Originally Posted by
ZygnusX1
While it may go quietly ignored by the forum, I hope the devs read this, there were a number of good ideas in the original:
the last part about data analysis is key. SC should literally have millions if not billions of data points of actual results from various offensive weights against defensive weights, and could simulate outcomes of theoretical matches statistically before rolling out a new MM or even a tweak to parameters. Is consistent fairness hard? You bet. With the number of wars being matched, even 1% bad matches would fill this thread easily.
I would like to see player offensive weight based on actual war performance. That defenseless max 11 that wipes your #1? Gets a big fat heavy weight that doesn't budge when they steamroll a TH8 for cleanup. And go ahead and take that "best" actual performance as a ratio to raw weight as a publicly-shown "skill" metric that feeds the MM. in other words - if you have max offense but have trouble clearing medium to light bases at or below your TH, you suck and should be fighting players at a similarly low skill level with similar weight. People should be able to see the weight a player can and does beat, as well as what they usually beat. Have public player and clan ladders based on war skill - the ability of players to successfully attack UP, to beat HARDER bases than their offensive troop weight. Fight your way up by beating HARD bases from highly skilled clans. The engineers can have their easy wins at the bottom of the skill ladder. Everyone will see a good war record that was not earned from a high attacking skill, but from attacking down. Good for you, you beat up the proverbial little kid on the block, don't expect the pro clubs to recruit you.
good players want to war against good bases and see their bases defend against good attackers. things like CWL were born because that was never going to consistently come out of the matchmaker.
