1579 replies to this topic

[Bulgroth]

Greetings,

I've taken some time to play several matches with the new combined settings and I have some observations and concerns about it.

First up, the observations:

i) The new settings have certainly decreased queue times and no longer do I sit in "SearchWarrior" for longer than a few seconds.

ii) Weight classes are far more even (3/3/3/3) than they were before. I saw a LOT of slanting towards the heavier 'Mechs in my average gameplay prior in Tier 2.

iii) The groups we're getting have a fair bit more teamwork and a lot of people talking to each other over voice comms.

And lastly, the concerns:

i) My friends are in the lower-skilled tiers (4/5) and lets just not beat around the bush on this... They're getting MURDERED. Its either they get in with the GOOD premade and have a quick stomp, or get in on the BAD one with lesser-coordinated groups and just DIE HORRIBLY.

ii) The dynamic of having premade groups in the solo queue has made the solo, low-skilled players second guess even PLAYING the game. My boyfriend plays less now and is considering taking a "break" until June when this solo/group thing ends. I've heard similar sentiment coming from various Social Media outlets like YouTubers and Twitch streamers.

iii) This one is sort-of a pro and con, but I feel its a bit more con when I'm on the context of the lower-skilled players/new players... Custom 'Mechs. In the low-skill tiers, you see a lot more trial and standard-configured 'Mechs. Add in the influx of the more-skilled ones in very coordinated groups on the enemy team... And you have a recipe for a bad time right there.


So... I guess what I'm getting at here is that perhaps there needs to be some consideration about skill tiers in the group matchmaking. My suggestion (and I'm sure others have voiced this too...) is to match the GROUP's highest skilled team member with THAT tier bracket. Ie: if a tier 5 groups with a tier 1, they'll get matched with tier 1-compatible players. This should help balance out this flip-flop between horrible and good matches that can be frankly, intimidating for lower-skilled players. NO ONE wants to go into a match, fearing they'll get stomped by a 4-pack of dudes in the same unit that are 3 or 4 tiers above them in skill.

On that note, the weight restrictions? Get rid of 'em. They're useless. ANYONE can slaughter a tonne of 'Mechs in any weight class. I've personally done 800+ damage with six kills in the PHX-2 trial 'Mech. I've done 1100+ damage in mediums, heavies, assaults... I've heard of a YouTuber that pulled off 7 kills in a LIGHT 'MECH with over 1600 damage. Weight doesn't matter. Skill will outweigh the 'Mech and serves little purpose in these groups. Let the players play what they want and let the matchmaker fix the weights in the matched groups.

Now, I know that it sounds like I'm doing a lot of complaining here and I realize that, but I want to take a moment to THANK PGI for even considering this change and even going so far as to implement it to TEST IT. Thank you. Thank you. Thank you. This test has changed the landscape of the solo queue and introduced an interesting dynamic to the drop system. No longer do we have to consider "sync-dropping" to get to play with our friends without ticking them off that we're on the enemy team AGAIN. I just feel we need to give some deeper thought into the underlying workings of this to better even-out the skill disparities we're likely to see with units queuing en-masse like I've seen of late.

[Grumpy Old Man]

First off, I love the work Anomalocaris has done in doing the calculations. Much appreciated to get better data than just our personal experiences, which are all filtered through our own opinions anyway.

Now, I did not have any more hugely uneven drops, but I cannot believe a rudimentary even weight distribution can be all that hard... I am not a programmer, but surely it cannot be all that difficult *after* choosing the 24 people that are going into the match to divide the total tonnage of the whole drop by 2 and then Tetris the weight together until you have a roughly even split. FFS, worst case we have 15 Solo Players in this match!!! Surely that is enough to shuffle them around some around the group tonnage?

***EDIT: I am trying to stay constructive. We can try to fix the problems that have been brought up one at a time...***

Edited by Grumpy Old Man, 03 May 2020 - 07:28 AM.

[Anomalocaris]

Champion of Khorne Lord of Blood, on 03 May 2020 - 06:41 AM, said:

I'm just confused a bit, why are we considering it as a group vs solos here when both teams get a group?

Because there is no skill based match making. Thus, best group wins. If only one side gets a good 4-man, the other 20 players in the match simply don't matter much. As we've seen in numerous streamed group drops where good players are getting WLR in the 10:1 range while grouped regardless of mechs.

Yes, you might have had a situation where 4 elite players got stacked on a team with the old soloq matchmaker, but it was _exceedingly_ rare. They were more likely to be split or even more likely to be in different matches (unless they were trying to sync). The old soloq matchmaker sucked, but at least it was random (can't believe I'm saying that).

