How To Make War Matching Balanced - Optional Players

[ClanItBe]

Our match up would be better if match making is timed e.g:

• starts every hour e.g 1pm, 2pm, 3pm etc....
• starts every 2 hours e.g 2pm, 4pm, 6pm etc....

that way we have a much bigger pools of clans to match against. also better timing when war starts/ends.


also they should dump all this 9.5, defenceless bases nonsense. if your town hall 10 with no infernos your base value should be maxed out TH9. if you have no defense and your TH9 you will be considered maxed out TH8. so the formula is basically any weak base = -1TH but maxed out.

The starting on the hours solution would definitely make for some more even matches for some but would also leave some others waiting longer if they can't match from a given period. I saw another suggestion that was not bad which was a button "Ill wait however long" for a good match.

I definitely agree with the rushed bases in war, seems to throw off the algorithm. If someone finds the right balance of rushed base and advanced troops I am sure their war team would be unstoppable. Really, I think SC should make it impossible to upgrade your town hall until certain things are done so we can just get rid of rushed bases.

[ClanItBe]

Interesting idea.

As for the algorithm, that is likely a complete rewrite of the code. some methods and classes can be reused, but for the most part execution would require a complete rewrite or major modification of the existing code. a few days to a few weeks depending on level of testing, lines required, etc.

I won't pretend to know SC code but as a software engineer I can make a few assumptions.

1. When war is started all the bases are scored. An attack and defense score are generated along with stats like number of townhalls at each lever, hero levels, xbow and inferno numbers. Those numbers are then stuck into a "War Matching" table.

All of that code can be reused, it just needs a foreach() loop wrapped around it so each variation of optional players are inserted into the table.

2. The matching algorithm runs over the table looking for even matches.

This code could also be reused but slightly modified to tighten the matching criteria, likely just a few configuration changes.

SO....

The largest changes are in the interface to add the Optional flag when editing the war roster and selecting it as a user. Showing the matches and allowing custom matches to be started would also be a fairly large effort but that could come later. I am sure a few weeks of development and testing could yield the basics of the Optional war player which would be fairly cheap price IMO.

[OMCKG]

this is very complicated. i woul support, though SC probably won't do it

[ClanItBe]

this is very complicated. i woul support, though SC probably won't do it

It would be a fairly large change but if it makes war better I am sure they can make it happen. They bring in about $1,000,000 per day so I doubt they judge changes on complexity and instead judge on will it make more money. I think more people going to war will make more money because more people will need to gem their heros so they can use them in war!

[ProfChaos]

Odd.

I'm not sure I've come across a scenario where a war was so unbalanced or unfair that I felt the need to ask for something this huge & complicated.

Also never had a search go for hours, let alone a single hour. Would making wars with so many varied numbers of mandatory participants , ie anything other than multiples of 5, make things faster? Seems like that'd be even harder and take much longer to search.

And are you sure that this would still be totally unexceptionable?

Other than mods & other such forms of ♥♥♥♥♥ing all wars are generally fair in my experience. Sometimes you get a beat down and learn. Other times you hand one out. All are in good fun.

[ClanItBe]

Odd. I'm not sure I've come across a scenario where a war was so unbalanced or unfair that I felt the need to ask for something this huge & complicated.

You must be in a clan of lucky people. A quick forum search will reveal a bunch of people unhappy with the uneven matches they get. This logic might sound crazy and complicated but it is not, just about every programming language has a method to permute combinations.

Also never had a search go for hours, let alone a single hour. Would making wars with so many varied numbers of mandatory participants , ie anything other than multiples of 5, make things faster? Seems like that'd be even harder and take much longer to search.

Again, you are lucky. Almost every war we are in takes an hour to match. The reason optional players will make it faster vs multiples of 5 is the extra people are OPTIONAL. So, you have that same 10 vs 10 but, if there is a good 13vs13 than you can do that too.