Guild Wars 2

Laserbolt.6731:

This comes up regularly.

The dynamics of the game make it really difficult to put an exact number on a player’s contribution to the win. The computer cannot record all the fight dynamics and interactions and cooperations.

The programming does not exist to record things like who helped who when a point was taken. For example, the person who actually stood on the point to “flip it” may have contributed almost nothing to making the flip possible. Two other people may have killed the enemy who was holding it. And all were in the vicinity.

I agree that it’s not easy, but I disagree that you can’t find a solution. I liked the approached detailed here:
https://forum-en.gw2archive.eu/forum/game/pvp/New-Scoreboard-K-B-C-N-R/

Basically, as long as the point is active (has the white swords), you can accumulate points for defending, neutralizing, or capturing it. Instead of basing it purely on the event, track who is in the vicinity while it’s active and either on the point or fighting players near the point. If it eventually flips, you get credit as long as you have some percentage of time near, a kill, etc.

Not perfect, but better than the current.

Justin ODell.9517:

You could do a linear interpolation between the solo MMR and team MMR based on the number of players in the party.

I haven’t thought about it since then, so it would be good to get people’s impression.

I don’t think linear is the way to go. A group of two doesn’t make a huge difference compared to solo players; it’s rare to see them as a specialized duo which can control a map. Even a group of three isn’t that overwhelming. However, a group of 4 has enough players to coordinate for both damage and area control.

Tao.5096:

Elite player spotted.

Bad player spotted.

Maybe I am a little elitist. But the truth is that some build choices just aren’t that good as others. Poor build choices go hand-in-hand with poor gameplay decisions. Minion master necros are supposed to guard points. I rarely see one do that; they fight in the open. Matchmaking should reflect the thought put into builds and roles and place people of similar skill levels together. If you want to make questionable decisions, you should be matched with players of similarly average skill.

Dixo.3876:

Are there plans for transparency. I’m certainty feel like I’m getting the short end of the stick tonight.

I can’t find the post, but Justin O Dell did say that they plan to add some in the future. I think a player mentioned a web-based detailed match history as an example.

Pretty detailed write-up here:
http://wiki.guildwars2.com/wiki/Matchmaking

Not sure if it’s been updated with current weights.

Okay, myself and my friends (usually duo or triple queue) have been exceedingly frustrated by the matchmaking the past couple days. It’s not really our opponents; it’s the teammates we keep getting.

I really have no clue where our rating is at. We’ve played against the top NA players multiple times. At other times, we have people on our team who are way below our level: minion master necros, warriors using signet of might and fury and not healing signet; the rally bots who run into a point outnumbered and die in seconds.

I don’t mind losing to better or more organized players. But when the loss is because we can’t carry the bad players hard enough, it makes me lose faith in this game. Not to mention the farce that is the leaderboard.

So I have a technical question. GW2 uses glicko2 for rating. But all the math for glicko2 is geared toward individuals or fixed teams against each other. How does the rating, rating deviation, and volatility get applied to individuals in a dynamic team? e.g. individual’s rating vs. average enemy team rating; individual’s rating vs each individual rating on the enemy team, etc.

What I’m getting at is the following: How can you accurately score contribution of an individual to the team? If Team A has a large deviation in individual rating and loses to a Team B of similar team rating, do the players on Team A with higher individual ratings lose more rating? If yes, should they lose more rating even if the loss was caused by other team members?

Further, what is the expectation of a team like Team A? Is contribution assumed to be a linear function of rating? If so, should it be? In my experience, a significantly lesser skilled player can cause more of a snowball effect than a very good player.

You could always use the “Practice” option where most people don’t care how bad you are.

Patchi.9061:

IP is one of the base reasons Engis are viable at the moment. Remove IP and what is left for engis to deal damage? You make it sound that you will eat 2500 damage over 10 seconds neglecting all sources of regen, condi clear, and whatnot.

Your example and explanation shows that you don’t understand game mechanics and why IP is too powerful.

