percentage points difference.


#33

Translation (Spanish): Hello greetings, the game seems to me sensational and I’ve been playing more 6 months … but I have a complaint and a disagreement at the time of the wars … since they always match us with alliances far superior to mine … “eye to elbow …” what I ask that wars be more equitable than that there inequality … my people are very unmotivated … I make a way to evaluate the matches and the most coherent in that part … Thank you keep improving.


No 5* heroes
#34

@Kor1sco … Correct, the opponents had a lower war score. That is exactly my point. As I posted previously, their defence teams were significantly higher power. So why a lower war score?

In fact, why even a match?


#35

i believe you know your teammates roster. Are there many 5 star heroes even if they are not developed?
criteria for war matching are:
Top 30 heroes with emphasis in the 5 most powerful
Top troops
Previous wars history

From what I read in the forum the algorithm gets better the higher the war score, though from experience in my ally lack of strategy was the main cause for defeat.
Selecting same color tank, defining targets and not attacking same target twice led us to victory more often.

@Kerridoc, am i missing something here?


#36

How many players were opted in from each side? There is penalty applied to the war score of a team with more players

Also, when were the images taken, before the war or after? If not directly before matchmaking the scores dont mean much, as people could have opted out right after matchmaking started.


#37

i took those images after he posted so I assume after the matchmaking


#38

Your list seems comprehensive. Not sure how the match occurred if it is as unbalanced as it appears. Perhaps @mhalttu can look under the hood.


#39

So it may not be what was used for match making.


#40

@General_Confusion I flagged your post inadvertently. So sorry! But I do have a question about part of it. How could someone take a shot of the opposing alliance info before matching begins?