377 replies to this topic

[Koniks]

Vlad, you should care about psychology and group dynamics because in a game with teams composed of strangers, how the individuals coordinate has a lot to do with success. It's why 2 players who operate together have good odds of thrashing an entire team of players who don't even if everyone's of similar skill.

Team cohesion and chemistry matter even if we don't have a good way to quantify it.

Even having a recognizably good player on a team can tilt the balance, not because they'll carry or call orders, but just because people will follow their blue Dorito. They're also just more likely to listen.

The same effect usually happens with a team that has a big group against a team of smaller groups. The rest of the players in the big group's team just fall in line.

Separating the queues rooms some of the worst edge cases that matchmaker would otherwise allow.

[Kjudoon]

As a contrary example, I. Hate skirmish. Yet i would resist anyone getting rid of that mode because it services a large portion of the population nor do I support making it the only mode. I prefer to opt out and leave tbose who love it in peace the same way i require them to stay out of assault mode and whine about capping like their opinion had validity.

[Johnny Z]

KinLuu, on 21 August 2015 - 11:29 PM, said:

Horrible idea.

The solo-queue was one of the best things that ever happened to this game. IMHO it is the only reason, that this game is still here at all.

Seems right to me. The guys that know the actually player stats know if this is true or not though.

Kjudoon, on 21 August 2015 - 11:45 PM, said:

And how do you then keep tier 5 and tier 1 from mixing? Ban grouping with other tiers? How do you solve the group average psr that has wrecked group queue due to large groups. The 5+ expansion was one of the greatest player pushed changes to the pub queues ever because now high skill players can easily be hidden in large groups for horrific mismatches. Your suggestion undoes a huge advancement in this game: the solo queue.

Having newbies not able to drop on the faction map is not a new idea. Thats one of the first suggested improvements made and for alot of reasons(not by me). So yes this could be an excellent use of the new teir system. Those below say tier 3 cant drop on the faction map. Stands to reason there could be a minimum skill level for advanced game play.

As a player this would do more to improve faction map game play than a match maker ever could. Mainly those that continue to troll the faction map to help Clan teams farm loyalty(which faction doesnt matter, but look at the faction map...) and take over the faction map wouldnt be able to do that constantly, and constantly ruining matches for legit players. I could be wrong and maybe a match maker is better?

Basically most faction map battles are close and enjoyable unless there is a lance of disconnected(or nearly so) green mechs dropping on the Inner Sphere side making it an 8 v 12 or worse lately. I always thought synch dropping would be the bane of the faction map since even before it was added by the way.



On topic. There is still a group queue?

Also on topic, dont let groups above 2 into the standard queue like world of warships.

If 12 mans are so advanced why they afraid to drop on the faction map?

Edited by Johnny Z, 22 August 2015 - 01:14 AM.

[Triordinant]

Vlad Ward, on 22 August 2015 - 12:21 AM, said:

Pretty much any player who's in Tier 1 is going to fit all of those criteria, whether they're in a group or not. Memorized maps? Maxed out, moduled mechs? Familiarity with standard tactics? Check, check, and check.

Train regularly together? Know each other's moves instinctively? Selected all their 'mechs ahead of time so that they work extremely well together and cover each other's weaknesses? Fail. Fail. Fail.

In any case, my original reply is still the best. Let PGI try it. It's the only way to be sure. Who knows? It might be for the best if they scrap everything, start from scratch and do it right. It won't even have to be a different company. PGI can close its doors and the people involved can start a new company using their friends/relatives names and "hire" exactly the same people who were "let go" before transferring the Mechwarrior license. The only difference is they'll be free from all their past mistakes and can't be sued for redoing everything.

[Rahul Roy]

I think the main problem with this idea is that players who don't want to play in (or against) groups.

This suggestion would mean they have to play against groups no matter what.

[Vlad Ward]

Kjudoon, on 22 August 2015 - 12:27 AM, said:

You whipped out your alleged degrees as an attempt to purchase credibility. That is an appeal to authority argument.

As for what does your purported bone fides as a comp player.... Retired....have to do with sociology it sets the context you are arguing from. You have played at a high level and have forgotten there are many... Potentially the majority who could not or do not desire or enjoy that level of play. By forcing them into your sociological paradigm you usurp their vision for yours devaluatinv their experience for your benefit.

I may not have an advanced degree, but i am smart enough to recognize an argument from selfish interest.

Error 404: Humor not found, man. Jesus. I mentioned my field as a jab at myself for not being all that great at applied sociology. I don't know what possible authority my degree in physics would actually grant me in a video game setting. Hooray, I can draw free body diagrams?

I retired from comp years ago after RHOD season 2, and have played casually since then. While I have played at a high level, and know what it entails, that's not what I do these days. Most of my drops now are solo or in groups of 2-4. I'm not trying to foist comp level play on anyone.