If you ever look at a death log after fighting an engineer (cele or full condi), you’ll find that burning is by far the largest source of damage. Not bleed. Not poison. You even said it. Very few engineers run flamethrower (incendiary ammo), off-hand pistol (blow torch), bomb kit (fire bomb), or any other significant source of burning or fire fields. So a large chunk of the damage from an engineer in a sustained fight is from IP.

Why does it do so much damage? That lies in the condition removal mechanics combined with the quantity of different conditions which engineers apply. With common engineer builds, a single use of poison grenades applies 3 conditions (Poison from the base skill, bleeding from Shrapnel, Vulnerability from Steel-Packed Powder) and can proc a 4th in IP. Very few condition removal skills can remove four conditions at once. Those which do are usually on long cooldowns – much longer than the 10 second ICD of IP. So you’ll mostly see abilities which remove 1 or 2 conditions. No one is quite sure what the rules for condition removal are, but bleeding is usually the first condition removed and burning one of the last. So unless it’s a group fight with lots of group condition removal, IP will often tick for its full duration.

Let me frame this differently. Let’s give a profession (don’t care which) a trait that says: Your next attack triggers <effect name>. <effect name> has a 85% chance to deal 3000-4000 damage (cannot crit). This attack cannot be dodged, blocked, or evaded. 15sec ICD. Do you think that’s fun and leads to better gameplay? Well that example is essentially what IP is.

It doesn’t matter if IP is a key element to engineer builds; it’s not conducive to good gameplay and needs to change. If engineers will be too weak with a drastically reduced IP, then other aspects of the profession can be buffed to compensate. But those aspects should allow for better counter-play.

I think the problem is more with the shoutcaster presentation of the game than the game itself.

Part of the reason conquest was chosen was that it’s more obvious to a spectator who is ahead or who has the momentum at any given point. And that’s true when you look at the map level. You can see the score, who has more nodes, which are in conflict, and who’s pushing their opponents back.

However, most shoutcasting of GW2 focuses on individual battles where it’s very hard to tell what is going on or who has the advantage. The scope of the game as a whole is lost. You don’t see the thief sneaking away for a decap or moving up to a teamfight in anticipation of picking someone off. You can’t easily tell where the DPS players are and where the bunkers are.

I think DPS had the best shoutcasting approach to GW2. They had a split screen and each commentator would look at a section of the map or a node. When the action picked up at one place or there was a key move, it was conveyed immediately. They could switch between screens to focus on that key move. None of those “oh hey look, they decapped this other node while we were watching this fight”. And that decap was just as important as the clutter of particle effects we were watching for the last 2 minutes.

Patchi.9061:

IP isn’t the reason engi’s are OP

If you think this, you should stop proposing balance changes.

Incendiary Powder is incredibly broken and should have been nerfed a year ago. With 6 points in Explosives, it’s a baseline 5 second burn which cannot be avoided and with all the cover conditions that engineers have, it’s near impossible to remove. With sigil of Intelligence and grenades, it’s reliably applied close to the internal cooldown. You end up eating the full duration, which even on a celestial engi after the nerfs is still ticking for almost 500 damage/second. That’s 2500 unavoidable damage for no effort on the part of the engineer. Every ~10 seconds.

You say Air and Fire are too good, yet totally ignore IP as being a problem. Don’t even say it’s the fault of runes because it’s too powerful on celestial rifle engi and they don’t use + condi duration runes.

Laserbolt.6731:

So how do do you determine when a player played well on a weak team?

Here is a pretty good idea for a scoreboard change which would better score intelligent play:
https://forum-en.gw2archive.eu/forum/game/pvp/New-Scoreboard-K-B-C-N-R/first

If the scoreboard reflects intelligent play, then it can be used as a metric for how well a player performed relative to his or her team.

Patchi.9061:

Hence adding voice comms in game would fix the issue~

Counter argument: Blizzard added voice comm to WoW. It delayed content by months. No one used it. I think they eventually removed it. I don’t think ANet has time to waste on an in-game voice comm feature.

