First, this is not a complaint thread; I’m doing fine, though I am playing more casually than last season, which is kind of a disappointment since I much prefer this style to the bunker wars.
I had three different things to say, so I figured I’d put them all in one post. Knowing how people fear words, I’ll summarize before and after.
First bit is about the weird way teams are made. Second part is a sports analogy. Third part is about the season structure.
I was here for the first night of S2, on which I just played to Emerald, then was out of town for five days. When I returned to the PvP action, it was a little uneven, but the most bizarre part was this:
Every time I got a win streak of 4 or 5 going, it would then put me on a team comprised of players from teams I had just helped to slaughter by 450. Predictably, the ensuing match was a headache of running around trying to do everything myself, sometimes resulting in a loss. I was confused. Why was this happening? Weren’t teams supposed to be built around similar mmr? Didn’t the prior embarrassment demonstrate that we don’t belong in the same group? Did the losses knock them down to my mmr? How were they higher?
Still, I rolled along (with one weird patch in late Emerald), cruised through Sapphire, and then began the trudge through the hell of soloQ Ruby. And as I watched the same players joined then opposed, match after match, with some very good and some very bad, I realized that the system’s mmr calculations are thoroughly out of whack. It’s just throwing you in a team with other people in the same pip range. If you’re queueing on a team, no sweat; you can overcome the vagaries of the draw. If you’re a top-tier player, no worries; you can trounce the competition no matter what millstones it throws around your neck.
But if you’re a regular soloQ hero, then you’re up against an uncomfortable facet of the automation: Every time you carry, the system thinks the people you carried as just as good as you are. Every time you’re dragged down, the system thinks you’re just as bad as the numbskulls who cost you. And you can’t escape them, because it’s going to keep putting you all in the matches together, dozens and dozens of which haven’t sorted out the weak from the strong, probably as a factor of low participation.
The worst part was this: In my first week (maybe two) of play, I had exactly one close, competitive match. It was on the first night, when by chance I was thrown into a game with several people I usually see in my mmr range (it was basically a 4v4, but it was fun and tight). Almost all my games since then have been lopsided victories or hairpulling frustration-fests. I’m not alone in this. I’m no pro scout, but I can spot a good or bad player in broad ranges, and I see them on the same team, over and over.
Look, Season 1’s method for matchmaking was stupid for a system about advancing through the ranks, but it made for good matches. I’m told that I will eventually return to the realm of good matches, but we’re nearly a month in (admittedly I’ve been rather casual about it) and it hasn’t happened yet. What are we aiming for here?
A lot of people have noted the parellels between GW2 Conquest PvP and basketball: They’re both five a side; all about rotations and transitions; have offensive and defensive specialists, though everyone is expected to contribute to the score; and some people will trashtalk no matter how bad they are. And one player either way can make a big difference.
You’ve heard of LeBron James. He started his career in Cleveland, where he led the previously moribund Cavs to the playoffs multiple times, including a trip to the Finals, along with back-to-back Conference 1st finishes, with a supporting cast featuring the likes of Mo Williams and Larry Hughes (who? exactly). In his last year before leaving, the Cavs went 61-21. Then he left for Miami, where he met up with other very good players. The next season (also without Ilgauskas, but c’mon) Cleveland finished 19-63, good for last place in the conference, relying on the likes of JJ Hickson and Anthony Parker, who, relative to the NBA, sucked.
James won a couple championships in Miami. In his last year there, the Heat had a rematch with San Antonio, and as much I was pulling for the Spurs, it was tough to watch. It was 1v5. Wade had tapped out, Bosh turned back into a pumpkin, and the rest of the roster had too many guys either broke down or thinking they could coast to a free ring. James returned to Cleveland and took them to the Finals the next season, while Miami languished in the bottom half of the standings and missed the playoffs.
None of us is the LeBron James of GW2 PvP. No matter what you think of the guy personally, it’s indisputable that during the aforementioned span, James was the best pro basketball player in the world. And the best player in the world could not carry all the way to glory. Granted, the NBA features the world’s top players, while Ranked Conquest features whoever happens to be sitting there at the moment, but we aren’t pros, and we can’t always carry.
Now, we can evaluate James not just through his Wins, but through comparisons of team performance with and without, as well as a heap of individual stats recorded in each game, and various derived metrics. Live people are watching.
This is not a ladder. In a ladder tournament, participants are seeded in advance of the competition; it’s a format usually limited to singles and doubles events, where the individual’s performance is the win or loss. We all started at the bottom.
Some people have protested, in response to complaints, that this is not a reward track . . . except that for the first three divisions, that’s precisely what it is. A participant just grinds and gets stuff. This is why whenever people discuss seeding for future seasons, the first question is always, “What about the rewards we’d get along the way?” So we all start at the bottom.
In a sports league, teams don’t start at the bottom. They start at 0 and 0, sure, but this is neutral: .500. Over the course of the season, a team might go above and below .500; some will drop below and never come back up, while some will go above and never fall back down. Thus, by the end of the season, all the teams are distributed across the winning curve, with the bad teams at the low end and the good teams at the high end.
By starting us all at the bottom, the GW2 League system is telling us to all start from a losing position, suggesting that the good will rise to the top and the bad will tread water and those in between will hit a wall wherever they belong. Unlike a sports league, however, some people can quit, schedules will be unequal, and solo players change teams every match.
Over time, almost everyone who keeps playing will eventually reach Ruby. This is why Ruby is hell. The current that prevents tier backsliding has carried everyone along to this point, and then we are all thrown out by the delta of division threshold into the standing water of Ruby, with all the sediment and crud collected along the way, and then the struggle to reach the sea begins. What’s the point of the first three divisions if nobody is supposed to be in them at the end except the people who gave up? They’re not really divisons, then; they’re just steps.
If a division is supposed to demonstrate how good a player is, then it makes more sense to start everyone in the middle — somewhere in Sapphire, I guess? — and have them climb or fall from there. In the end, you’ll probably find yourself matched with similarly skilled players and have a sense of where you stand. Sure, maybe some people will quit if they find themselves dropping, trying to preserve a rank they don’t actually merit, but so what? Maybe you have to play X games to keep your badge at the end of the season.
Now, I say “probably” above because this still doesn’t correct one issue: If your teammates are chosen randomly, then there are portions of the population that will get stuck on good or bad streaks and find themselves not where they belong, because excepting outliers, one player’s capacity to absolutely affect the outcome of the game is essentially nil. If the matchmaking worked, then eventually you’d, in theory, find yourself on a team with 9 people who were equal and you’d either win or lose if you were better or worse, but the matchmaking doesn’t always work.
In sum: This season’s matchmaking, for a large number of players, is a crapshoot. It seems to be taking everyone from a pip range and throwing them into a game, regardless of how good or bad the players are.
One player’s skill cannot always determine the outcome. This system works for teams, I guess, but not for soloQ.
If it’s a League, why are we all grinding up when we should be starting at 0 and gaining or falling based on performance?
If the League is all about promoting team play, then either: 1. We need a separate queue for teams; or 2. We need to acknowledge that the great majority of us are there only to fill out the participation roster and our experience doesn’t matter.