71 replies to this topic

[Grus]

However...

If you're fighting fairly, you're doing it wrong...

Ill just leave that here.

[RickySpanish]

Why even call them c-bills anymore? Perhaps all currency should be converted to participation tokens and everyone gets a flat amount each month. Also let's solve the problem of Clan vs IS tech by making them all shoot fluffy hugs and rainbows instead. Yay progress! Everything is fair now thanks OP you da man.

[Dee Eight]

The only way to fix the matchmaker is do as ASH suggested in another thread.... hard global reset of all players back to Tier 5, change the scoring formula so that match scores less than 200 on a win don't increase your psr (instead of 100 where it is now), and then wait a few weeks for the active players to move up in tiers to where they should be.

Edited by Dee Eight, 21 November 2018 - 08:29 PM.

[Nightbird]

Anomalocaris, on 21 November 2018 - 07:28 PM, said:

Don't let the perfect be the enemy of the good. AMS is still a reasonable (if broad) indicator of the kind of player you're getting. As a modifier to WLR, it's far and away better than what we use now in the MM. Within the existing stat structure we have in MWO, you need a modifier to WLR because many people don't play enough matches each month to flatten out the impact of overly good or bad teams and random distribution.

If we have W/L, Kills/Match, then AMS is useless. We already have the former two.

[Grus]

Dee Eight, on 21 November 2018 - 08:13 PM, said:

The only way to fix the matchmaker is do as ASH suggested in another thread.... hard global reset of all players back to Tier 5, change the scoring formula so that match scores less than 200 on a win don't increase your psr (instead of 100 where it is now), and then wait a few weeks for the active players to move up in tiers to where they should be.

That just sounds like a LURMagedon 2.0.. but all in all it would work. Although i would add in a slight bias to light and med mechs just due to the steep curve they have to gain MS.

[Khobai]

justcallme A S H, on 21 November 2018 - 07:05 PM, said:

As usual, horribly incorrect.

If the bell curve has the bulk of players in Tier 3 (which it should) there would be no real impact based on how the MM currently looks up/down 2 tier levels to find players.

Im not talking about "tier 3" though. Im talking about the players at the upper end of the bell curve. Obviously the middle of the bell curve wont be affected (at least not during peak hours) because thats where most players are. Thats how a bell curve works. derp.

The players that wont be able to find games are the players at the upper end of the curve because theyre the lowest population group. At the very least they will have considerably longer queue times. Thats a fact. Because the more tiers you create the harder it will be for players at the top end of the bell curve to find games because youre creating additional buckets and each bucket has less players than before.

The other problem is during non-peak hours. When theres barely enough people to fill up even one game. And you want to start adding more buckets? Yeah no. The game just doesnt have the population to support that. Game needs LESS buckets not MORE.

If your solution involves adding more buckets its a bad solution. It doesnt address the population crisis of the game at all and actually makes queue times worse. And theres no guarantee it will even improve game quality, because as you yourself have admitted, theres stomps even in high level competitive play. Longer queue times with no guarantee of improved game quality.... no thanks. bad idea is bad.

Dee Eight, on 21 November 2018 - 08:13 PM, said:

The only way to fix the matchmaker is do as ASH suggested in another thread.... hard global reset of all players back to Tier 5, change the scoring formula so that match scores less than 200 on a win don't increase your psr (instead of 100 where it is now), and then wait a few weeks for the active players to move up in tiers to where they should be

no that idea is terrible. when you have a low population you dont further divide the population up into buckets that cant play eachother. the whole tier system is a fail idea and should be removed. this game doesnt even get new players anymore so theres no need for a tier system to ease new players into the game.

instead they should add some gamemodes that are more forgiving and less prone to snowballing and dont punish teams as much for having bad players. the gamemodes are a huge part of the problem; if weve learned anything thus far its that a lot of players cant handle how brutally unforgiving skirmish is. Those players will never "get gud", its not going to happen.

thats exactly why other games like war thunder and world of tanks dont make one-life gamemodes their main gamemodes. Its one of the big reasons those games are more commercially successful too because they appeal more to casual gamers. MWO did it wrong thats why its not doing well.