You seem to be getting way too heated over this, man. Take a chill pill for a minute. I'm not calling you out or anything.

Mizeur, on 22 August 2015 - 12:28 AM, said:

Vlad, you should care about psychology and group dynamics because in a game with teams composed of strangers, how the individuals coordinate has a lot to do with success. It's why 2 players who operate together have good odds of thrashing an entire team of players who don't even if everyone's of similar skill.

Team cohesion and chemistry matter even if we don't have a good way to quantify it.

Even having a recognizably good player on a team can tilt the balance, not because they'll carry or call orders, but just because people will follow their blue Dorito. They're also just more likely to listen.

The same effect usually happens with a team that has a big group against a team of smaller groups. The rest of the players in the big group's team just fall in line.

Separating the queues rooms some of the worst edge cases that matchmaker would otherwise allow.

It's true that coordination is helpful in MWO. I just don't see it making a huge difference in mid-level play, to be honest.

Players that either already coordinate well with randoms or at least know the most common strategies on each map are already going to be ranked higher than players who don't coordinate well or who do strange things and end up by themselves.

Teamwork is essential at high level play, whether you're all solos or not. You don't get to Tier 1 by herping off to a corner every match. And if a large group is obliterating people with their pristine teamwork and coordination, they'll quickly find themselves in Tier 1 or Tier 2 going up against pugs that know how to handle them.

Edited by Vlad Ward, 22 August 2015 - 12:41 AM.

[Triordinant]

Rahul Roy, on 22 August 2015 - 12:38 AM, said:

I think the main problem with this idea is that players who don't want to play in (or against) groups.

You're talking about the vast majority of the MWO population.

[Koniks]

Vlad Ward, on 22 August 2015 - 12:21 AM, said:

Ehhhh. ELO and PSR operate under fundamentally different grouping philosophies, both in the solo and group queue.

ELO attempted to match the average rating of one team as closely as possible to the average rating of the opposing team. While players closest to the current mean team ELO were always selected first, this wasn't a guarantee that you didn't get extreme variations in pilot skill.

That's imprecise and inaccurate. Elo was just the algorithm used to rate individual players based on their likelihood to win. The rating itself was adjusted by comparing their wins and losses against the likelihood of their team to win relative. That prediction used average team Elo and weighted the post-match change to the rating, if there was one, based on both the result and how unlikely it was compared to the prediction.


The matchmaking algorithm incorporates those ratings. But Elo rating doesn't inherently force teams to be balanced based on the team average. Matchmaker also could have controlled for the spread between the highest and lowest individual Elo ratings on a team.

Theoretically it did but the release valves overrode it to prevent unreasonable search times. Complicated by having to match on 4x3, game mode choice, and now server choice.

Elo ratings could have been tiered just as easily as PSR.

[Vlad Ward]

Triordinant, on 22 August 2015 - 12:34 AM, said:

Train regularly together? Know each other's moves instinctively? Selected all their 'mechs ahead of time so that they work extremely well together and cover each other's weaknesses? Fail. Fail. Fail.

This is pretty much all covered under knowing standard tactics, really. Most of mid-level MWO boils down to "Take good Mech. Take good position. Shoot until dead."

Most top tier mechs are not so niche as to be pigeon-holed into a role which could be rendered useless without the right team combination. Is a Tier 1 team going to be screwed because Random Player A took a Thunderbolt instead of a Timber Wolf? Not really.

[Appogee]

Vlad Ward, on 21 August 2015 - 10:50 PM, said:

Ideally, however, those people who don't listen and thus perform badly will end up in lower tiers and won't be matched with you against strong teams in the future.

AFAIK PGI still intend to use 'PSR-averaging' in the same way they used 'Elo-averaging' to try (and once again, fail) to create equally matched teams in the group queue.

In the group queue, teams still consist of a mix of players of widely divergent skill levels.

I don't see anything in the new matchmaking or PSR system that will change that in the group queue.

Perhaps PSR will prove to be a more accurate way to determining a player's skill. It can't be any worse than the ridiculous former idea that one lance worth of good players are only good (ie Elo goes up) if they can carry 9 noobs and beat a team with a higher average Elo at the same time.

Edited by Appogee, 22 August 2015 - 01:04 AM.

[Vlad Ward]

Mizeur, on 22 August 2015 - 12:42 AM, said:

Elo ratings could have been tiered just as easily as PSR.

Could have been. Weren't. Was a disappointing state of affairs. I understand the distinction you're making, and you're correct. I'm using a lot of imprecise language, mostly due to the volume of language I'm having to pour out right now.

[Vlad Ward]

Appogee, on 22 August 2015 - 12:45 AM, said:

AFAIK PGI still intend to use 'PSR-averaging' in the same way they used 'Elo-averaging'.

