War Matchmaking Issue -- Proposed Solutions (Developer response: post 107)


Our proposal to fix Alliance War in Empires & Puzzles
@ChelleATC, @zombies, @math4lyfe, @SADERSpanda, and many others who have contributed and are not listed. We have spent the better part of 2 years tracking this data, and would like to make the following proposal:

Current problems as seen by an overwhelming amount of posts on the forum:
• Unfair matching in alliances with less than 30 members.
• Unfair matching between new alliances and established alliances based on lack of war history.
• Matchmaking exploited to get easier matchups to fill more chests.

Solutions to these problems:

1. Roll back the War Block from 5 to 3 teams.

  • Matchmaking is done on a ladder. In alliances that are less than 30 members the gap between the rungs of the ladder is too big. Thus, creating a great disparity in teams that are matched.

  • Reducing the number of rungs, the alliances have to fight on from 5 to 3 will only help match alliances with closer scores, thus creating better match ups, that will be more “fair.”
    image

  • Above is an example of how the rungs maybe spread out. In this example the weakest team from the top tier alliances can choose to drop a member, thus making their opponents much easier to defeat, basically turning the war results from 50% wins to 100% wins. It not only gives an unfair advantage to the less noble team, but also penalizes teams that haven’t dropped members to game the system.

2. Start new alliances at some number of penalties greater than zero, for example, 15 penalties.

  • In the current system, a team can decide to remake their alliance, when they do this their war history is reset. If the team was above average, they are now in line for between 2-4 war chest fills without a loss.

  • The fact that this can be done, throws weaker and less established alliances to the proverbial wolves. The match ups that have been illustrated by one particular alliance on the forum, have caused players to place 1* defenses and even quit the alliance altogether. This can cause severe mis-matches for up to 2 months, affecting 16-20 teams.

  • After an alliance gets to their “normal” war penalty number - they can repeat the process to string war chests indefinitely. So, this problem can affect an unknown amount of underdeveloped alliances.

3. Tiered rewards for War Chests, (or individual war loot,) based on a comparison of opponent strength.

  • As we have all learned throughout our lives, it is never a practical decision to take something away from someone in regard to motivating them toward an intended behavior.

  • Similar to how titan loot is tiered for killing bigger and bigger titans, war loot should be tiered. This will motivate alliances to stay full and to not want to restart for easier match ups.

    • Staying full – This explains itself, the more members an alliance has the higher the war score they have, the higher tier they fall into, the better rewards they get.

    • Easier war match ups – The easier match ups will come with a reduction in rewards from where the alliance was before the restart.

  • Another bonus, based on a tier system being put in place, could be similar to a rare titan loot roll for long term alliances.

    • After filling 5 war chests as an alliance, the alliance will get a rotating bonus roll for a 4 star ascension material on every or every other war chest.

    • The odds for such a roll could be based on longevity in the alliance.

      • For example, 5% for 30-60 days, 10% for 60-150 days, 25% for 150-365 days, 50% for 365 days or more.

In summation, the changes detailed here should help even out matchmaking, by creating stability with alliances and removing two exploits of the current system. It is not easy to have a plan the caters to each and every individual, but we believe that, if the entirety of this plan, is put into place, we could expect to see it help the attitudes of 99.7% of the player base. In a time, of #nospend and several members leaving the game. I believe these changes could help reinvigorate the player base and provide some much needed motivations going forward.

Thank you for taking the time to read our proposal,

Your friendly neighborhood Panda.
image

17 Likes

I wish I could mash the like button 1000x’s!

Thank you for taking the time and Energy to write this up. This adjustments to our current war system will make the War/Match making spot on.

16 Likes

@mhalttu

I’d be curious to hear your thoughts on this

8 Likes

This is VERY important. Just took a short break from the top 100 to help out a few training alliances. it was amazing the disparity in wars once you leave full alliance, honorable play territory.

14 Likes

That all makes sense to me. It brings together some of the proposed ideas I’ve seen into a coherent system of proposed change. Has my vote.

A question out of curiosity, though whether anyone has the answer is another matter.
Do we know the difference in war score between the bottom of 30 member alliance group and top of the 29 member alliance group? (In an ‘ideal’ scenario I’d imagine the gradation in war score to be fairly smooth, if that makes sense.)

3 Likes

Excellent writeup, you have my vote. :slight_smile:

7 Likes

Absolute and utter bump for me. This is not optional this is necessary. Why does it take members to think of these things rather than the developers of the game. It isn’t rocket science. Sorry I just find it annoying the lack of care and due diligence allowing for issues like this to even exist when the game has been around for years.

2 Likes

Math

While it makes sense from an alliance point of view to track titans based on alliance, it makes very little sense to track wars based on alliance.

Just track individual proficiency

(Warscore Manipulation (not matchmaking) - #51 by Gryphonknight)

For first war, just use the current raid trophies as a seed.

This system has a theoretical infinite “war penalty”.

Tiered

Tiered loot like Wanted Hero Mission chest would be the easiest to implement. And would address the problem.

