44 replies to this topic

[Mystere]

Eisenhorne, on 08 August 2018 - 12:42 PM, said:

The correct way to avoid this is to just not bring brawlers to quick play since they're less useful like half the time because they're so map dependent ...

That's really so funny because I have seen many times people say the exact opposite, that LRMS are "less useful like half the time because they're so map dependent ...".

Edited by Mystere, 08 August 2018 - 01:46 PM.

[Eisenhorne]

Mystere, on 08 August 2018 - 01:46 PM, said:

That's really so funny because I have seen many times people say the exact opposite, that LRMS are "less useful like half the time because they're so map dependent ...".

They're both true statements though. LRM's are less useful on brawl-focused maps, brawl builds are less useful on long range maps. That makes the "meta" builds ones that can work at long range AND close range decently well... AKA mid-range traders. Gauss rifles, er mediums, large lasers, ac 5's, ac 10's, etc. You know, things you see on Warhammers, Mad Cats, Hellbringers, and other "meta" chassis.

[Mystere]

Eisenhorne, on 08 August 2018 - 01:58 PM, said:

They're both true statements though. LRM's are less useful on brawl-focused maps, brawl builds are less useful on long range maps. That makes the "meta" builds ones that can work at long range AND close range decently well... AKA mid-range traders. Gauss rifles, er mediums, large lasers, ac 5's, ac 10's, etc. You know, things you see on Warhammers, Mad Cats, Hellbringers, and other "meta" chassis.

Then my solution is to triple the number of maps, so that we have around 1/3 as highly claustrophobic urban environments with the scale and general layout of Hong Kong, and 1/3 with the scale and general layout of the desert landscape near Tobruk.

Edited by Mystere, 08 August 2018 - 08:56 PM.

[General Solo]

Some people seek the easy wae rather than having to think how to counter their disadvantage on certain maps.
Others relish the challenge of thinking how to counter their disadvantage on certain maps.

MM needs to put each group in different skill pools.

Edited by OZHomerOZ, 08 August 2018 - 08:43 PM.

[Mystere]

OZHomerOZ, on 08 August 2018 - 08:42 PM, said:

Some people seek the easy wae rather than having to think how to counter their disadvantage on certain maps.
Others relish the challenge of thinking how to counter their disadvantage on certain maps.

MM needs to put each group in different skill pools.

I disagree on the last part. Those who successfully adapt will thrive; those who do not deserve to be left for dead on the side of a ditch.

Edited by Mystere, 08 August 2018 - 08:59 PM.

[Weeny Machine]

The problem is PSR. In its current state everyone is T1 or if you are really reaaaally bad T2 at one point. How should the matchmaker sort teams that way when there are "real" T1 guys and "potato" T1 one guys.

I have had matches were I recognized names of really good players but I recognized not one of my team. You can guess what happened next.

Sidenote: I do not consider myself T1, yet I have been T1 for maybe 2 years already.

[General Solo]

Mystere, on 08 August 2018 - 08:58 PM, said:

I disagree on the last part. Those who successfully adapt will thrive; those who do not deserve to be left for dead on the side of a ditch.

Bit harsh
but I can understand.
Problem is without the separation the dead rise to plague you in future games.

[General Solo]

Bush Hopper, on 09 August 2018 - 12:10 AM, said:

..... How should the matchmaker sort teams that way when there are "real" T1 guys and "potato" T1 one guys.....

Like this

OZHomerOZ, on 08 August 2018 - 08:51 AM, said:

I agree MM needs fixing its upward bias before weapons are balanced. How is the part I disagree.

I would make a small tweak where the conditions for PSR movement on a loss are mirrored for wins.
This would provide equal opporunity to increase and decrease PSR.
Currently their are more PSR increase conditions than PSR decrease conditions, they should be the same.

For example: On a win the carriers (strong performers) carry (help) the carried (weaker performers) due to the current PSR conditions for wins, both parties increase PSR effectively placing them in the same skill pool.