Edited by Khobai, 21 November 2018 - 09:22 PM.

[Vxheous]

Khobai, on 21 November 2018 - 09:00 PM, said:

Im not talking about "tier 3" though. Im talking about the players at the upper end of the bell curve. Obviously the middle of the bell curve wont be affected (at least not during peak hours) because thats where most players are. Thats how a bell curve works. derp.

The players that wont be able to find games are the players at the upper end of the curve because theyre the lowest population group. At the very least they will have considerably longer queue times. Thats a fact. Because the more tiers you create the harder it will be for players at the top end of the bell curve to find games because youre creating additional buckets and each bucket has less players than before.

The other problem is during non-peak hours. When theres barely enough people to fill up even one game. And you want to start adding more buckets? Yeah no. The game just doesnt have the population to support that. Game needs LESS buckets not MORE.

If your solution involves adding more buckets its a bad solution. It doesnt address the population crisis of the game at all and actually makes queue times worse. And theres no guarantee it will even improve game quality, because as you yourself have admitted, theres stomps even in high level competitive play. Longer queue times with no guarantee of improved game quality.... no thanks. bad idea is bad.

The queues won't be affected, since you can still have Tier 1 through 3 matching into games. The difference though is that if you put 1 tier 1, 3 tier 2, and 8 tier 3 on each side of the match maker, at least you guarantee that both side's tier 1's are closer in skill level than say, a 99% tier 1, and a 43% tier 1.

[Khobai]

Vxheous, on 21 November 2018 - 09:11 PM, said:

The queues won't be affected, since you can still have Tier 1 through 3 matching into games. The difference though is that if you put 1 tier 1, 3 tier 2, and 8 tier 3 on each side of the match maker, at least you guarantee that both side's tier 1's are closer in skill level than say, a 99% tier 1, and a 43% tier 1.

If you start moving Tier 3s to Tier 4 and 5 of course its going to affect the queues.

If the game pulls players from Tier 1-3 to be in the same game, and suddenly theres less Tier 3 players because a lot of them have been bumped down to Tier 4-5 theres going to be less players available in the Tier 1-3 queue. So the queue times will definitely be longer. And since stomps happen in top level play too, theres no guarantee of the game quality being improved. Stomps happen; theyre part of the game because of how flawed skirmish is. Moving players around to different tiers isnt going to get rid of stomps.

Again creating more buckets is not the answer. Thats just not what you do when the game has a population crisis. And the game is still losing players... if PGI wants to save their game they need to fix the reasons theyre hemorrhaging players like crazy. Thats more important than screwing around with a broken tier system. The Tier system isnt why people are quitting the game in droves. Its the repeated nerfs, unresolved weapon balance issues, lack of new content, failure to fix bad gamemodes, and the utter stagnation of the game thats the problem.

Edited by Khobai, 21 November 2018 - 09:39 PM.

[Vxheous]

Khobai, on 21 November 2018 - 09:24 PM, said:

If you start moving Tier 3s to Tier 4 and 5 of course its going to affect the queues.

If the game pulls players from Tier 1-3 to be in the same game, and suddenly theres less Tier 3 players because a lot of them have been bumped down to Tier 4-5 theres going to be less players available in the queue.

No, because tier 3's will stay as tier 3's, same with tier 2's. It's the oversaturated tier 1's that'll get thinned back out to tier 2 and 3 (even 4 and 5 if they're actually that bad)

However, if that means tier 1, 2 and 3s have actual player populations that have similar skill level, the matchmaker putting equal amounts on each side should also mean better matches. Not like right now where there's 8 tier 1s, 2 tier 2s, 2 tier 3's on each side, but team 1 in skill level actually has 2 tier 1s, 3 tier 2s, 6 tier 3's and a tier 5, while team 2 has 4 tier 1s, 2 tier 2, 4 tier 3's, and 2 tier 4's. The current system thinks that it's a equal match (8 tier 1s 2 tier 2s, 2 tier 3s each side) but in actual skill reality, team 2 is going to rolfstomp team 1 12-2.

Edited by Vxheous, 21 November 2018 - 09:34 PM.

