Clan War Matchmaking Improvements

[WerewolfXS]

I'm not sure anyone is really bothered by engineers playing the game differently than others. What most are bothered by is that the way engineers play gives them an unfair advantage against people who choose not to when an engineer clan matches a non engineer clan. That makes the game less enjoyable for those who dont engineer.

Ok, u are right most of times, a clan with engineered accounts will have a advantage, but u cant say its a unfair advantage, simple because matchemaker is eaqual to everyone....
I would preffer that everybase base, dosent matter how bad it was, could have fun at war without worries, but this isnt possible....

[PurpleDeathStar]

Match offense capabilities to defense capabilities. If you have a warden and lvl3 bowlers, you have to eat an eagle for war. Shouldn't be a problem right? Not looking for an advantage, just want to play your way, that would be fair.

[louisspawn]

This is one of the reasons why the whole engineering got so popular: if you have a lvl5 warden you we often fail even to two star a well developed th11. The odds for a th11 3 star are even worse - the defensive strength on th11 is just too much for those who don't have above average skill.
When you look at the CWL recaps even the highly skilled players of Clash prefer to use th11 vs th10 for reliable 3 stars.

With that in mind we can understand why engineers prefer matchups where they don't have to hit the highest townhalls - at least this way you don't have to be a Powerbang to 3 star a top base on the war map.

[Darian[Supercell]]

Just to assuage any fears, again, nothing. has. changed. in the matchmaking algorithm. There has been no stealth change, no minute modifications, no adjustments. And just to restate it, there will be no adjustments to it until the update is released. I don't have a release date pinned down yet as it's still in the testing stages and it's something QA is analyzing at the moment.

If any changes had been implemented, I hope you trust me enough to say I would let you know if anything had changed. Furthermore, implementing stealth changes to the system before we implement more changes would not only be counterproductive but it would probably be largely overwritten by the upcoming changes as well.

Any perceived differences over the last week or so are most likely and highly probable to be statistical anomalies. Given the number of Wars that go on, the number of Clans active in Clan Wars, statistically it's very possible that odd matches will happen over a set period of time.

So I say again, rest assured, I will let you know when the changes are going to happen and until they happen no changes will be made or have been made.

Carry on :-)

[Superfinch30]

Thank you Darian for the clarification.

On that note though..... minor stealth tweaks do apparently get made relatively often.
Are we to be made aware of these in future?
Purely from a standpoint of curiosity, and being able to provide feedback.

[RayTyson]

Thank you darian for clarifying.

i figured that nothing was going to be changed since it would be for nothing since the update will already bring this new change.

[Ridhorz7]

We win 10 strike now

[OnyxDS]

Thank you Darian for the clarification.

On that note though..... minor stealth tweaks do apparently get made relatively often.
Are we to be made aware of these in future?
Purely from a standpoint of curiosity, and being able to provide feedback.

This seems to be a key point. SC can measure the random mismatches - in most cases the server won't think the clans are entirely equal, and SC can monitor that. But the mismatches that people are complaining about don't seem to be something that is objectively measurable by SC - when a certain type of base gets a consistent advantage that's not random and will usually be because the MM thinks it is equal to its opponents and players disagree. Which means the only real evaluation of whether it's working is whether players think the matches are OK and it is working. And players can't really comment on whether changes have helped or hindered if they don't know there were any changes.

(PS and checking for statistical correlations between upgrade types and scores/win rates on the server probably doesn't help detect MM advantages since there look to be very strong correlations between some upgrades and player skill - so you'd expect certain upgrade patterns to win more often even with a perfect MM)

[JFSoul]

This seems to be a key point. SC can measure the random mismatches - in most cases the server won't think the clans are entirely equal, and SC can monitor that. But the mismatches that people are complaining about don't seem to be something that is objectively measurable by SC - when a certain type of base gets a consistent advantage that's not random and will usually be because the MM thinks it is equal to its opponents and players disagree. Which means the only real evaluation of whether it's working is whether players think the matches are OK and it is working. And players can't really comment on whether changes have helped or hindered if they don't know there were any changes.

(PS and checking for statistical correlations between upgrade types and scores/win rates on the server probably doesn't help detect MM advantages since there look to be very strong correlations between some upgrades and player skill - so you'd expect certain upgrade patterns to win more often even with a perfect MM)

Excellent point and observation. I've stated this before when this topic comes up. Often we see threads that point out mismatches--and in some cases they do appear to be a mismatch. However the underdog wins--so when SC looks at the data, was it really a mismatch? It's rhetorical question, but I'm pointing out how difficult it might be to rely on the data when it comes to lopsided bases.

[United06]

Thank you Darian for the clarification.
On that note though..... minor stealth tweaks do apparently get made relatively often.
Are we to be made aware of these in future?
Purely from a standpoint of curiosity, and being able to provide feedback.

Like the loon change I guess .. it was entirely a stealth tweak ..