77 replies to this topic

[TruMantas]

Not to make too fine a point, but I believe there are improvements to be had in matchmaking.
Especially when it comes to allowing stacks just dropping into QP...

I'll just let the the post-game screen speak for itself.

Spoiler warning: It's not pretty.

Spoiler

[Kotis77]

Matchmaker can be better, but PGI/Sean Lang dont want it to be improved. But how much it would make games better thats the question.

There is matches that you will know outcome when you see players on each side. I think everyone hates that kinda matches. If you got 0% chances of winning matches it will drive players away.

My POV is that matchmaker should balance matches that you got a chance to win the match, even 1% is better than 0%. This what we got now doesnt allow that on some matches.

There is lots of problems/variables other than matchmaker, but 0% of winning sucks.

Even if we got matchmaker fixed, one of the biggest problem for stompy matches is snowballing that happens on that there is no hitpoint regeneration/heals(or ppl just feed) like other FPS games that will allow comebacks, but its what makes this game different.

Edited by Kotis77, 13 July 2023 - 09:54 AM.

[pbiggz]

all things considered the matchmaker actually performs exceptionally well. This game doesnt get more than 1000 concurrent players at peak hours, and its about half that during off hours. There are only so many people, and only so many matches to be formed. To paraphrase the PGI devs, the longer you wait in the matchmaker, the worse your match will be, the best matches to the matchmaker are the ones that form statistically, so saying you'd be willing to wait longer for a better match, as many seem to say, actually doesn't really help here.

The reality is, this thing isn't magic.

Now there are solutions about. Some of them are real solutions, and some of them, frankly, are fake.

To start with the fake solution, which, lets not beat around the bush, is, effectively, banning group players:

TruMantas, on 13 July 2023 - 08:34 AM, said:

Especially when it comes to allowing stacks just dropping into QP...

This is not a solution. This is suicide. Not only would removing groups have little effect on your match quality, it would, in my view, have a seriously negative effect on match quality, mostly because there would be no more matches, because this game would have shut down 3 years ago.

As for real solutions, some are actually floating around:

More granular tiers; im not sure how this would play out, but giving the match more meat could theoretically let it assemble more statistically even matches.

Qualifying games; this one, im frankly not sure why we didn't get from day one; most matchmade games force new players to play a number of gently matchmade games to test their skills; from there, the game generates an appropriate player rating for them, and directly inserts them into the rankings. In MWO terms, this would mean maybe 5 or 10 games initially played without your tier visible, from which the game would produce your appropriate PSR, and place you in the rankings directly. A new player who is very good at shooters could get placed right in Tier 1 or 2, while a more casual player who doesn't play shooters would probably end up in 4 or 5. From then on, PSR would behave as it does today, and players with a dramatic skill gap would get some much needed insulation from one another. The common, if dubious complaint that elite players frequently make smurfs to farm newbies and casuals would be eliminated by this proposal, because it would be far more difficult to end up in an artifically low tier unless you deliberately threw games, which is TOS.

Nine-Ball, on 13 July 2023 - 09:35 AM, said:

4 group drop =/= automatic win

tbh given his match results it looks like they lost despite the 4 man on his team doing alright.

[RickySpanish]

^ In-lobby PSR weighted team balancing would be the best fix. It'd require programming though, which apparently the MWO codebase is allergic to. Theoretically the game would look at the players and groups it has found to form a match, and juggle a few around to balance PSR in some way.

[Duke Falcon]

Hmmm, five RIMS in a single match on the same side?
Not remember ever happened that so much people from my unit were deployed together. Usually forms groups of three and usually funny to drop with them...

As for MM, we have no better but we may could. Some nice options listed above, some are not as hard to implement.

RickySpanish, on 13 July 2023 - 11:13 AM, said:

^ In-lobby PSR weighted team balancing would be the best fix. It'd require programming though, which apparently the MWO codebase is allergic to. Theoretically the game would look at the players and groups it has found to form a match, and juggle a few around to balance PSR in some way.

We agree that theory is good. Just rarely works as it should've be. But population number is also a factor. If we have 2-3 times more players than currently MM would work better for sure. Now, despite it bleeds from a thousand of wounds, it is forced to split by zero...

Edited by Duke Falcon, 13 July 2023 - 11:28 AM.