[El Bandito]

Nightbird, on 21 November 2018 - 06:32 PM, said:

https://www.dictionary.com/browse/fix

fix:
1. to repair; mend.
2. to put in order or in good condition; adjust or arrrange:
3 to make fast, firm, or stable.

The MM is broken, and if corrective measures are taken, it will be fixed.

And again I disagree. PGI will not be able to fix it with such population. To cue your dictionary, the MM will not be stable, or firm, because to try that in current population will mean 10+ minutes of wait time per QP match--which is longer than the match itself. And that's during prime time. At other times it could take more than 20 minutes.

PGI can only improve it so that the MM is better than before (but not fixed) so that it will still leave enough leeway to allow half-decent wait time. Cause as flawed as MM is, I'd rather play the game with its flaws rather than looking at the "searching" text, and I think many other players agree.

Edited by El Bandito, 21 November 2018 - 09:37 PM.

[Nightbird]

El Bandito, on 21 November 2018 - 09:34 PM, said:

And again I disagree. PGI will not be able to fix it with such population. To cue your dictionary, the MM will not be stable, or firm, because to try that in current population will mean 10+ minutes of wait time per QP match--which is longer than the match itself. And that's during prime time. At other times it could take more than 20 minutes.

PGI can only improve it so that the MM is better than before (but not fixed) so that it will still leave enough leeway to allow half-decent wait time.

Already answered on page 1. 0 increase in queue time to fix MM

Nightbird, on 21 November 2018 - 09:11 AM, said:

You're saying that the population doesn't allow balanced matches, so the match maker can't be fixed. Even today, with any 24 players, you can create a balanced match. It wouldn't mean no stomps, just that the chance of a lopsided result is minimized.

Just do a simple thought exercise, do you believe that if you sort 24 players by skill and mech selection and map selection and place an accurate value on their expected contribution, then put the 1 best player and 11 worst players on one team, will this team lose against the team with #2-13th players? If the answer is yes, this team will lose, then you can shuffle players until the two teams are balanced. The problem has always been the MM used stupid metrics like match score and tiers to create teams, so yes, the MM is fixable and can create balanced teams with today's population. You just need to hire someone who can math. I'd do it myself except PGI doesn't share enough data.

[El Bandito]

Nightbird, on 21 November 2018 - 09:36 PM, said:

Already answered on page 1. 0 increase in queue time to fix MM

And I answered in my edited post. With current population, any real fixes to MM will mean noticeably longer wait time on an already long wait time. So yes, improvements should be made--and I am not against it, but it must keep the wait time in mind.

Quote

Cause as flawed as MM is, I'd rather play the game with its flaws rather than looking at the "searching" text, and I think many other players agree.

Edited by El Bandito, 21 November 2018 - 09:53 PM.

[Nightbird]

El Bandito, on 21 November 2018 - 09:40 PM, said:

And I answered in my edited post. With current population, any real fixes to MM will mean noticeably longer wait time on an already long wait time. So yes, improvements should be made--and I am not against it, but it must keep the wait time in mind.

If you read my post, you'd realize that it's completely false that "any real fixes to MM will mean noticeably longer wait time"

[El Bandito]

Nightbird, on 21 November 2018 - 09:57 PM, said:

If you read my post, you'd realize that it's completely false that "any real fixes to MM will mean noticeably longer wait time"

I read it and I still disagree that using different calculation to better shuffle the available 24 players is gonna fix the issue, because that will still mean scrubs will be matched with/against those who are much better than them, which will result in dissatisfaction on both ends. Any real fix would require the player to play with/against 23 others who have similar skill level--something that is not possible with current population, or 10x that, without affecting the match wait time noticeably.

Reverting to 8v8 is one of the easy ways to reduce MM disparity in QP though, so that could be combined with different MM algorithm for making a real impact while keeping wait time roughly the same.

Edited by El Bandito, 21 November 2018 - 10:22 PM.

[Nightbird]

El Bandito, on 21 November 2018 - 10:17 PM, said:

I read it and I still disagree that using different calculation to better shuffle the available 24 players is gonna fix the issue, because that will still mean scrubs will be matched with/against those who are much better than them, which will result in dissatisfaction on both ends. Any real fix would require the player to play with/against 23 others who have similar skill level--something that is not possible with current population, or 10x that, without affecting the match wait time noticeably.