The people with map troubles get carried and thus go down PSR on a win, effectively placing them in different skill pool from the carriers if PSR change conditions for wins and losses are alike.

Problem currently they are not the same

Edited by OZHomerOZ, 09 August 2018 - 12:48 AM.

[MrMadguy]

OZHomerOZ, on 09 August 2018 - 12:45 AM, said:

Like this

The people with map troubles get carried and thus go down PSR on a win, effectively placing them in different skill pool from the carriers if PSR change conditions for wins and losses are alike.

Problem currently they are not the same

I guess, initial bias is implemented to reflect simple fact, that on average when you lose - you have lesser Match Score, than when you win. Devs do it to avoid double dependence. So, nothing bad here.



But my researches show, that it's not the only problem with PSR. Initially I thought, that it was that problem, cuz I wasn't able to go down even with extremely terrible stats, but actually it's not. Rating changes are at least 2x biased towards increasing - that's the real problem.

Edited by MrMadguy, 09 August 2018 - 02:16 AM.

[General Solo]

MrMadguy, on 15 January 2018 - 12:17 AM, said:

Well I did say mirror the conditions

The Current problem is the fact that out of 10 possible PSR Movement conditions, as shown in your graphic for win and loss,
6 are PSR increase conditions
2 are no PSR movement
and only 2 PSR decrease conditions

So of course its bias upward to my untrained eye as their just more go up in PSR condiitions.

EZ

So just mirror them already and have equal number of PSR increase conditions as PSR decrease conditions
IMO

[Oldbob10025]

Papaspud, on 07 August 2018 - 11:19 AM, said:

The game has deteriorated into a hot mess- worse than it has ever been. It is pretty clear there is no longer any attempt at matchmaking, I have seen the worst players in every match for the last week, really no fun. I keep trying everyday to play a few matches, but when you are on or against a team where the first 3-4 guys just run out and die before most of the team has even gotten to take a shot, it just makes me throw up my hands and say nope, not fun anymore.

I have loved this game since beta, but with the lack of matchmaking, playing the same 2-3 maps over and over- and over and over- and the game just running like crap- I am getting to where I have just about lost any desire to play, takes about 1-2 matches before I just shut off and call it a day.

Looks like the end days are here sadly, because I don't see anyway for them to reverse this trend with a 6 year old game- most of the older players have left and the new ones, well if you can't say anything nice..........

This is not a elo system where its based on population & time played not ELO. Once you get to know that you would understand why things they way they are and have more fun at the game. As far as calling out players for doing stupid moves according to you, well the game is build for all and not just you. They thought what ever reason to goto point B etc and you cant stop that because you dont control them. If you want a controlled environment then Faction Warfare would be your thing with your own super duper team that you hand pick and then you wont have that problem.

Yes the Matchmaker Systems is broken, and not much we can do about that and I dont think PGI is going to fix that anytime soon with Faction Warfare things coming. If your not having fun then the best thing might be to play another game. Why would you play a game and not have fun at it?

I hope that helps you to understand some things... Later and remember have fun in the game..

[MrMadguy]

OZHomerOZ, on 09 August 2018 - 03:06 AM, said:

Well I did say mirror the conditions

The Current problem is the fact that out of 10 possible PSR Movement conditions, as shown in your graphic for win and loss,
6 are PSR increase conditions
2 are no PSR movement
and only 2 PSR decrease conditions

So of course its bias upward to my untrained eye as their just more go up in PSR condiitions.

EZ

So just mirror them already and have equal number of PSR increase conditions as PSR decrease conditions
IMO

I'm not 100% sure now. Do we have some victory match score bonuses, that aren't tied to personal performance?

[Dragonporn]

MrMadguy, on 09 August 2018 - 03:30 AM, said:

I'm not 100% sure now. Do we have some victory match score bonuses, that aren't tied to personal performance?

I'm pretty sure we have. As to what Homer said, that would definitely work, and I'm all for it, as it was already been
suggested countless times. But for it to make a difference, PSR must be reset for all players starting in T5 again.