I would prefer a leaderboard reward system but that is another rant.

Streak bonus

I would rather see a loyalty bonus

Use all 6x war energy, get a bonus.

Participate in 8x consecutive wars, get a bonus. 8x average of 3 points= 24 points towards a chest.

Open 3x consecutive war chests in the same alliance, get a bonus. Ratio of many rewards in Empires was originally 1 to 3.

13 Likes

@Gryphonknight

I’m not sure if you mean the current system has infinite penalties or and individual system has infinite penalties. I can assure you, based on the data collected since November of 2019, the current system does not.

Individual penalties mean more alliances taking smaller match up advantages. How many players do you know with a very good alternate account? Imagine that player, rotating accounts in and out to basically never create a “personal war history.” It would continually cause the alliance match ups to be skewed in the favor of the “rotater’s alliance.” While this isn’t as egregious as starting a new alliance, it probably could effect a lot more alliances.

I was not around when individuals carrying personal scores was the war system, but I have discussed at length with @Zombies. He may be able to explain better than me.

Panda

1 Like

Would be great if tournament rules would also apply to wars.

2 Likes

I love it, except… I’m anti-teired loot especially in this atmosphere. It removes one problem, creates another.

By doing so, you’re unintentionally punishing others for not living up to the A+ Expected Standard.

As it stands, it’s rather difficult and insane to try and recruit. People are like wolves if you talk in almost any room whilst unaffiliated. Beyond that, you’re forcing those who don’t engage in such behavior and prefer small groups by basically their loot and effort.

Whether 5 or 10 wins to get a chest, that is still effort and dedication for 100% participation.

Teired loot will unintentionally yet most definitely have collateral damage. War is already predominantly repetitious as is, and tiering everything only adds to that trend of helping the paying to wins increase that gap over cheap and free to play-players.

Reminds me of this nerfing of Mercs all over again. Not saying the course of action was inherently right or wrong, irrelevant.

Regardless, it affected far more than the few it was aimed at. Just because one can do something, doesn’t mean everyone is.

Cheaters and shadiness will always exist. Aside that, whole heartedly agree. Tiered loot, however, is just punishing everyone who isn’t a top alliance, with all paying members, and easier to maintain numbers.

3 Likes

Those in small alliances already opted out of caring about loot though or else they wouldn’t be in small alliances, they would be in full alliances tryin to climb the titan ladder and reach war cap…

3 Likes

Not sold on Tierd loot more on participation… but the rest of the proposed changes are in line with players and leadership in most of the alliance

1 Like

No way players are using a immense amount of res to get the best set-up emblems and this kills that aspect of the situation

And exactly that makes wars almost the same every time. Spreading emblems all over the war teams would be much more fun than holding all blems on one def.

#play2haveFun

1 Like

Tiered Loot

It works as incentive to motivate players in the direction of making the already in place algorithm work better. Seeing that the algorithm was written by SG, the assumption is they would prefer the game to move towards alliances that are more full.

Secondly, it would push those that don’t have strong feelings either way to fill alliances rather than start new ones.

6 Likes

This is incorrect.

Current system

Teammate war history 2x zero attack wars ( automatic opt out ).

Alliance war history 20x wins maximum, or 20x losses maximum

Elo system

Teammate war history 6x war attacks per war PLUS X war defends per war, from current war to first war ever attacked in.

Or a minimum of

12x adjustments per week

624x adjustments per year

Alliance war history, sum of all personal war histories for everyone CURRENTLY opted in to a war.

Or a minimum of ( for a 30x teammate alliance)

360x adjustments per week

18,720x adjustments per year

The current RAID matchmaking system has theoretically infinite penalties.

Due to math is is effectively 100x wins maximum, or 100x losses maximum

I find it interesting that SGG does not use top 5 hero power for raid matchmaking.

Please gather your thoughts in a way that is easier to follow. Nothing you have stated below that statement seems to support or not support the assertion that what actually happened in the game is incorrect. Again, regardless of the math you are doing or the NEW proposal you are making, individual war scores happened in the game and were exploited.

All please remember this post is to discuss the merits of the proposal in the OP. If you have a new proposal to discuss please create your own topic.

Thank you for understanding.

@Guvnor @littleKAF I have flagged several posts as off topic, could you please remove them?

1 Like

OK. I like it. Question is how does SG actually know how long someone has been in a particular alliance? Now with POV they have created an environment that encourages people to leave and hit a titan and return. Just solely looking at the number of days in an alliance is not accurate. Does this affect war matching at all even if no war was completed whilst the member or members were away killing titans to fill their POV?

I have seen all the tricks and poor gamesmanship that can happen and it would be nice to see some support for more fairness. Mismatching as well has been very skewed lately too. Anything to improve the system would be great. I love war, and as a former leader and current co-leader, I am frustrated by just what you mention.

Thank you for the time you have spent over the past 2 years in your data collection. Yes this is a game, but always better when we can have an equal playing field