Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

Matchmaking criterias

STAR WARS: The Old Republic > English > PvP
Matchmaking criterias

MidichIorian's Avatar


MidichIorian
01.26.2013 , 11:39 AM | #1
We need them, 12 months ago. I'm aware that many players are fine with simply joining games and screwing around, it's IMO as productive as watching paint dry, or rolling pugs in a premade but a few of us are actually interested in the experience and are doing our best to win, regardless of it being non-rated. It is impossible to have a fun expereince in combination with being somewhat competitive in the current enviroment and I, just like many others, am on the verge of moving on just because of it.

Disclaimer: "Join a group" is not a valid counter argument, fact remains that a lot of people, if not the majority, are solo queueing, either by choice or because their friends arent on. It's not going to change so you might aswell accept that this is the reality. Nor is claiming that it's a MMO a good argument because more or less every game on the market has a multiplayer mode these days and in every game we have people solo queueing despite that it often would have been to their advantage to queue in a group. Solo queueing is there and it's not going away

Problem 1.Skill disparity. I'll be getting to the premade issue but this would be one way of making sure that in the event that you have to face a premade you won't be doing it with people who not only can't communicate but also, individually, are much less expereinced/skilled than the players on the other team. Countless are the games where I walk away with zero or very few deaths, top dps and top objective points at the same time as my team mates are more or less farmed. I find it incredibly frustrating to be surrounded by 7 black nameplates on the scoreboard so I cannot understand how my team mates can be ok with placing 10-16 in game after game. Is it really better to get a game two minutes faster and have an awful warzone expereince than have the matchmaking system search for compatible players? By simple matching players based on valor or games played we'd improve (not fix, only improve) the situation, due to the new/inexpereinced chars/players being placed in their own games.

Problem 2: Backfilling. I cannot unerstand how someone can be so casual and care so little about the outcome of games that they are ok with backfilling a game where the outcome is determined long before you've loaded the map or the game simply is about to end and you barely have time to leave the spawn. I don't even care about the comms and I hate when people say that it's great "because they'll get a quick daily if they only have to play 2 minute games"!. Why are these people even playing the game? They are obviously more interested in the comms than the expereince and what are they going to do with the comms in the first place if they're not going to play properly? You can get a few medals in recruit gear by simply camping a node so you don't need gear for that. This phenomenon is much older than stab farming but stab farming might have made it worse, I don't know.

Problem 3: Premades. Yeah I said it. I've reached the point where I've come to despise some premades on my main server because I cannot understand how it can be fun for them to enter games more or less knowing that they're going to win. And I shouldnt despise them because it's not their fault that the matchmaking system is lacking. Sure, they could play rateds but if players have problems finding people to play normal warzones with it's even harder to have 8 people on for rateds, not to mention that it's not a given that anyone els is queueing at that very monent. Either way, it's beyond me how people aren't willing to trade slightly longer queue times for better matchmaking experiences. And again, the solution to Problem 1 would improve the situation without having to put solo players and premades in different queues. Actually, I'm not even asking for them to be in different queues, I'm asking for the system to add a 3 or so minute buffer time span during which the system only looks for either other solo players or other premades, depending on how you queue. Once that time span is up it will grab the first players it can find, like the current system. We can't trade a few minutes for a better expereince?

Problem 4. Class balance. there's not much to say about this, you'd be a fool to not acknowledge that there are issues. I've come to tolerate it but it would have been great if the matchmaking system somehow could take classes into consideration when forming the teams. It shouldnt affect queue times since the system wouldnt be looking for different classes, it would simply split the players who already had been accepted into a game based on their classes. Might reduce some of those 4 healers VS zero healers or where one class is making up an entire team- games.

Also, notice how I didn't list "gear" as a problem? While it mathematically is boosting a person's performance I don't see it as the main reason to why people are underperforming. If you can't break 100K in recruit gear it's not the gear, it's you and you're not going to top charts regardless of the gear you're wearing. There is however a strong correlation between being inexpereinced and wearing crappy gear. I'm willing to bet that the inexpereince/skill has more to do with it than the gear. I mean, if you are in full recruit when you ding 50 you've obviously not saved up any comms, which by itself suggests that you arent that interested in PvP to begin with and perhaps not the best player on the field.

tl:dr - Lot's of QQ'ing about the current system. inb4L2P

Warshades's Avatar


Warshades
01.26.2013 , 11:52 AM | #2
I dont trust bioware with something like that. Look at the group finder. Hour plus weight time for the group finder to match the right classes. And I get the premade argument, but I dont get the dismissive attitude to grouping before queing. Not that I disagree with solo queues, just not about complaining about other people grouping while I chose to solo queue. Make sense? Basically Your asking for the game to attempt to find you other solo people to be grouped with and ensure the other team doesnt also have grouped players. In this case I have to disagree and say if you want to do group activity, but dont want to make a group to do it......accept that it wont always work to your advantage.
“Light thinks it travels faster than anything but it is wrong. No matter how fast light travels, it finds the darkness has always got there first, and is waiting for it.”