Communication isn’t the issue so much as players being able to learn roles and smart strategies from gameplay. All too often you get the guy who goes far node at the start, can’t get there before it caps, and has no sustain or decap ability. He ends up dying in an easy +1 and causes the match to snowball for several minutes. But instead of realizing what they did is dumb, they blame the rest of the team.

Good

• Addressed the true problem of might stacking instead of giving into the uninformed with a celestial amulet nerf.

Bad

• I was really expecting a nerf to Incendiary Powder. This trait is way too strong for being impossible to avoid. Cutting one second off would be an appropriate starting point.

Not Sure

• The patch is small and shaving some strong things while slightly improving weak things. If this happens every 1-2 months, it’s good. If this is all we see until summer, it’s really bad.
• Sigil of Torment: Adding another condition to engineer to cover the multitude of conditions they already have makes their condi pressure even stronger.
• Rune of Nightmare: Lower duration is much needed, but lower cooldown may make the seemingly random aspect of it even worse. I’d rather the fear just be changed to something that isn’t a CC.

If the OP’s question means the celestial meta: No, not in the least. Mesmers have little influence on the meta. Thief does influence the meta because of unrivaled map mobility, but it didn’t cause the rise and dominance of celestial builds.

The celestial meta is professions with both power damage and damaging condition application who swap weapons often and have overly powerful skills. This was caused by the following:

• Un-nerfing Elementalist Signet of Restoration completely, even though it had been cut in half in PvP when d/d was too powerful in the past.
• Various buffs to d/d skills (evade, additional blast finisher, etc)
• Engineer Incendiary Powder has been in need of a nerf for over a year.
• Not properly balancing sigils when the shared cooldown was removed. Sigil of Battle being the most obvious example.
• Rigid stat choices on amulets cause most profession builds to be either too tanky and lacking in damage or too squishy as DPS.

Only a few builds can make celestial work too well. They all have the following in common:

• Have access to multiple damaging conditions which can be applied with little effort, especially burning. Combustive Shot, Incendiary Powder, and Drake’s Breath + Ring of Fire for burning.
• Trigger swap sigils often as a natural part of play. For most professions, a weapon swap isn’t something they should always be doing on cooldown, limiting the power of swap sigils.
• Stack might naturally with abilities or blast finishers or heavily use their power trait line to offset the low power values on the celestial amulet.

The hack problem could be mostly addressed by requiring the lord to be dead in order to take it.

But that still leaves the problem of “the winning server wins more”. With the original orb mechanic, you had a statistical advantage over opponents. The more orbs you had, the bigger that advantage. It was also hard to get an orb out of a keep. So when you have a slightly stronger server made stronger by the orbs and you have to beat them to get it back, it just snowballs.

Um no.

Elementalist, Engineer, and to a lesser extend warrior are so common because they can do everything as well or better than other professions and builds which only do some of that. They’re just too good. And thief is there because it can move around the map and burst much better than anyone else.

I think I’ve seen one time where Courtyard pop up and it had more than 5 votes. Most of the time it gets 1 or 0. So no, it’s not popular.

Justin ODell.9517:

I’ve been looking at the data, and by my estimate it disagrees with what your saying. Not only are wide differences in MMR rare, we more accurately predict the outcome the wider they become.

That’s not good evidence that the system is working. If a bunch of people are around the same MMR, but some should be higher and others lower, the outcome over many iterations still ends up looking correct because it averages to 50% ratios. Although the accuracy of predictions with wider MMR ranges seems to imply that MMRs are correct, it doesn’t. As long as you have some people higher or lower towards the directions they should be, it helps offset the people with incorrect MMRs when you lump them into teams. In theory, that should eventually settle, but if you keep injecting new players, it stays chaotic.

I have to agree with the feeling that MMR calculation isn’t correct or there are many outliers. While I see many opponents and teammates that I would say are roughly around my skill level, I often that one really bad player on either team that is nowhere close. Problem is, that one outlier has a huge impact on how even the match-up feels.

Do you mean rating (as in match-making rating) or leaderboard rank?

