1579 replies to this topic

[Panzerschiffe]

Magic Pain Glove, on 30 April 2020 - 02:55 PM, said:

PGI or whoever is observing this thread . Can we get a conformation that you can ( or cant ) differentiate between someone dropping in a group and someone dropping solo in your DATA collection now that the queues have been merged ? ( Solo and Grouped WLR )

Because those two are gonna be two VERY different things .

I too would like to know this, as you can go to the current leaderboards, and see that a whole bunch of folks WL and KDR has skyrocketed. You have to go to the 4th page before W/L ratio drops below 5. Before the queue change, W/L ratio was in the 2-3 range by the first page.

[DAEDALOS513]

justcallme A S H, on 30 April 2020 - 02:27 PM, said:

Problem being that now invalidated some 30+ something posts you have made which go against your position.

If you are going to randomly pick a position, at least stick to it.

I liked his post because he didn't speak in absolutes.. like for example: "No good can come from this." He tempered his constructive criticisms with positives and gave insightful advice. Some guys should take a page from his book.

I never said grolo was perfect, I never said we'll consistently get competitive matches or that MM works.. but I did say, for the state MWO is in at the moment, this update was a huge boon. Instant drops with friends where you can try lance tactics.. when's the last time you heard people saying they were reinstalling MWO? I'm still in awe at how great this update was.

You don't want to stomp? Drop solo or in a 2man of non-meta.. it's up to you.. top 1% will never do so because .. you guessed it.. ego.

Edited by DAEDALOS513, 30 April 2020 - 03:08 PM.

[Larsh]

Don't know how much time this would take on the dev team. But, it might help to place a Tier ranking to each person in game. This could give us some more info on who is being matched at any time of day.

Example: Before a match, you see your team mates with a T1, T2, T3, T4, T5 next to it. Hell make it color coded too.

Then when the match ends, show tiers for both blue and red team.

[K O Z A K]

Magic Pain Glove, on 30 April 2020 - 02:31 PM, said:

OK guys , I just saw the latest tweet from Russ 1st 2nd 3rd and since I commented about 50 matches . They might be looking at my WLR . And if they are I want them to disclose it , every single match.

Because I think whats going on is that PGI is unable to differentiate solo games from group games much like the regular leaderboard , jarls and every other thing working with the MWO stat collection API . I got stomped on in solo but won most of the group games . However PGI statistics treat that as 35 / 15 KEKW .

we played you guys once yesterday and we won, so I think it's safe to say you lost 100% of your games

#pgimath

Jokes aside, you know they don't differentiate between solo and group stats in this new que. I think the only thing they were looking to see is if more people would play, if matches would start quicker and if stomps were more frequent. The reality is that more people are dropping (for now), games are faster and the actual amount of stomps is more or less the same. Stomps were frequent in QP before, and they are still so now, it's just before stomps were much more erratic, and now basically the strong groups control the stomps. If you're a lonely solo T4 puglet, you will get stomped more often now, and if you're playing in a group with good players, you will be doing a lot more of the stomping. The fun factor will hold up for a while until all the solo Q regulars realize they are losing more often as a result of this. Improvement in the overall reduced level of stupidity in the que may result in the opposite effect and make people not as unhappy that they're dying and losing more often.

[Brain Cancer]

DAEDALOS513, on 30 April 2020 - 12:03 PM, said:

Good point.. I know good players won't need extra ammo to take down a mech with extra armor because they can aim. Maybe they can increase ammo count a bit as well as armour for tiers 5 through 2.. the point is new players need some kind of handicap. This should be relatively easy to implement.

What do you suggest? They're not about to start work on fixing PSR.. they didn't do it when they were at full force so I doubt they do it anytime soon.

There is no fix that doesn't hard-separate lower-skill and higher-skill players, really. They don't belong together in the first place. And without a functional ranking system, there is no way to distinguish them either. The only fix at this point is a hard reset, re-ranking players with stats along the lines of Jarl's list, separating them out, and repeating the process with weekly checks and calculating said rankings with the most current 30-60 days worth of data and no more than a single tier shift per check (or less).

And enforcing that separation. Groups in grolo queue are automatically placed as the highest tier of their players, of course.

Not by radically remodeling stats. People go ultrasalty if a missile suddenly gets 10 more velocity. 50% more armor? Nuclear.

[Willard Phule]

So far, my experience has been that every single match has been a seal clubbing. Which side you're on depends on where the 4 man is. Being a solo, you tend to get long streaks of suck.