Carter_Mathis's Avatar


Carter_Mathis
01.26.2013 , 12:23 PM | #3
Quote: Originally Posted by Warshades View Post
I dont trust bioware with something like that. Look at the group finder. Hour plus weight time for the group finder to match the right classes. And I get the premade argument, but I dont get the dismissive attitude to grouping before queing. Not that I disagree with solo queues, just not about complaining about other people grouping while I chose to solo queue. Make sense? Basically Your asking for the game to attempt to find you other solo people to be grouped with and ensure the other team doesnt also have grouped players. In this case I have to disagree and say if you want to do group activity, but dont want to make a group to do it......accept that it wont always work to your advantage.
The group finder takes so long because there either isn't a tank or a healer queuing at the time, not because it's broken. There are plenty of instances where my tank or healer toon solo queues in the lfg tool and insta-pops because I was the only one queued. It's not a mechanical issue it's a player one.

On the original topic, asking people to separate based on skill is nearly impossible, especially since i've had many horribly long conversations with dps who average 125k damage and 0 objective and will swear that they're amazing. And basing it off of valor rating doesn't do much, since absolutely horrid players seems to stay that way despite their valor rating, and it allows people who legacy gear over to their other 50's to ding in with lower valor/probably undergeared people (Ex. my optimized EWH smash guardian is valor 45).

NoTomorrow's Avatar


NoTomorrow
01.26.2013 , 12:27 PM | #4
Bioware currently doesn't have the muscle, nor experience to pull out something like this. Matchmaking is a paramount feature for a team based game like SWTOR and the lack of it ruins the experience for many players.

You'd think that a game with so much PuGing would have a voice com integrated in the client.. And we are asking them for a matchmaking system? It will not be better than this.
Quote: Originally Posted by EricMusco View Post
Scatter Bombs are meant to be a fun bit of extra damage that occurs when you roll into or away from the action. That said, we’re okay with you trapping an unsuspecting enemy for a “wall bang” every now and then.

Warshades's Avatar


Warshades
01.26.2013 , 01:23 PM | #5
Quote: Originally Posted by Carter_Mathis View Post
The group finder takes so long because there either isn't a tank or a healer queuing at the time, not because it's broken. There are plenty of instances where my tank or healer toon solo queues in the lfg tool and insta-pops because I was the only one queued. It's not a mechanical issue it's a player one.
In his 4th point he mentioned taking into account class composition, which is what group finder does. If you think there are long waits for dps, what do you think will happen with wz if similar requirements are implemented? In his example he gave 4 healer groups or a group full of sentinels. Well what if your playing an over populated class like sentinel? Are you suddenly going to have to wait an hour to join a wz? Better yet do you think Bioware is capable of creating a match making system for pvp that keeps wz pops quicker than a few minutes?
“Light thinks it travels faster than anything but it is wrong. No matter how fast light travels, it finds the darkness has always got there first, and is waiting for it.”

FREDDOSPWN's Avatar


FREDDOSPWN
01.26.2013 , 01:25 PM | #6
Solution 1) Games with closer skill levels are generally more enjoyable. With that said, you cannot separate players based upon skill by using valor. Valor is an indication of experience, not skill. You would have to introduce a separate statistic to keep track of skill. That would create even more issues (losing rating because you only ever get 7 'bad' players on your team). The easiest solution would probably be to hide the skill value.

Solution 2) Some people try to make the best of a bad situation. The issue is with people leaving warzones early. If they did not leave, there would be no back-filling required. You should put your focus on making leaving a warzone more unattractive.

Solution 3) Not all groups of players are the top-tier PvP players. I would say that most groups are formed to initiate a buffer against some of the players you have been describing (along with friends just wanting to play together). A group would result in a maximum of 4, rather than 7, potentially less-optimal players.
A group vs group queue would not work, simply because being in a group does not mean you will win. It is very possible to beat an enemy group with a full solo queue team. This is where Solution 1) comes in. Groups would simply need a modifier added to them, to factor in the communication advantage.

Solution 4) Sorting by class is not feasible. If you sorted by AC potential, then you would still end up with your example of 4 healers vs none. If you sorted by specialisation, then people could just change their specialisation in the warzone (which would also end up with teams of no healers). You would get people queueing up as a healer to get a faster queue pop too.
How to fix that exploitation? Do not allow people to re-specialise once a warzone queue has popped (people waiting for a pop, then refund points). This also then creates the problem of people forgetting to swap to the correct specialisation.

