Warning... Wall of text! (Man this seemed shorter when I was typing it)
I want to preface this by saying I have not read every comment in this thread and so sorry to anyone who may have suggested this before me. As well, you may read this as somewhat negative but let me assure you this is simply my passion as I want nothing more than for the game to succeed. Without further adieu, allow me to proceed.
I feel that limiting the match maker in ways such as 4 man groups or 1/1/1/1 is only making the job of putting a match together needlessly complicated. Firstly, preventing other interested folks from playing in a group of friends larger than 4 is just not acceptable in my books. This is just too big a negative factor to even consider.
Secondly, I cannot see how combinations like Locust/Cicada/Dragon/Awesome is well balanced against Firestarter/Shadowhawk/Thunderbolt/Atlas. I'm not saying matches are always this disparate in terms of tonnage, but the point is that there are many factors as to how "good" a Mech is that is not taken into consideration by the current match maker. There are obvious benefits to be had of a Firestarter that a Locust will just never have.
This brings me to what I believe is the ultimate fix for the match maker. I believe that a Battle Value "table top"-like system would make life much easier for the match maker. If values were assigned to each weapon, system and Mech, the system simply has to gather a group of players with similar BV to each other.
My specific thoughts on how the system works are as follows:
BV of a weapon is calculated based on all statistics (Damage, heat, duration, cooldown, tonnage, slots, ammo/t etc).
BV of a Mech is calculated based on the total BV value of weapons, systems, engine, armor, tonnage, etc.
BV of a player is calculated based on the total BV value of the selected Mech, PSR, and whether or not they are in a group (BV modifier for those in a group to represent the significant coordination)
The matchmaker now takes the BV of all available players and groups and tries to fit them into a match of near even BV value (or a steadily increased margin as time passes).
The benefit of this system is that all factors are considered to attempt to produce the most enjoyable game possible.
I'm working on some specific examples with appropriate BV calculation algorithms to further promote the ease of use of a system like this but I hope those that read this can understand what I'm trying to explain.
Especially to the Devs, this system will make life so much easier. When using the algorithm, if you make a change to weapon damage or range, the BV is automatically recalculated based on the change. No further adjustments needed. But! If you need to, just have a modifier field to compensate for any discrepancies in the perceived effectiveness of the weapon. Just as the BV modifier for those in a group to show the additional effect of teamwork in a group will allow those that are matched against them to have the benefit of either a) more powerful Mechs
skilled players (based on PSR BV modifier) or c) another group (based on group BV modifier).
In my mind this is so simple for me to see the benefits. Perhaps I am blinded by this so please let me know what you think and whether or not my logic is flawed. I am sure there are those of you who will not waste time in doing so
Apologies for the wall of text but thank you very much for taking the time to read my rant. And thank you Russ and crew for taking the time to discuss this with your community!
Edited by FoXabre, 08 September 2015 - 05:07 PM.