First and foremost: This is a wall. But it’s deeply Analytical.
I really do appreciate the communication and intrigue into trying to solve the scoring and stale matchup problems. I’m also not really sure if it really solves the core problems with why we see runaway scores, though.
Quite frankly, it doesn’t really reflect reality that much, either.
Fundamentals to why we have Runaway matchups:
- Timezone coverage/player distribution
- Organized guilds
The timezone coverage issue doesn’t end up changing. Certain groups will still contribute disproportionately to a given server’s score. Unless you force servers to split up (WHICH WOULD BE HORRIBLE. DO NOT DO THIS.), you’re still going to see coverage imbalance and people stacking for off-hours. Any server that ends up splitting will just be at a bigger deficit compared to those that don’t, and the only incentive to splitting is re-stacking to make a stronger low-tier server move up ranks. The problem is this already happens and has happened, and has lead to what are still stagnant matchups.
Thinning out players across servers won’t do much for the latter, either. Even if particularly strong servers destack for a period of time, guilds, which are the driving force of winning WvW, will eventually cull out other servers, and people will transfer and re-stack to follow the winners, just like now. Then you’ll just have more servers and more links being needed to be performed, which while it works mathematically from a strictly player count perspective, does not work from an organizational/performance perspective. The re-stacked or newly-stacked servers will be led by huge organized guilds with potentially very little link influence that are dominant in WvW. Its opposition will feature similar numbers, but none of the groups will be organized.
The way it currently is, it actually makes more sense from a balanced-matchup perspective that the team featuring the sides least-likely to win and facing the biggest community-merging issues (Which of the closely-ranked servers’ Teamspeak should be used?/Which server really deserves to be the host?/other misc community problems), which further reduce the likelihood of winning.
The big issue with WvW right now is that the problems with the format are so deeply nested in other aspects of the game and are so entrenched in the nature of the format itself that fixing them at this point requires very surgical and complex procedures.
The model you present is simplified to convey a point, but take a good look at the actual numbers between each server. I doubt that between any two adjacently-ranked servers (granted this metric is very hard to gather now with links) the gap is as substantial as 20%. It’s likely single-percentage within tiers, maybe a little more at each tier step. The combinatorics don’t lie in that 27 servers is already plenty to make a ton of combinations. I think doubling-up would probably end up just confusing the system more by putting in extra bloat of pairings that look good on paper but don’t reflect what’s actually happening in-game.
To get a better understanding of the reality of WvW versus what’s happening mathematically, I highly suggest you try an experiment. I know it’s not good practice to meddle in the affairs of players for the most part, but I think this is a necessary evil that would benefit everyone greatly, especially if you announce it as an experiment to learn more about WvW.
Make an announcement you’ll be leading a given T1 server during prime hours. Get some pugs, don’t use some kind of communication media, and attack the opposing servers’ keeps or try and fight them with what you have. This organization and proficiency of commanding on that scale is perhaps what you might get from those mid-tier servers and large groups of hodge-podge players being forced/matched up together. You should still have roughly even numbers, potentially moreso if you hype it up. No disrespect, but you’re likely going to not get very far. Groups who have dedicated hundreds/thousands of hours to being the best of the best are going to continue to be. Evening the playing field for more “fair” player distribution doesn’t really mean much when the organization is so scattered, and communities so fragmented. We’re already seeing big complaints about community fragmentation. I think that’s the bigger priority for getting servers to be more stable; on some level, they need integrity and identity back. This keeps people in cohesive units between matchups and would enable servers to build up with what they have, rather than what they depend on from the higher-tier ones, to then immediately be shut down on the next re-link. Server chat and a two-to-four-letter code title for server membership identification would go a long way, I think, in helping revitalize failing communities.