[General Solo]

MrMadguy, on 09 August 2018 - 03:30 AM, said:

I'm not 100% sure now. Do we have some victory match score bonuses, that aren't tied to personal performance?

I 'll try to explain it with the graphic below which I obtained by butchering your graphic in photochop. soz



In my mind MM worked pretty good when it was first released due to the fact that the good players were in T1 already or they got their shortly (Months afterwards)

Now the problems is that after all this time (years), every man and his dog has also eventually arrived at T1 due to the upward bias shown in the graphic above.

An analagee would be traveling to mars, some use rockets, others walk (I dunno rainbow bridge maybe), they all get their in the end. Just the better ones get their alot fasta.

[MrMadguy]

OZHomerOZ, on 09 August 2018 - 03:10 PM, said:

I 'll try to explain it with the graphic below which I obtained by butchering your graphic in photochop. soz



In my mind MM worked pretty good when it was first released due to the fact that the good players were in T1 already or they got their shortly (Months afterwards)

Now the problems is that after all this time (years), every man and his dog has also eventually arrived at T1 due to the upward bias shown in the graphic above.

An analagee would be traveling to mars, some use rockets, others walk (I dunno rainbow bridge maybe), they all get their in the end. Just the better ones get their alot fasta.

What I try to say, is that on loss we usually have lesser Match Score due to several natural reasons, that don't mean lower personal performance. When you lose, you have lesser amount of kills, die earlier, deal lesser damage, even if your enemies are perfectly balanced and have exactly the same skill, as you. And you should understand, that you just can't look at some stats out of context. For example you just can't say, that with K/D = 1 and average MS = 250 player will most likely go up, so PSR is biased towards increasing. No, you should look at results of all possible matches as whole. That's, what my simulator does. As you can see on a first picture - average PSR change is 0 for current system, if WinAvgMS = LossAvgMS + 100.

[General Solo]

I propose getting a lot of screen shots say over 100 where people got carried.
Then get EXPERT players to grade the losing side players by Carrier, Pull their weight and Got Carried.
Grab the match scores of each group to tune your numbers.

Also you said "What I try to say, is that on loss we usually have lesser Match Score due to several natural reasons, that don't mean lower personal performance."

I say who cares about the losses as the imbalance is caused by the win conditions allowing folks to get carried and increase in PSR same as the people that carried them.

Instead of the carriers gaining PSR and the carried going down PSR.

Edited by OZHomerOZ, 09 August 2018 - 11:47 PM.

[Tetatae Squawkins]

I have no idea why the tier system works the way it does. You basically have to be AFK for half of your games to see any significant downward movement.

[General Solo]

Probably to appease the I should be in T1 coz Im a da best, stats dont mean nuthing crowd

[Dogstar]

MrMadguy, on 08 August 2018 - 02:41 AM, said:

Wrong. Matchmaker should measure, how player actually plays - not how he is supposed or assumed to play. I play since Open Beta and what? Am I top-Tier player due to that fact? No, I'm actually below average. I'm supposed to be playing in Tier 2 now, while in reality I belong to Tier 4.

I'm not wrong it's that you don't understand. Your saying the matchmaker _should_ do something whereas I'm telling you what it _actually_ does.

MWO does not have a skill based matchmaker

[Tarl Cabot]

Dogstar, on 11 August 2018 - 11:41 AM, said:

I'm not wrong it's that you don't understand. Your saying the matchmaker _should_ do something whereas I'm telling you what it _actually_ does.

MWO does not have a skill based matchmaker

Yeap, MWO has an experience based matchmaker, which most can hit the minimum threshold to move up, some slower than others... But MrMadguy knows that, he may have misinterpreted your intentions with the post.

Now if the Jarl's list could be setup to winnow out the seasons, ie calculate the last 5 seasons, then MrMadguy average stats would not be as show a lower average status at where he may have actually started moving up.

Edited by Tarl Cabot, 11 August 2018 - 04:08 PM.