[GW2] Tiers

Is it that you believe it is impossible to have a League that doesn’t consist mostly of a few strong teams overmatching a large number of weaker teams? Is it not feasible for ANet to come up with a draft or seeding system that mitigates against that? I believe they can learn from the mistakes of this last season and give us a much improved version next time. Maybe you think that’s beyond them – if so you may well be right, given the litany of failures in similar circumstances since launch but I prefer to remain optimistic.

Yes. No. Even balanced tiers may be beyond possibility given the current server structure due to the spread in weight classes.

To have a good matchup, you need three servers of roughly comparable strength. (See DAoC for ways of combining the smaller servers against the bully or siphoning off the bandwagon; the comments on that post have a bit of discussion.) What are the odds that the 24 NA severs and 27 Europe servers neatly fall into 8 and 9 sets of three? Empirically, we know that did not happen, most notably in the cases of servers that bop between “crushed by 2 bigger servers” and “crushing 2 smaller servers.” It’s worse than that: the best case scenario is to have just a few servers’ WvW experience be crap. Imagine instead that we have 4 super-heavyweight servers and everyone else is neatly in groups of 3. Except for that AAA match, every single matchup is one bigger server against two smaller servers, and we risk getting tossed into a situation of all those marginal servers entering that bop cycle of crush or be crushed. It is only because the system has so many problems matching 3s that the matchups are as good as they are, because a duo in the middle can offset a foursome above.

Did that last sentence make sense? AAA ABB BCC CDD DEE, where A beats B beats C. That’s really bad. But if we have AAA ABB BCC DDD EEE, the double-C offsets the quadruple-A, giving us two good matchups at the bottom.

It may be even worse than that, because I don’t think we have just 8 weight classes in NA. If we are closer to 10 or 12, given the server spread, we are trying to find a “least bad” solution, because there are no good ones that yield anything close to AAA BBB CCC DDD EEE FFF GGG HHH. Bronze folks, what do you say? Are you closer to FFF GGG HHH or FGG HIJ JKL? I hope it’s not that bad, because those would all be crap matchups.

Unless the idea for Season Two is radically different and includes dramatic ways of balancing population advantages, the Leagues only make it worse. That was discussed at length in Weight Classes, but let’s put it simply: if we have trouble getting a EEE match, mixing up C through G and randomly drawing three will yield a worse match in almost every possible case.

As WvW is currently structured, the tiers are the least bad way of getting matches that are not utter blowouts. Putting Blackgate and Sea of Sorrows on the same battlefield is just mean. You cannot make Gold any smaller without just recreating tiers, and you cannot make it any bigger without just making victims of the top Silver servers. You cannot go smaller than tiers, and going bigger just makes the blowouts worse.

: Zubon

The suggestion of adding a Darkness Falls-style dungeon is interesting and will be explored in a later post.

3 thoughts on “[GW2] Tiers”

  1. During the League the most exciting and the most controversial match I experienced was the one in which Yak’s Bend worked first informally with and then, late in the match, formed an alliance with Ehmry Bay to defeat Stormbluff Isle, a server that at the time was considerably stronger than either of the allies standing alone.

    It was, as I say, highly controversial but as originally conceived alliances were expected in WvWvW (note the third W). Like a lot of pre-launch expectations that never came to pass. Perhaps it’s not surprising. I can affirm from that one match that holding a formal alliance of two servers together even for 36 hours is incredibly hard.

    Not only did it require communication channels to be set up and maintained between the two sets of commanders but ad hoc instructions and explanations needed to be given to the militia on a rolling boil as new casuals log in constantly and need to have the unusual circumstances explained to them before they start unwanted firefights with the allies.

    A suggestion I’ve seen made several times is to have an autonomic function that combines two servers into a single team when set thresholds are met. This would mean that unbalanced matches where one server was consistently ticking 400+ would be forced into a 2vs1 state regardless of the wishes of the players. If Blue was winning, Red and Green would find themselves literally unable to attack each other and would have no choice but to double-team Blue.

    This seems to me to remove too much autonomy from players and I wouldn’t advocate it in such a harsh form. It should, though, surely be possible to create incentives that persuade two weaker servers to work together. Perhaps giving second and third double or more WXP for all activities against first once a threshold is passed.

    I just raise this as one possibility. It seems to me that there must be many far better ways of organizing WvW than any of the versions we have seen so far. No-one wants to be the gimp that everyone beats on. I very much don’t want to be the one beating on said gimp. Everyone holds in mind a perfect image of Fair Fights and Close Matches. We all want the same thing. We just don’t know how to achieve it.

    And we aren’t paid to come with the solution. Arena Net employees are.

  2. I wonder if your seeming obsession with population and coverage isn’t running up a dead end alley. There’s -always- going to be imbalance down that front.

    Even with a Darkness Falls siphoning away the primarily PvE crowd, there is still the potential scenario that you can have a server with a larger hardcore WvW crowd determined to stay on the map and ignore the dungeon. Will they not proceed to dominate?

    How about addressing the scoring system to see if there are any other ways of producing the possibility of small wins, rather than a gradual blowout that saps morale from the less committed?

    Random idea: Let’s say we divide the day into 4 periods of 6 hours and make each display as a mini-match. Participants get a reward chest (ie. similar to dynamic events) at the end of each tiny match according to how the three servers fared for that round. You can still let the big scoreboard add up to form a victor at the end of the day (week), but perhaps the possibility of doing well for a localized 6 hour period might be encouraging to smaller server morale.

    Edge of the Mists appears to be trying to address the issue of population imbalance in a few ways – size of map strings out players and makes it harder for a poorer playing zerg to regroup, terrain and chokepoints, siege and friendly NPCs seem to be aimed toward giving the side with less players a better fighting chance.

    It is still left to be seen how successful this will turn out. The increased difficulty and perceived potential stagnation of territory gave me an initial frustrated and turned-off feeling, but I need to play it more to get a fairer take on the map. The lack of social and community structure during beta may have contributed to this – I needed the comfortable backing of my server working in unison, rather than wandering in as a lone sheep amongst wolves. I did spot a lot of potential for new tactics and clever siege locations though.

    1. Coverage doesn’t matter as much to me. The only off-hours activities that bother me are resetting the waypoints at my server’s EB keep and BL garrison. Other than wanting to keep those, I don’t care how randomized the map is when I get back. I can ignore the PPT as long as the individual fights are good (tomorrow’s topic). Population (quantity) affects that a lot more.

Comments are closed.