I admit that I don’t have much of a competitive bone, so I need someone else to help me understand the meaning of this elusive term, “good fights.” Exactly what would be defined as one?
I characterize “a good fight” as one with roughly balanced forces, where “balance” mixes quality and quantity. This applies to the two levels at which WvW takes place.
WvW competition happens at two levels: individual fights and overall score (“PPT,” points per turn). Individual players are rewarded according to individual fights, where you get points for kills and captures. (There are rewards for defending, but the system is weighted towards rewarding offense over defense. More on the virtues of defending below.) Servers are rewarded according to how many objectives they control over time (plus points for stomps). This post addresses the server level (“a good matchup”), and tomorrow we discuss the individual fights (“a good fight”).
At the PPT level, population and coverage are key. How many fights you win is more important than your winning percentage. A server with a small, elite WvW population can win every single fight but still lose the week because that small force cannot be everywhere. Having a large population lets you attack and defend multiple points across multiple maps. More bodies means more caps and longer defenses.
Quality also matters, but probably less than quantity for PPT. All things being equal, you prefer having more skilled and experienced players who communicate well. As population shrinks, quality matters more; as population grows, as Stalin said, “quantity has a quality all its own.”
Coverage is the other element, particularly as server population shrinks. Consider two servers with equal WvW forces, where one is concentrated in the American time zones and the other is spread evenly over the globe. The American server will dominate American prime time, appearing to have twice the population or more; the global server will lose for four to six hours a day and then sweep the map once the Americans start going to bed. Controlling the map for 18 hours yields more points than controlling it during prime time.
Note that off-hours flips affect prime time play. When I went to bed Sunday night, our server controlled Stonemist Castle and had it fully upgraded. If there is one hour in which we have no meaningful coverage, SMC will flip and lose all its upgrades. It is enormously frustrating to win the battle but lose the war because you have a job and need to sleep. This can lead to not bothering to upgrade or to WvW at all. Off-hours ninjas can demoralize opposing forces.
Quantity, quality, and coverage are the three major factors I see in what makes a good matchup. Theoretically, over time, most servers will settle into balanced matchups based on that because of how scores come out. That settling out time will involve painful mismatches as one server dominates or gets utterly crushed due to mismatch. The first weeks after launch were like that, and Season One has caused the same on a smaller scale as transfers shifted WvW populations in ways that will take weeks or months to settle out.
Note that WvW population matters; server population matters only as a pool of potential participants. The dedicated WvW playerbase will be a fraction of the overall population, so small shifts like the transfer of a dedicated WvW guild will have a disproportionate effect on WvW population. It is the more dedicated WvWers who are likely to transfer on that basis, increasing the effect. There is also a feedback loop whereby winning servers get a higher percentage of the overall population participating (“bandwagon” or “fairweather”). There is a terminal point on that feedback loop as there is a WvW population cap and as the average player skill drops when casual WvWers keep dedicated WvWers in the queue (“the bandwagon before the horse”?).
To demonstrate the outsized effect of small WvW population shifts, consider Fort Aspenwood. Before Season One, Fort Aspenwood was rated as potentially competitive in Silver League. Hardcore WvW guilds transferred to Fort Aspenwood. After Season One, Fort Aspenwood is winning Tier 2 in Gold. 40 or 50 people is just a blip on the overall server population, but if most of them are spending most of their time in WvW as a coordinated group, that can flip an entire map. The effect is smaller with existing Gold population levels, but it’s hard to argue with success.