In the group queue still, teams consist of a mix of players of widely divergent skill levels.

I don't see anything in the new matchmaking or PSR system that will change that in the group queue.

The average they're using is of players who are grouped up together.

A group consisting of 3 Tier 2s, a Tier 1, and a Tier 4 will show up in the matchmaker as a mid-low Tier 2 entity.

Whole teams are not averaged. A Tier 2 match will draw exclusively from Tier 2 players unless the MM is forced to draw from Tier 1, Tier 3, or eventually Tier 4 due to population constraints. It will never draw a Tier 1 entity and a Tier 5 entity and call it "Tier 2.5"

[Triordinant]

Vlad Ward, on 22 August 2015 - 12:40 AM, said:

You seem to be getting way too heated over this, man.

I think the real issue here is only a small minority of MWO players like to group up and form Units, while at the same time most of the other MWO players don't want to play with that small minority. Forcing them to play with you will not work.

[Koniks]

Vlad Ward said:

You don't get to Tier 1 by herping off to a corner every match.

In solo queue you absolutely can take one of the Ravens, run off, and snipe the entire game to rack up a high match score, win or lose.

Your assumption about players quality is beside the point. If you combine the queues and use match teams based on PSR, then players who are coordinated pre-drop have a non-trivial advantage over those who have to coordinate on the fly.

This shows up in drop deck omposition, mechlabbing, pilot assignment, and even having people join a group in order so that their Assaults all end up in the same lance.

[Kjudoon]

Actually i am not getting too heated. I am an intense and passionate person though with a very good command of the vulgar ergot.

A long history studying sociology, religion, anthropology, history and politics out of personal interests has made me accutely aware of subtle contextual manipulations in communifation. Drives some people nuts but stops a lot of people running games cold.

Everyone has their own context bias and view. It is best to give people freedom to go their own direction.

[Vlad Ward]

Triordinant, on 22 August 2015 - 12:52 AM, said:

I think the real issue here is only a small minority of MWO players like to group up and form Units, while at the same time most of the other MWO players don't want to play with that small minority. Forcing them to play with you will not work.

And in the meantime, you get threads like this every time a casual dares to play with friends and hop in the group queue: http://mwomercs.com/...ing-good-teams/

You know what'll kill this game faster than anything? Launching on steam without the ability for people to play with their league of legends buddies without running into 228 and SJR because there aren't enough mid or low-level players in the group queue to get a good match.

See? I can make doomsday predictions, too.

Edited by Vlad Ward, 22 August 2015 - 12:56 AM.

[Johnny Z]

Triordinant, on 22 August 2015 - 12:40 AM, said:

You're talking about the vast majority of the MWO population.

Most people are casual players and do not want to join a guild. It doesnt matter what game it is.

Edited by Johnny Z, 22 August 2015 - 01:04 AM.

[Triordinant]

Vlad Ward, on 22 August 2015 - 12:55 AM, said:

You know what'll kill this game faster than anything? Launching on steam without the ability for people to play with their league of legends buddies without running into 228 and SJR because there aren't enough mid or low-level players in the group queue to get a good match.

That means my 3 queue solution is still the best IF there's a population explosion due to Steam.

Edited by Triordinant, 22 August 2015 - 01:02 AM.

[Vlad Ward]

Mizeur, on 22 August 2015 - 12:53 AM, said:

In solo queue you absolutely can take one of the Ravens, run off, and snipe the entire game to rack up a high match score, win or lose.

Your assumption about players quality is beside the point. If you combine the queues and use match teams based on PSR, then players who are coordinated pre-drop have a non-trivial advantage over those who have to coordinate on the fly.

This shows up in drop deck omposition, mechlabbing, pilot assignment, and even having people join a group in order so that their Assaults all end up in the same lance.

Er, well, hold on. Drop deck composition? Pilot assignment? I'm confused.

If Johnny is Unit A's best Light pilot, and Unit A sticks him in a Light, how is this different from Jessica who drops solo in her best Firestarter?

Mechlabbing should already come out in the tier lists. People with highly optimized Mechs are going to float to the top, and vice versa.

You can also change lance composition before the game starts. This won't help spawn locations, but almost every team meets up before any real engagements these days, no?

[Yokaiko]

Triordinant, on 22 August 2015 - 12:52 AM, said:

I think the real issue here is only a small minority of MWO players like to group up and form Units, while at the same time most of the other MWO players don't want to play with that small minority. Forcing them to play with you will not work.

I would love to see that where exactly you get that idea from?

You when I saw a mass exodus from my friends list? When they forced the 1-4 and 8 man groups. A few came back when they got over that ridiculousness.

Edited by Yokaiko, 22 August 2015 - 01:02 AM.