Reverting to 8v8 is one of the easy ways to reduce MM disparity in QP though, so that could be combined with different MM algorithm for making a real impact while keeping wait time roughly the same.

That's not a MM fix you want, but a population fix. The best that any MM can do is to create two teams each with a 50/50 chance of winning with any 24 players. The current MM does not do that. If you give the MM 48 players instead of 24, it can further separate out the best and worst into 2 groups of 24 before making teams. If you give the MM more players you can do further divisions with a narrower band of skill within each group of 24. That is a proper MM.

A functional MM one of the fundamental features that lead to a good population. That's why I said asking for more pop to fix the matchmaker is putting the cart before the horse.

Edited by Nightbird, 21 November 2018 - 11:02 PM.

[justcallme A S H]

El Bandito, on 21 November 2018 - 10:17 PM, said:

Reverting to 8v8 is one of the easy ways to reduce MM disparity in QP though, so that could be combined with different MM algorithm for making a real impact while keeping wait time roughly the same.

Yeah go 8v8 with a totally stuffed MM letting better players assert even high levels of dominance... Cause that'll work.

As NB said - that's a population fix not a MM. Especially as it'll exceccerbate the issue further.

[RJF Volkodav]

This type of matchmaking is just BAD - https://mwomercs.com...ost__p__6172076

Each stomp game i've analyzed there were a possibility to switch already matched players between teams within weight classes (assault for assault etc) to get 5% difference of ALL the stats between teams - WLR, KDR, MS. PGI dont even need to throw tier system. Just need to add skill balancing. But reworking tiers could potentially help even more.

Edited by RJF Volkodav, 22 November 2018 - 01:23 AM.

[El Bandito]

Nightbird, on 21 November 2018 - 10:55 PM, said:

That's not a MM fix you want, but a population fix. The best that any MM can do is to create two teams each with a 50/50 chance of winning with any 24 players.

The best that any MM can do is to create two teams each with a 50/50 chance of winning, plus making sure all the participating players are of similar skill level. That is the true fix.

Such true fix is, again, not possible with current population, unless one is willing to wait for a looong time.

Edited by El Bandito, 22 November 2018 - 03:16 AM.

[justcallme A S H]

RJF Volkodav, on 22 November 2018 - 01:21 AM, said:

This type of matchmaking is just BAD - https://mwomercs.com...ost__p__6172076

Each stomp game i've analyzed there were a possibility to switch already matched players between teams within weight classes (assault for assault etc) to get 5% difference of ALL the stats between teams - WLR, KDR, MS. PGI dont even need to throw tier system. Just need to add skill balancing. But reworking tiers could potentially help even more.

Yeah and as if there is no way to balance the 24 players before a lobby is formed to at least TRY make it more even

[Khobai]

Vxheous, on 21 November 2018 - 09:28 PM, said:

No, because tier 3's will stay as tier 3's, same with tier 2's. It's the oversaturated tier 1's that'll get thinned back out to tier 2 and 3 (even 4 and 5 if they're actually that bad)

what youre saying makes no sense.

if theres players that dont belong in tier 1 then theres also certainly players that dont belong in tier 2 or tier 3.

if you take players out of tier 1-3 and put them in tier 4-5 instead, theres going to be less overall players in tiers 1-3. Less overall players = longer queue times.

by distributing players more evenly among the 5 tiers you will in fact be increasing queue times, because not all 5 tiers can play against eachother. thats common sense.

And once again, dividing players into more buckets is not how you address a population crisis. We need less buckets not more. Especially since you havent proven that dividing people into tiers will even improve game quality. Its pure conjecture on your part with zero evidence to back it up. Stomps still happen ALL THE TIME even in high level play, so too many players being in tier 1 is obviously not the reason stomps are occuring. Stomps mostly occur because skirmish is a really bad gamemode due to the snowballing effect that occurs once one team gets down 2-3 mechs. Skirmish gamemodes are the biggest problem in quickplay.

Edited by Khobai, 22 November 2018 - 10:32 AM.