Now, however, those 4 98% elite, tier1 (not that tier means too much) players can ensure they are on the same team every single game while the 16 solos are randomly selected from all tiers. Even if all the solos were tier1, we know that the skill level ranges from 30th percentile to 99th percentile - randomly selected.

Thus, unless the matchmaker puts another high skill 4-man opposite the first, the match conclusion is pretty much decided the moment the teams are finalized.

Let me put it to you another way. Imagine you're playing 7-card stud poker. You get to guarantee you get 3 aces (high) in your hand every round, with the remaining 4 cards being random. How often are you going to lose? Not much unless one of the other guys at the table gets the same advantage as you (the chance of drawing 3 of a kind in 7-card stud is about 5%).

Now, you'll say that you won't face an elite team every match. Of course not, but you'll see them plenty and no one wants to face a foregone conclusion, especially when equally matched teams are a lot more fun.

BTW, I found this interesting thread from 2013 regarding separation of the queues.

https://mwomercs.com...age__mode__show

Look at the votes. 60% wanted separation, but 53% wanted separation with the ability for solos to opt into group. 7 ******* years and this was still the preferred option of people today, yet PGI wouldn't do it. How ******* hardheaded can they be?

[Anomalocaris]

ERSmurf, on 03 May 2020 - 06:56 AM, said:

@Anomalocaris: Lots of interesting notes. I do have one question and I do apologize in advance, if it seems stupid, but please allow me explain.

Are you absolutely sure about wins and losses zeroing out? I am asking this, because there have been players in those old matches that no longer play and new players, who do not have old data to compare with, but still affect the current season one way or another. And players, who have too low total numbers of matches in either or both, but who still played, lost, won or tied.

You make a good point about people who play fewer than 10 games per season not showing up. If that number were sufficiently large it could skew the measured WLR in stats. Otherwise, WLR for the population should be very close to 1.0 each season.

[Champion of Khorne Lord of Blood]

Anomalocaris, on 03 May 2020 - 07:34 AM, said:

Because there is no skill based match making. Thus, best group wins. If only one side gets a good 4-man, the other 20 players in the match simply don't matter much. As we've seen in numerous streamed group drops where good players are getting WLR in the 10:1 range while grouped regardless of mechs.

Yes, you might have had a situation where 4 elite players got stacked on a team with the old soloq matchmaker, but it was _exceedingly_ rare. They were more likely to be split or even more likely to be in different matches (unless they were trying to sync). The old soloq matchmaker sucked, but at least it was random (can't believe I'm saying that).

Now, however, those 4 98% elite, tier1 (not that tier means too much) players can ensure they are on the same team every single game while the 16 solos are randomly selected from all tiers. Even if all the solos were tier1, we know that the skill level ranges from 30th percentile to 99th percentile - randomly selected.

Thus, unless the matchmaker puts another high skill 4-man opposite the first, the match conclusion is pretty much decided the moment the teams are finalized.

Let me put it to you another way. Imagine you're playing 7-card stud poker. You get to guarantee you get 3 aces (high) in your hand every round, with the remaining 4 cards being random. How often are you going to lose? Not much unless one of the other guys at the table gets the same advantage as you (the chance of drawing 3 of a kind in 7-card stud is about 5%).

Now, you'll say that you won't face an elite team every match. Of course not, but you'll see them plenty and no one wants to face a foregone conclusion, especially when equally matched teams are a lot more fun.

BTW, I found this interesting thread from 2013 regarding separation of the queues.

https://mwomercs.com...age__mode__show

Look at the votes. 60% wanted separation, but 53% wanted separation with the ability for solos to opt into group. 7 ******* years and this was still the preferred option of people today, yet PGI wouldn't do it. How ******* hardheaded can they be?

So it seems the best thing would be to just refine what we have a bit so the groups that fight are a bit more evenly matched with eachother in the fight. I still personally don't see it as really different than the normal queue chances of winning or losing. It's a 50-50 chance if your 4 man team beats their 4 man team each match similar to how prior it was a 50-50 chance if your team of randoms had any sort of coordination or otherwise skilled players that'd make them more useful to have around than the other team's players.

Good players grouping up and getting 10:1 WLR sounds honestly expected considering even solo they have very high winrates and KDRs especially. Infact when you have guys out there who have KDRs even just at 2 or above, they're already worth 2 players on average, the solo matchmaker never accounted for KDR or winrates anyway and just uses their tier, so an actual good player with a KDR that means he is worth an entire lance of players is considered the same as the guys who take 10k matches to grind to tier 1 and get a kill every other match.

