Guild Wars 2

You can’t even get Social Security and you think you’re old? Bah. As your elder, let me give you some advice junior. For PVE, the easiest way to go is necro minion master, especially after you get the Reaper skill “rise”. If you like playing El, then staff is fine, with fire, water, and either earth or arcane, depending upon how much you like to dodge to use evasive arcane. After you get tempest, I like the auramancer build (checkout metabattle for some builds) for support. In PVE, I like glyph of storms, arcane wave, and feel the burn if you are a tempest or cleansing fire if not, all for area of effect damage. Arcane wave is also a blast finisher for your water fields from staff 3 & 5 when in water. I recommend celestial armor over the zerkers, and runes of the traveler for the speed boost. Also, I like reassigning the dodge key to the space bar, which is much faster for me to hit.

As optimization problems go, this doesn’t appear to be very hard, especially if you have a month to do the calculations, but you never know until you see the data and try a few solutions. I’m sure there are plenty of smart players with the required skills that would find it a fun project if some test data were made available.

On the other hand, if Anet really has no data on the number of WvW players by server and time, then yes, we can’t create a reasonable solution.

I’m just saying if the data became available, I’d be happy to coordinate the effort and try my hand at a solution.

First, this is a serious post and not intended as sarcasm. I’ve observed that the grouping of servers in WvW has produced some grossly imbalanced matchups, where for a whole month some players have to suffer being routinely outnumbered and outscored 2 to 1. To me, the most fun in WvW is when there are even matchups and the weekly winner is in doubt right up to reset.

My view is that the matchups should be created so that the three teams have nearly the same number of active players at all times over the match. Obviously, this can’t be perfectly achieved, which is where optimization and time series analysis come in. Selecting the server groupings for even matchups falls with these disciplines which are not available at all software development companies.

I suggest we players offer to help Anet come up with a practical algorithm for even WvW matchup selection. If Anet will tell us what data they have about WvW player activity (ideally, number of players in WvW for each server at each tick), I’m sure there are people here willing to help (I’ll be the first to volunteer).