Leaderboard rank doesn’t directly factor in your match-making rating at all. No matter what your rating, your ladder point potential per game is the same. As a result, the more you play, the more ladder points you get. People who get a disproportionate number of blowouts or have a lower match-making rating than they should get slightly better returns on ladder points.

Switching to a team ladder rank won’t change anything if this broken system is still used. The broken system is the problem, not rating individuals vs. teams.

GW2 chose their own system because those other systems aren’t perfect either and have different matchmaking rules.

World of Warcraft requires players to set rosters, which have ratings, and you can only be on one roster at a type per game type. GW2 rates players and doesn’t require fixed rosters. WoW had problems with its matchmaking and still does. Initially they used Elo with a maximum change of 32 and starting rating of 1500. That didn’t work out well and they switched to a system where everyone starts at a rating of 0, but has accelerated gain early on. The big problem that WoW still has is top players sitting on ratings after they reach a certain level. This makes it extremely difficult for anyone getting a late start in the season to pass that threshold.

Starcraft2 is primarily geared toward 1v1 play (though it does support teams). With 1v1, you have a much larger matchmaking pool, which makes it easier to match players of similar skill. It also had issues with league promotion early on.

Hey, that sounds exactly like what ANet is already doing! The punishment for desertion is just a little different (ANet prevents ranked play for a duration, Blizzard forces one round of unranked play).

It’s easy to state how a system should work conceptually. Getting it to function that way in reality is the hard part.

Did you hit “leave queue” after getting to the “searching for match” phase and then re-enter the queue? Currently “leave queue” doesn’t work after your enter the search phase and doing this gets you double queued.

You don’t want full customization. You’ll get things like nomad and dire. You want multiple pieces of jewelry like amulet and 2 rings in the existing combinations so that you can mix and match.

Errant Venture.9371:

Most of us want a game where gear plays as little impact on success as possible.

Except with the current amulet system, gear has a huge impact because it limits which builds can be viable. If you’re an elementalist or guardian, good luck running an amulet with no vitality. There’s no option to get a little bit of toughness which would make DPS builds not die nearly as fast.

Might not be able to do that easily.

Alternate suggestion: Do win by profession type (soldier, adventurer, scholar) rather than individual profession.

Warlord of Chaos.7845:

If Grouch thinks engis need a buff then this “New and solid game plan for the year” (For balance) has me very worried lol.

Engineers need nerfed and buffed. There are a few things in the meta builds which are too strong, but if you take those away, engineer will be a little weak.

You could tell it was fake because some of the changes were ripped directly from the forums.

No:

https://forum-en.gw2archive.eu/forum/game/pvp/Leaked-Upcoming-Balance-Changes-not-sure

Simon.3794:

They really need to nerf power scaling on engi skills, because engis dont might stack. and still crits for a ton.

Power scaling on engineer weapons is fine; it’s too weak if anything. The typical celestial rifle engi build has 1840 power before any might stacks because they put 6 points in their power trait line yielding +300 power. Throw in 6 stacks of might and they’re over 2000 power; 9 stacks and they’re over 2100, which is around what you’d see on a zerker profession.

SobeSoul.6910:

Exedore.6320:

Saiyan.1704:

Sigil of battle isn’t overpowered.

No, an almost 15% increase in damage from one sigil is perfectly balanced…

You do realize there’s other sigils which do this right?

Like Air? Maybe Doom? Those need toned down too.

Saiyan.1704:

Sigil of battle isn’t overpowered.

No, an almost 15% increase in damage from one sigil is perfectly balanced…

I’m glad OP isn’t a developer. This list is just silly. Adding a UI element doesn’t get rid of passive play. No nerfs to Sigil of Battle or other overpowered sigils.

Worth watching at least for the math that shows how incredibly powerful Sigil of Battle is. I agree that Sigil of Battle on other sigils need nerfed. However, I disagree with the approach to nerfs:

First, I think reducing the duration on Sigil of Battle is a better choice than the number of stacks. A lot of its power comes from duration increases and the lower the base, the weaker the increase becomes compared to other choices in runes and sigils.