My point is that many of your gripes with the system cannot be easily remedied with exploit-free solutions. Trying to fix the problem can (and probably will) create more issues than it solves.

Washingtoon's Avatar


Washingtoon
01.26.2013 , 02:08 PM | #7
Quote: Originally Posted by Warshades View Post
In his 4th point he mentioned taking into account class composition, which is what group finder does. If you think there are long waits for dps, what do you think will happen with wz if similar requirements are implemented? In his example he gave 4 healer groups or a group full of sentinels. Well what if your playing an over populated class like sentinel? Are you suddenly going to have to wait an hour to join a wz? Better yet do you think Bioware is capable of creating a match making system for pvp that keeps wz pops quicker than a few minutes?
Hey, it's the OP here again, currently on the account I'm using from work.

To clarify, which I tried to do in the original post too, I'm not asking for the system to look for specific roles. What I meant was that when 16 players have been found the system will start to take classes into consideration. Hence it will only work in same faction games. So, as an example, if seven out of those 16 happen to be snipers it won't put six of them on the same team.

As for specializations, I'm fully aware that it's impossible to fix all issues, this is more of damage control, but even if not every sorc out there plays healer it must be better to split them evenly over two teams instead of having 7 of them face 5 warriors, or two other AC'es being distributed like that. I've been in plenty of games where it basically has been AC vs AC and it's never fun.

Quote: Originally Posted by FREDDOSPWN View Post
Solution 1) Games with closer skill levels are generally more enjoyable. With that said, you cannot separate players based upon skill by using valor. Valor is an indication of experience, not skill. You would have to introduce a separate statistic to keep track of skill. That would create even more issues (losing rating because you only ever get 7 'bad' players on your team). The easiest solution would probably be to hide the skill value.

Solution 2) Some people try to make the best of a bad situation. The issue is with people leaving warzones early. If they did not leave, there would be no back-filling required. You should put your focus on making leaving a warzone more unattractive.

Solution 3) Not all groups of players are the top-tier PvP players. I would say that most groups are formed to initiate a buffer against some of the players you have been describing (along with friends just wanting to play together). A group would result in a maximum of 4, rather than 7, potentially less-optimal players.
A group vs group queue would not work, simply because being in a group does not mean you will win. It is very possible to beat an enemy group with a full solo queue team. This is where Solution 1) comes in. Groups would simply need a modifier added to them, to factor in the communication advantage.

Solution 4) Sorting by class is not feasible. If you sorted by AC potential, then you would still end up with your example of 4 healers vs none. If you sorted by specialisation, then people could just change their specialisation in the warzone (which would also end up with teams of no healers). You would get people queueing up as a healer to get a faster queue pop too.
How to fix that exploitation? Do not allow people to re-specialise once a warzone queue has popped (people waiting for a pop, then refund points). This also then creates the problem of people forgetting to swap to the correct specialisation.

My point is that many of your gripes with the system cannot be easily remedied with exploit-free solutions. Trying to fix the problem can (and probably will) create more issues than it solves.
Again, these are suggestions on damage control and not actual fixes. I'm of the opinion that the matchmaking system can never be truely fixed until we have x-server queues and the population is big enough for BW to add real criterias to the system.

1. I used valor as an example since it's already an implemented system. I'm trying to avoid big ideas such as independent skill level systems since that, looking at the current situation, never will become reality. Even if valor wouldnt exclude the possibility that a really good player on a new char gets to face inexperienced players it would most likely keep the valor 80's and up seperated from the freshies. No guarantee that a valo +80 in any way will be skilled and even if I've already stated that the gear gap is exaggerated we'd probably avoid games where a team got stuck with 14K people agaimnst a full WH team. I too see lot's of ways to exploit a system like this but fact still remains that the current system has no criterias and only a part of the community would take advantage of exploits. The net outcome is that we'll see some improvement.

2. That's a valid point. On the other hand, disconnects arent "that" common so if players on a team drop it's probably because the team is losing. That means that they deserve to lose and by not backfilling it will probably happen faster. If anything I'd say that it's unfair to the winning team that the other team gets replacements and potentially win, although that doesnt happen particulary often. I'm all for balancing teams but once the game has started it shouldnt be affected by new people. I'm in favor of a combination; No backfilling and a lockout on comms for quitters.

3. True that far from all premades are good. Some are gosh darn horrible but I'm of the opinion that more or less all premades set the tone for the game and inevitably determine the outcome of games, win or loss. I don't want to solo queue and win or lose just because I or the other team happened to get "that one premade". I don't see why solo queuers should have to suffer just because premades can't keep up with eachother. If someone is in a premade they have all the tools to communicate and improve so let them do that while playing other premades.

4. See above.