I can see this new "feature" was approached with the same reasoning and planning as just about every other "improvement." The matchmaker seems to ONLY be worried about making sure the premade group is kept together. No balance in weight class or PSR anymore.....as if there ever were in the first place.

[Zirconium Kaze]

I don't know. I always play solo, even in faction play. I think it's fine.

[Sergeant Destroy]

Is finee. Just up your tonnage......

[Brain Cancer]

Dakkalistic, on 30 April 2020 - 02:55 PM, said:

<endgame pic snipped>

Or 12/9

That was a gg indeed, and it's no exception in my experience, having dropped solo and grouped. Matches have indeed been more engaging, leagues above the ol' run to the middle and circle each other.

Yes, good groups will stomp, but that has been the case in FP, filthy sync drops, and frankly, the previous meta was stompy to begin with. I can honestly say, even if I get stomped now, I probably had a more enjoyable match than before the merge.

#1bucketforever

Interesting side notes:

If the alpha lance was actually a group, they have respective K/Ds of 0.25,0.84,0.31,and 0.70 from recent play.

John Tetradon, despite topping the kill count and nearly damage has no stats, indicating the account was likely activated this month. Given his skills, probably also smurfing. Kyle Harper who did top damage is really on the good side with well over a 2 K/D.

Other fun in the group: A 0.17 K/D player, and pretty much with the exception of you, John and Kyle everyone there dies more often than they live, frequently by very large margins. You can pretty much call that a very good carry, especially considering the player with the most kills is almost 100% the alt of someone with very nice stats on their regular account. (Edit: Or your opponents were horrid.)

Edited by Brain Cancer, 30 April 2020 - 03:46 PM.

[RickySpanish]

So a (the?) worry is that without differentiating solo vs group drops, it won't be possible to see that while overall stomp rate hasn't changed much, players in groups will pick up more wins than solo players?

But, as long as the MM always ensures there are the same number of grouped players per team, surely this won't matter? Grouped and Solo players as a whole will be winning and losing at a similar rate. Sure there will be very good players in groups who win a lot, but those groups will be carrying at least 8 solo players to victory as well. In my opinion the question is whether or not there are more "quality" matches as a whole, and that's a hard metric to measure. If pulling in groups shakes up the meta and makes matches more interesting, then it's surely a good thing for now. However, there absolutely *should* be a group only queue as well, so that top tier teams can elect to play there and face stiffer competition.

[Dee Eight]

I don't see why this was so complicated for PGI to understand... the majority of players abandoned using group queue because they had no desire to group up to play matches. Now they're being forced to deal with groups if they wish to play quick matches at all. How is this going to help player retention ? The time to have effected changes to improve the group play experience for those who wanted to do it was several years ago before all the previous fiasco decisions of PGI decimated the player base size... what do they think is going to happen when the pandemic shutdowns and isolation orders end ? That players are going to stick around for this sort of abuse to see what they do next to screw up the game ?

[Magic Pain Glove]

Hazeclaw, on 30 April 2020 - 03:15 PM, said:

Stomps were frequent in QP before, and they are still so now, it's just before stomps were much more erratic, and now basically the strong groups control the stomps. If you're a lonely solo T4 puglet, you will get stomped more often now, and if you're playing in a group with good players, you will be doing a lot more of the stomping. The fun factor will hold up for a while until all the solo Q regulars realize they are losing more often as a result of this.

Pretty much this . Altough Russ did admit on twitter that the number of stomps has increased by 5% overall . And is planning to mitigate that by (deep breath) tonnage restrictions ... All while people are posting screenshots like this with completely tonnage borked matches. 915 tons vs 560 tons .

[RickySpanish]

Magic Pain Glove, on 30 April 2020 - 03:48 PM, said:

Pretty much this . Altough Russ did admit on twitter that the number of stomps has increased by 5% overall . And is planning to mitigate that by (deep breath) tonnage restrictions ... All while people are posting screenshots like this with completely tonnage borked matches. 915 tons vs 560 tons .

That won't even make a difference since beta and charlie balancing is toast as well.

I bet the queues use different code for balancing rules, and it isn't modular, so porting over solo queue's tonnage balancing is either infeasible or will take more time than anyone wants to spend on it.

[Brain Cancer]

Dee Eight, on 30 April 2020 - 03:48 PM, said:

I don't see why this was so complicated for PGI to understand... the majority of players abandoned using group queue because they had no desire to group up to play matches.

I disagree. The majority of players abandoned group queue because they generally got whomped repeatedly by the good groups (usually larger ones at that) and found they were more likely to get wins in the chaotic tatersea that was solo queue on their own (or sync dropped, naughty naughty).