Groups being thrown into the matchmaker isn't a problem, the problem is the matchmaker itself has never accounted for skill in any meaningful way so of course in any situation where grouping of any sort is allowed skilled players will come out pretty much unopposed since the matchmaker is just tossing entirely randomly skilled players with them.

[Nesutizale]

I agree, the matchmaker / PSR system is the bigger problem. When both sides get equaly matched teams it should be less of a problem.

[DAEDALOS513]

Weeny Machine, on 03 May 2020 - 06:55 AM, said:

And still you don't get it.

It is quite likely that you and the other group players will have as long fun as the normal solo queuers will turn around and leave. Sooner or later you will have your group queue back (it is just called QP now) and maybe it is equally dead then.

I drop just as much solo as group.. if not more.. I enjoy a challenge.

[Champion of Khorne Lord of Blood]

Nesutizale, on 03 May 2020 - 07:52 AM, said:

I agree, the matchmaker / PSR system is the bigger problem. When both sides get equaly matched teams it should be less of a problem.

I had mentioned it before my exodus, but my idea for a new matchmaker system is simply to consider the player's average KDR in the mech they're in or at least just overall. It's not absolutely perfect, but its far far better than the current system. Essentially their KDR is just a consideration of how many players they're worth, so a guy with 0.5 KDR is worth half a player, a guy with 5 KDR is worth 5, people who just farm KDR would be very quickly knocked down a few pegs if their 10.0 KDR meant that they're facing very tough opponents. Give the matchmaker the job of giving both teams the same number of "players" worth of KDR. So one team could end up having 12 people with 1 KDR on average each while the other team has 6 people with 2 KDR and 6 people with 0.5 KDR.

Seems simple enough to me though would lead to good players being stuck more on teams of worse players while against competition that is a bit better than the players on their team on average... though that's sort of what a skill based matchmaker is supposed to do in a situation where there are not enough players at every skill bracket on at all times to form full matches.

[Nesutizale]

Interesting idea. I also wondered how to take into concideration the mech the person brings but was more looking into giveing each mech a BattleValue similar to the tabletop system...also maybe not an exact copy.

Your system has the advantage that it takes into consideration how good a player is with the mech as even some strange configurations, that might fall under the radar with a BV system, can still outperforme as this particular player is extremly good with it. I like that.

[CptJudas]

ShooterMcGavin80, on 01 May 2020 - 08:06 PM, said:

Just played a game where a 4-man of SA guys on Polar highlands running 4 RLF-IIC-2's with 6xAC2's carried and stomped. Weird thing though was when we have just one pug left, they all pointed their arms at the sky and stopped shooting, even saying on the chat 'HANDS UP DON'T SHOOT'. I assume to let the enemy team get another kill or two.

Never saw anything like that nonsense happen when it was good ol' solo queue.

Disgraceful.

Wow. I'm sorry our fun offended you? We pointed our guns in the air at the end of the match b/c we had ran out of ammo from, as you pointed out, carrying you silly billy. 2 of our guys even died pretty early that match. So I'm wondering what is so disgraceful about playing the game and running out of ammo? Assuming our intentions is rather absurd friend. I would have loved to kill that last guy. B33f is running that deck for hours and literally creating a mushroom religion and we did three drops with riflemen and show up in the forums.

That being said, this shouldn't be a surprise to anyone as us more competitive players said that this would be the result of combining queues. Even if you adjust the group tonnage the end result is the same. Good players playing together are going to win almost every time and they will win by a lot.

Also, the disgraceful, heinous act for all the world to see.

Absolute monsters

Edited by CptJudas, 03 May 2020 - 08:05 AM.

[Anomalocaris]

Champion of Khorne Lord of Blood, on 03 May 2020 - 07:46 AM, said:

So it seems the best thing would be to just refine what we have a bit so the groups that fight are a bit more evenly matched with eachother in the fight. I still personally don't see it as really different than the normal queue chances of winning or losing. It's a 50-50 chance if your 4 man team beats their 4 man team each match similar to how prior it was a 50-50 chance if your team of randoms had any sort of coordination or otherwise skilled players that'd make them more useful to have around than the other team's players.

Good players grouping up and getting 10:1 WLR sounds honestly expected considering even solo they have very high winrates and KDRs especially. Infact when you have guys out there who have KDRs even just at 2 or above, they're already worth 2 players on average, the solo matchmaker never accounted for KDR or winrates anyway and just uses their tier, so an actual good player with a KDR that means he is worth an entire lance of players is considered the same as the guys who take 10k matches to grind to tier 1 and get a kill every other match.