[KingCobra]

TruMantas, on 13 July 2023 - 08:34 AM, said:

Not to make too fine a point, but I believe there are improvements to be had in matchmaking.
Especially when it comes to allowing stacks just dropping into QP...

I'll just let the the post-game screen speak for itself.

Spoiler warning: It's not pretty.

Spoiler

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------
How many times does this topic come up on the forums? 1000 times a month? How many times do many old players to MWO have to answer this topic? 10000 times a month? It is very simple to answer the matchmaker has not worked for years why? because 500,000 players bailed MWO leaving it with a very low population of players which means the match maker cannot work properly so the DEVS or JR DEVS said hey just lump players together no matter the tier and let's play stompy robot battles and not wait 10-20 minutes for a battle to start.

Then they added 4 man groups into the drop equation which has made it even more unbalanced and totally idiotic as 90% of the battles now are one sided crappy games. So there you have the correct answer to your post op is there a cure? only one put all the game modes back the way they once were and gain a few 100 thousand new players to make the matchmaker work correctly.

P.S As I see MWO this last year there have been improvements in a few areas but the game is getting closer to World Of Tanks the game MWO most resembles and the devs wanted it to be a Clone of WOT with Mech skins and a E-Sport game.

[pbiggz]

KingCobra, on 13 July 2023 - 11:53 AM, said:

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------
How many times does this topic come up on the forums? 1000 times a month? How many times do many old players to MWO have to answer this topic? 10000 times a month? It is very simple to answer the matchmaker has not worked for years why? because 500,000 players bailed MWO leaving it with a very low population of players which means the match maker cannot work properly so the DEVS or JR DEVS said hey just lump players together no matter the tier and let's play stompy robot battles and not wait 10-20 minutes for a battle to start.

I will repeat, the matchmaker is actually working remarkably well, especially at peak hours. It could absolutely be better, but this doom and gloom attitude does nobody favours. The population is low, but its stable. 1000 players a night, for 3 years running. Efforts need to be made to shore that up, but we aren't standing at death's door here, so the fatalistic whining is entirely uncalled for.

KingCobra, on 13 July 2023 - 11:53 AM, said:

Then they added 4 man groups into the drop equation which has made it even more unbalanced and totally idiotic as 90% of the battles now are one sided crappy games. So there you have the correct answer to your post op is there a cure? only one put all the game modes back the way they once were and gain a few 100 thousand new players to make the matchmaker work correctly.

And I will repeat again. Groups are not the reason you lose games, and the narrative that players have fled en masse from groups is not borne out in any of the data that is available. PGI doesnt make much data available but steam does, and if people were running away from groups, the steam player chart would show that. It doesn't. In fact, it shows a remarkable degree of stability. 1000 players at peak hours, for 3 years straight.

A cursory look at jarls lists stats also suggests there wasn't a major change in KD ratios before and after queues consolidated. I have personally checked the stats of many of the loudest anti-group guys, and their KD rates were basically the same on average. If they thought they were losing more after the soup queue, it was entirely in their heads. That isn't representative of the community at large; that kind of info is difficult or impossible to get, but it is certainly revealing when you consider who is most enthusiastically opposed to groups.

I welcome everyone to be critical. The first step to that though, is knowing what you are talking about. If you feel the need to talk out of your *** you are not helping anyone, PGI, or players, to improve and enjoy this game more. If you feel the need to lie to make a point maybe the point isn't worth making.

Edited by pbiggz, 13 July 2023 - 12:15 PM.

[Kotis77]

pbiggz, on 13 July 2023 - 10:09 AM, said:

This is not a solution. This is suicide. Not only would removing groups have little effect on your match quality, it would, in my view, have a seriously negative effect on match quality, mostly because there would be no more matches, because this game would have shut down 3 years ago.

I think the opposite way. Before soup que the matches feel better and you could influence matches with your own actions/skills. I think the main issue now is that matchmaker hasnt been adjusted after adding groups in QP. But i can be wrong or remember wrongfully.

But still i wouldnt go back to soloque only just fix matchmaker.

[crazytimes]

Your team had two absolute potato assaults including an atlas that did 13 damage, plus a lurm heavy that got exactly one alpha out before dying.

Not sure the group is the "problem" here.

[Haipyng]

MM has never been really great. The population is not big enough to properly balance and the safeties that keep you from waiting more than a few minutes for a match trip too often. Devs already have spoken on it. It's been this way since the beginning and no one really questions it anymore.