Second, celestial amulet having more stats than other amulets is by design and should remain that way. Tweak other broken mechanics first. The celestial combo has the same ratios in PvE/WvW as PvP, yet you don’t see anyone going full celestial there. Celestial only seems so powerful because the other amulet choices are too extreme; there’s no other middle ground choice.

Third, don’t just look at on-swap sigils. Air and Fire are incredibly powerful as well, even when not paired together.

This is by design. No one understands the reasoning.

There should be

• A window to allow you to reconnect to the game, as long as it’s with the same character you were on before
• If you DC and the game ends within the window and your team wins, the game isn’t counted compared to getting deserter score.

Hate to rain on the parade, but solo queue was just as bad as the combined queue. You won in solo queue by having less completely terrible players than the other team. And that still happens in the combined queue. The only difference is that people try to 1v1 less; if you try to do it, you’ll most likely lose.

I finally got it to go through at a real speed by closing and re-opening the client if it was at 30KB/sec or below. Took 20-30 tries before I got a good connection.

See this thread (lots of people with the same problem):
https://forum-en.gw2archive.eu/forum/support/support/Start-up-Client-stuck-at-0kb-s/

Same problem since Tuesday patch. If I close and restart the client a lot it sometimes starts downloading at several hundred KB/sec before stopping.

AT&T U-Verse
Chicago area

In regards to map selection, I know the map selection on ranked is restricted to Temple, Khylo, Nifelhel, and Foefire. But can unranked play still get Courtyard, Skyhammer, and Spirit Watch?

Jim Hunter.6821:

That is terrible news. You should add more maps like courtyard so that the entire game doesn’t revolve around who can bunker better.

Right, so on Courtyard it’s who can spam the most AoE damage and boons and long-cooldown elites. I’ll take conquest any day over the no-skill mess that is Courtyard.

ANet may be tweaking the match making parameters every couple days to see the outcomes. When you know the true effect of each tuning parameter, you can do a pretty good job of choosing each for the best overall outcome.

Also, the social experiment aspect is probably entertaining.

Yasha.5963:

Am I crazy?

Yes.

First time I played I lost 0-500, next time I won 500-0, but since then I have had a lot of scores inbetween. But putting aside the scores, I just love how you can take a completely different approach to conquest on the map. You can flank, mass stealth and surprise, get in and out of battles, there seems to be a “front” to manipulate etc.

This happens in conquest too. Watch some EU matches of the go4gw2 cup. However, conquest favors not zerg balling like you have in courtyard.

MonMalthias.4763:

That’s definitely one way to go about it. I think Anet definitely was right in “simplifying” the Amulet/Jewel/Rune system for PVP;

Agree with the rune changes, but completely disagree with the amulet changes. We need more stat customization in PvP, not less. It’s very hard to create a DPS build with even a little bit of extra armor. As a result, most DPS builds are super-squishy, which makes them easy prey for thieves and keeps them out of the meta.

MonMalthias.4763:

I think that it’s fair to say that the meta builds these days are at least mostly defined by their Sigils and Runes.
IP on Engi doesn’t really hurt until you put in Balthazar Rune.
Terrormancer thrives on Nightmare Rune.
D/D Ele running Strength or Hoelbrak have something like 15-20% higher effective Power than runes without Might Duration. (CMC has the exact calculation; I forget the exact percentiles)
Mesmer and Thief derive their auto pressure from – you guessed it – Air and Fire Sigil.
Warrior can hit 2-3k weapon swaps with Hydromancy Sigil; and guarantee an Earthshaker combo triple crit with Intelligence Sigil.
Condi Ranger’s Entangle is scary; but even scarier with Krait Runes.
Guardian’s down state control – already the highest in game – becomes ludicrous with Mercy Runes.

Disagree with some of this:

• Incendiary Powder is still far too strong even without Balthazar runes.
• I don’t know where you’re getting 2000 damage from Hydromancy procs. Even 1000 damage against low armor is stretching it (0.25 coefficient, but can crit).
• Guardians still run Trooper (formerly Soldier) runes over Mercy because they need the HP, and the condi cleanse is arguably better for them.