Groups being thrown into the matchmaker isn't a problem, the problem is the matchmaker itself has never accounted for skill in any meaningful way so of course in any situation where grouping of any sort is allowed skilled players will come out pretty much unopposed since the matchmaker is just tossing entirely randomly skilled players with them.

We agree that the matchmaker is an issue. Always has been. But I think its critical to understand that the addition of groups to solo drops multiplies the magnitude of matchmaking issues. It makes it a _lot_ worse. More importantly, Russ has said just this week in response to questions on this very issue that improving PSR or implementing a skill based matchmaker just isn't a priority right now. So if they have no plans or desire to improve matchmaking, they never should have thrown the monkey wrench that is groups into the machinery of soloq. It has made competitiveness worse (overall stomps up 5% - although we don't know if that means stomps went from 10% to 15% of games, or from 10% to 10.5% - and Russ won't clarify which makes me suspect the former. He wouldn't even tell us what percentage of matches were stomps in the first place so they're playing politics with the data IMO).

[Anomalocaris]

DAEDALOS513, on 03 May 2020 - 07:55 AM, said:

I drop just as much solo as group.. if not more.. I enjoy a challenge.

If that's the case and you're winning your solo matches at about the same rate as previous seasons (I'll bet its a little lower though in solo drops) then that means your group win rate is between 4 and 5:1. So the impact of grouping up with a good 4man is an improvement of 300+% in your WLR. Useful data point on the impact of groups......

[Rick88]

The true disgrace is that red shooting an unarmed mechwarrior. Is there no honor anymore? What would his/her/their/thy/thou/thims/herms mother or father or wolfpack (if raised by wolves) think of such behavior? Also, this was one of a handful of matches our group our team brought "meta" mechs to the field. Typically we bring random stuff to make it more fun, but the results are about the same. We'll see where all this data goes...

[Marshal Jim Duncan]

Combining the Queues has brought a bunch of salty, negative, can't play as a team ppl into quick play. You would think the opposite would have happened. Perhaps the reason why so many people were not dropping in groups is because it allowed them to avoid the idiots that ruin their game play.

Maybe instead of focusing on stupid changes like this, focus on adding stuff like adding a hands-free option for VOIP comms and adding a 'always make me ready' button so you don't have to wait until someone finished hitting their bong before they ready up to start a match or reduce the number of clicks it takes to check your event status, or a whole bunch of other crap that would make this game less of a chore to get into an actual match. You guys did the same crap with MW5. It takes take forever to do something that provides no value with regards to enjoying the point of the game: shooting things.

[Anomalocaris]

CptJudas, on 03 May 2020 - 08:03 AM, said:

Wow. I'm sorry our fun offended you? We pointed our guns in the air at the end of the match b/c we had ran out of ammo from, as you pointed out, carrying you silly billy. 2 of our guys even died pretty early that match. So I'm wondering what is so disgraceful about playing the game and running out of ammo? Assuming our intentions is rather absurd friend. I would have loved to kill that last guy. B33f is running that deck for hours and literally creating a mushroom religion and we did three drops with riflemen and show up in the forums.

That being said, this shouldn't be a surprise to anyone as us more competitive players said that this would be the result of combining queues. Even if you adjust the group tonnage the end result is the same. Good players playing together are going to win almost every time and they will win by a lot.

Also, the disgraceful, heinous act for all the world to see.

Absolute monsters

Alright Shooter, I think you gotta retract the criticism here. Outta ammo and pulling that was pretty funny and harmless.

[Anomalocaris]

Alreech, on 03 May 2020 - 06:47 AM, said:

Your conclusion: get the groups out of MWO, because most players dont want working together "thats the reason why group queue died".
My conclusion: improve grouping up to get more players into working together.

Could you be more wrong? I say let people play with who they want to play. I don't want to play with pre-mades. Lots of people felt like I did so they gave us soloq, the most successful queue of them all in terms of population. You don't want to play solo, fine, play group queue or faction. But....

Those who wanted to play groups didn't play enough, or couldn't keep a sufficient player base. If you think you have ways to get them back, fine, apply them to GROUP queue. Don't force solos to fill in teams with groups because when you force people to do something they don't want in a F2P game, something amazing happens. They leave.

I will never play in a mode where I'm forced to be a solo vs pre-mades, and I've tried group queue in the past and have no desire to be _part_ of a pre-made. That's my decision. PGI may choose to make it permanent in 3 weeks, in which case, oh well, move along. I'm still here because I love playing MWO in the old soloq and I'm hoping to convince PGI and enough players that reverting is the best option. We'll know in 3 weeks.