Now the people from the good groups can group up and 1) still beat the bad groupers from before + 2) stomp even harder on PUGs while having an 8-man disposable meat shield group that is a hash of mixed skill levels and gross disorganization, miscommunication, and misanthropy. The previous sanctuary has been violated.

[K O Z A K]

Magic Pain Glove, on 30 April 2020 - 03:48 PM, said:

Pretty much this . Altough Russ did admit on twitter that the number of stomps has increased by 5% overall . And is planning to mitigate that by (deep breath) tonnage restrictions ... All while people are posting screenshots like this with completely tonnage borked matches. 915 tons vs 560 tons .

Yeah drops like that are dumb now, if they won't bother to do any sort of matchmaking based on skill, at least do some tonnage based balancing before even touching the group tonnage. In regards to group tonnage, I honestly don't see it making any difference. As long as 4 solid players are able to take good mechs of any class, they will continue to slice up QP solos like cutting grass. They can push the limit down to 4 piranhas/fleas and it's still going to be the same. They will just be hurting all the "I'm tier 5 I don't care and I just want to play with friends" people.

[Spare Knight]

I have tried to have an open mind about it and I said I would give it the month, but it doesn't look good so far. I have put a good deal of money into this game and I regret it. I can only hope that things will get better.

[Magic Pain Glove]

RickySpanish, on 30 April 2020 - 03:44 PM, said:

Players in groups will pick up more wins than solo players?

Yes. Which is why if PGI needed to differentiate between solo WLR and group(ed) WLR , even before the queues were merged. Not many players can maintain WLR above 10 when playing solo , hell even 5-4 is impressive. However with groups this is different , especially well coordinated sweaty groups can get their WLR to crazy numbers like 8 or above . Impact you have on a match as a single player is far lower ( regardless of how good you are ) compared to a coordinated sweaty group with players of similar skill .

But since PGI ( if my assumption is correct ) cant ( or wont) differentiate between those they are getting misleading data and showcasing it as a basis for their future patch decisions.

Taking the last tweet from Russ as an example to illustrate. They see someone having won 35 games and lost 15 . Not too crazy right ?

But they dont realize that most of the matches won happened while the player was grouped and the losses happen when he was solo. So it looks something like 31/4 WL grouped and 4/11 WL solo. Which was pretty much the case with me after the patch. Which is why I suspected that they looked over my stats and asked them to disclose all matches and say if they can differentiate matches where I ( or someone else who they looked at ) was grouped and matches where I was playing solo and got stomped by other groups . Or they dont have to disclose anything they can just answer if they can differentiate between the two .

Like Hazeclaw stated earlier

Quote

Stomps were frequent in QP before, and they are still so now, it's just before stomps were much more erratic, and now basically the strong groups control the stomps.

[DAEDALOS513]

Brain Cancer, on 30 April 2020 - 03:19 PM, said:

People go ultrasalty if a missile suddenly gets 10 more velocity. 50% more armor? Nuclear.

hehe

[Liveish]

Have a read

https://twitter.com/...7327015936?s=20


Russ


quick analysis of final week of separated solo queue vs combined solo queue saw a rise in “stomps” but only by 5%. Stomp = 12-4 or worse. Running a more precise day to day comparison now to see if that holds true. I think that 5% can be improved with tweaks like group tonnage


Paul

So... plan is to tune down max tonnage of 4 person groups. Going to leave 3 and 2 alone for the time being. This is a test to see overall impact of tonnage variance in the combined queue. Because this will mess with group launching we will announce a time for the change.


Details will be coming shortly. One other thing I'll be doing in the background over the weekend is testing the tier separation. Problem with this is that it has a much bigger impact on wait times so these changes will not be persistent.

Final note on tier separation: There are the infamous release valves that allow the MM to kick off a match. Tightening these too much will put us back into the land of lonnng wait times for any match. Wash your hands
Also a poll for Lolz

https://twitter.com/...1701062658?s=20

Edited by live1991, 30 April 2020 - 05:05 PM.

[Excessive Paranoia]

Returned to the game recently after 3 years away, only to have this dropped in my lap. Mixing group queue with solo queue, is, was, and always will be, a bad idea. There are some possible qualifications that might have made it bearable (i.e. literally rewriting the entire MM / PSR to properly segregate players based on skill, including groups), but they are things PGI will literally never do, so they aren't really worth discussing. I'll go back on break for the next month, and if this is made permanent, whelp... GG...

Edited by Excessive Paranoia, 30 April 2020 - 05:12 PM.