That Battle, Energy, Fire, Air, Doom, Intelligence, Hydromancy and Geomancy are so favoured over others for the aforementioned builds should speak volumes as to the internal imbalance in GW2. These same sigils were favoured before April 15; they’re still favoured now.

Agree here except Hydromancy. If you use Sigil of Force (+5% damage) as the metric for balance, then Battle (without boon duration) and Air are clearly way too good.

It’s a good change that fixed a loophole with the locking of weapons and abilities. Yes, it hurts some builds and team compositions. But now there is consistency between the start of the game and the rest of it, so in the long run, it will be easier to balance.

Using necro Deathshroud as an example – how can you balance a change to deathshroud mechanics when necros (except minion masters) could get about 20% at the start of a match? If you want to give all necros a base percent at the start but didn’t fix this loophole, then you’d have necros starting with 40%, which is a bit too good. Or maybe increase life force generation? Again, that extra 20% from minion death makes them extra powerful at the start.

Doing a “surgical change” to sigils or sigil combinations won’t work. If you change Sigil of Fire, people will replace it with something else like Sigil of Blood. If you want to add a new sigil in the future, then you’re really restricted as to what you can do.

Here’s where power creep came into play when each sigil had its own cooldown:

• On-crit sigils can activate at the same time for high burst.
• On-swap sigils are balanced around “normal” weapon swap mechanics and you aren’t necessarily going to swap weapons on cooldown. So for elementalist and engineer, which swap constantly, that’s no longer a factor.
• Warriors with Fast Hands can have 4 on-swap sigils activated within 10 seconds. That’s like having 4 sigils in one weapon set – basically two air and two fire sigils at once.

Having all sigils of the same activation mechanic share a cooldown is the best way to address all of those issues. And as long “shares a cooldown with other <on-swap, on-critical, on-hit> sigils” is in the text, it’s not overly confusing.

Sigil of Battle still needs a duration reduction, even with these changes.

The only other alternative I see is to significantly nerf the effects of all sigils so that they have less overall impact on gameplay.

style.6173:

I thought we were told that the matchmaking would prevent a pug team from playing a premade.

They never said that. The matchmaking parameters will try to match the sizes of groups who queued, but it’s not guaranteed since there are other criteria like your matchmaking rating and how long the system waits to match you.

Vigor traits were nerfed, but at the same time, the shared cooldown on sigils was lifted. So builds which didn’t use Sigil of Energy before that change snatched it up, and as a result there are more dodges, not less as ANet intended.

So basically the OP is raging about things where are not explicit? When players say to nerf elementalist and there’s one dominant build for elementalist, it’s typically implied that that’s what they want nerfed.

Elementalist Abilities
Agree with a lot of the changes, especially to burning duration.

Celestial Amulet
Celestial Amulet is fine. If the amulet was the problem, we’d see a lot more builds running it and being a nuisance. Many people forget that the d/d elementalist build was buffed/un-nerfed in the same patch which made the celestial amulet viable.

Might
Nerf Sigil of Battle to 12 seconds of 3 stacks of might instead of 20 seconds. This reduces damage by over 5% on builds which have boon duration traits and might duration runes.

The Berserker PvE “meta” exists because the AI is too easy to exploit and relies too heavily on low frequency, high damage, telegraphed attacks. When practically all the damage is negated by some dodges or evades or just chain CC’ing mobs into a corner, most defensive traits will never be useful in PvE.

This is way too harsh. People disconnect through no fault of their own from time to time. By locking them out immediately, you’re also punishing the people they play with.

How about some of these ideas?

• Give everyone a “get out of jail free” card for the first disconnect with a cooldown of a few days.
• If the player queued with others and disconnects but his or her allies don’t leave the game, lessen the penalty.
• If you re-enter the game after disconnecting, forgive or lessen the penalty for that disconnect.
• The quantity of ranked games played can impact the length of the penalty. If you play a lot without leaving, it’s more likely it was an accident before. Especially if you play the full length of losing games.

Also, is there anything being done to combat people who just stand around in spawn when their team isn’t doing well?