[ERSmurf]

@Anomalocaris: Thank you for the additional clarification + extra notes.

[FRAGTAST1C]

Larsh, on 03 May 2020 - 04:52 AM, said:

Can't dispute live footage. I've been watching many of the streamers to try to get a solid idea of the playerbase.

One thing I will mention though, is that some here are mentioning the most top viewed streamers. Have you had a chance to view the other streamers that may get 1 or 2 views the whole night? They may not be the best quality, or may not know how to interact with a full audience, but it may be a good idea to get their insight as well.

While you can view some of the lesser known streamers, you should take the matches into account and not what they say. I wouldn't take anyone's opinion seriously when the facts are clear from their own material. That's the point I'm trying to make here. Let's go with the facts instead of personal experience and opinions.

To make it clear, the merger can work if there are a few things put in place. The way it is now is just nonsensical. I mentioned them before but I'll list them again so we can hopefully compromise and let the 3 eejits in PGI understand the real issues.

1. Tiers should be separate and T5s shouldn't be in T1. Furthermore, fix the PSR such that it actually reflects the player's skill instead of the number of matches played regardless of performance.

2. Bring back old spawns where Alpha, Bravo and Charlie were meant for Light/Medium, Medium/Heavy and Heavy/Assault respective. The current way is stupid. There's no need to make the 4-man always be in the same lance.

3. End this garbage of treating an individual as a group-of-1 and spread the tonnage evenly between the teams.

4. Assess the threat value of a group. Stop treating a weekend group the same as a group of highly skilled players. Restricting tonnage to even 160t for 4-man won't affect them.

5. Stop placing 3-man and 2-man groups in the same team. Don't create matches if one team has a group and the other doesn't. Wait for another group. Extending the wait time in such instances to 1 min won't kill anyone.

[Alreech]

Anomalocaris, on 03 May 2020 - 08:28 AM, said:

Could you be more wrong? I say let people play with who they want to play. I don't want to play with pre-mades. Lots of people felt like I did so they gave us soloq, the most successful queue of them all in terms of population.

Solo Queue lost less players than Group Queue doesn't make it more successful.
And losing players doesn't make a game succesful.

Quote

You don't want to play solo, fine, play group queue or faction.

"And if you don't find matches in group, stop playing MWO?

Faction play is indeed an option to group play, with one big problem:
PGI tells everyone to stay the hell out of it if not being highly "competetive or organised".

It's a better option because of the drop decks (mech selection & respawns), it has the same problems as Group & Solo:
No coordination between Groups or Players
Maps to small for 12 vs 12 or even 8 vs 8.
No balancing between Groups or Players.

Quote

Those who wanted to play groups didn't play enough, or couldn't keep a sufficient player base. If you think you have ways to get them back, fine, apply them to GROUP queue.

Too late, Group Queue is dead, killed by PGI a long time ago.

Group queue is needed for new players who want to play with their friends, the way a game normally grows.
Having to play without friends won't help getting new players into MWO, no matter how "succesfull" Solo Queue is.

[Anomalocaris]

Alreech, on 03 May 2020 - 09:19 AM, said:

Solo Queue lost less players than Group Queue doesn't make it more successful.

And losing players doesn't make a game succesful.

"And if you don't find matches in group, stop playing MWO?

Too late, Group Queue is dead, killed by PGI a long time ago.

Group queue is needed for new players who want to play with their friends, the way a game normally grows.

Having to play without friends won't help getting new players into MWO, no matter how "succesfull" Solo Queue is.

To use your logic, if I can't find a solo game anymore I should just quit too. So its you or me? I choose me. Some people will leave over this merge. Some will come back. I highly doubt many newbies will come in. But you will sacrifice game play quality in the process. Which will accelerate the drain and reduce newbie retention. If they had kept tweaking group queue and added the solo opt-in, maybe they could have improved retention without forcing anyone out. That was a potential win-win, but with the merge you're hoping to gain more than you lose - risky given PGI's history.

Every game gets old and loses players. Short of actually releasing a new MWO:2 with an actual marketing campaign and good gameplay, this game is not going to grow its playerbase, all you can do is hope to keep the drain as slow as possible.

[color=#222222][font=&amp][/font][/color][/left]

Look at the player numbers on Jarl's

https://leaderboard.isengrim.org/stats

Average add over the last 20 months is about 1500 new players per month. And, with the exception of the quarantine bump last month, the game had lost 50% of its population in that time frame. You're not going to reverse that.

Edited by Anomalocaris, 03 May 2020 - 09:53 AM.