RickySpanish, on 13 July 2023 - 11:13 AM, said:

^ In-lobby PSR weighted team balancing would be the best fix. It'd require programming though, which apparently the MWO codebase is allergic to. Theoretically the game would look at the players and groups it has found to form a match, and juggle a few around to balance PSR in some way.

^This. I've said that for years. Just have a balance pass in the lobby. Consider PSR and just something simple on tonnage. Call it good. Has to be better than just random team mashing so you get outlier odd ball games, like 7 LRM boats on one side.

[Ihlrath]

It puts 12 players vs. another 12 players. What more do you want, competent matching of skills? Sheesh.

[crazytimes]

Haipyng, on 13 July 2023 - 04:07 PM, said:

^This. I've said that for years. Just have a balance pass in the lobby. Consider PSR and just something simple on tonnage. Call it good. Has to be better than just random team mashing so you get outlier odd ball games, like 7 LRM boats on one side.

So you want skill, and weight, and Loadout to be balanced? Seems easy. Can't imagine why it hasn't been done already. Groups? We balancing them still? What if one group had a LRM mech in it? Just wait for the next group with a LRM mech? What if that player is slightly less skilled then then the other groups LRM player?

I'm happy with random MM. In the long run the potatoes all balance out.

[LordNothing]

one datapoint is not sufficient to describe the current state of the match maker.

you are going to get bad games off peak days/hours because the mm is simply starved for players. valves open, and the mm effectively gives up and just slaps games together with the hand full of available players.

[sycocys]

Almost 2/3s of your team did under 200 damage, wouldn't have mattered who you dropped against.

3 players did = to slight + psr amounts of damage in any average match. - on the losing team.

3 players that are playing average aren't going to carry 9 players that are playing way under.

[Kamiko Kross]

Nine-Ball, on 13 July 2023 - 09:35 AM, said:

4 group drop =/= automatic win

No, not by a long shot. I've seen some shockingly bad play from premades.

[Saved By The Bell]

Just turn off that damn thing and embrace Random God.

[Ignatius Audene]

Only 3 DD. Had 5-6 in some tier 1 matches (on one team not added).

[VeeOt Dragon]

honestly yeah Population is the current biggest factor to MM problems followed second by groups (this is more because the MM doesn't seem to know how to place said groups into a game, i think a group should be calculated as the highest Tier player in the group but thats just me hating to see T1s going into matches with T5s). Groups do actually effect a match more heavily than say 4 single solo players because they can coordinate what mechs they are bringing (the classic is 3 LRM boats with a NARC/TAG light) and tactics where the 4 solos just have to pick a mech and hope for the best. on the flip side i rarely see groups using the in-game chat to coordinate with the rest of the team.

sadly without a bigger population anything that would improve the MM (such as say more Tiers) isn't worth the effort as they would just have to lift the flood gates all the time anyway.

now my first and maybe easiest (not necessarily cheapest) solution would be for PGI to do some damn Advertising. well over 3/4 of the people i talk to on Discord don't even know that MWO exists or what it is. i have managed to get a couple people to try it but the steep learning curve exacerbated by the low population putting T1/T2 players with T5s means that few stay.

MWO has a LOT of flaws some can be fixed some can't. the code is not only old but a patchwork mess so going into anything more than surface lvl is likely impossible at this stage or just to expensive in work hours to take on.

Edited by VeeOt Dragon, 14 July 2023 - 02:43 AM.

[LordNothing]

track a separate psr value when dropping as a group than dropping solo. that should differentiate the skill stackers from the buddy groups quite well. you might be t3 solo and t2 in a group, so if grouping up is the force multiplier people say it is (its more correlative than causative), your higher ranking in groups will be properly represented in the mm and steer you away from the lower tiers.

[KingCobra]

For poor baby Jesus sake everyone needs to stop beating this dead horse topic most players just want to play battles and yes i accept the fact I personally will play anyone from tier 1 to tier5 win or lose. PGI need to do away with tiers altogether its causing more problems than its worth keeping.

And PGI just make 2 drop ques solo play with no groups and a 8v8 or 12v12 group only que keep faction play and events if they want and call it good.

P,S and get rid of the MM as well just group up the players needed for a battle and drop and play its as simple as that. .

Edited by KingCobra, 14 July 2023 - 10:46 AM.