51 replies to this topic

[Revis Volek]

Sandpit, on 23 February 2016 - 12:37 PM, said:

The OP is complaining that he's "saddled" with "bad" players

In other words, "I'm elite and should only be teamed with other "elite" pilots"

I know, and im saying the OP doesn't under stand the system and how large the skill gap is between a 1 and a 3.


His issue, not the games. Carry harder?

[Sandpit]

Revis Volek, on 23 February 2016 - 01:57 PM, said:

I know, and im saying the OP doesn't under stand the system and how large the skill gap is between a 1 and a 3.


His issue, not the games. Carry harder?

Agreed, but most of the people I see posting on the game and issues in the game are doing so from a VERY ignorant position and should honestly do a little research into the subject before they post.

We need a Tier 1 forum section. You have to have x amount of posts, x amount of likes, and x amount of time in the community so I don't have to bother posting alongside the peasants here. I only want elite ideas.

Git Gud or don't post

[Alistair Winter]

Honestly, I think it's partially because of 12v12 and I wish public queue would go back to 8v8. The amount of firepower that a full team of 12 can unload is just atrocious. If a couple of dudes get caught in the open, they're reduced to dust in seconds. And then the ball starts rolling.

I could be wrong about this, but I feel like there's a greater margin for error in 8v8, and it also makes it easier for individual players to carry harder. You're not just a lemming in the great Nascar, with the option to either participate in a horrible strategy or to die alone later.

It's why I'm so excited for 4v4. It'll be a great test.

I agree MM is currently bad, but it used to be so much worse. So, so much worse.

[Bulletsponge0]

don't blame the matchmaker, blame selfish players that don't want to play as a team

[Sandpit]

Alistair Winter, on 23 February 2016 - 02:12 PM, said:

Honestly, I think it's partially because of 12v12 and I wish public queue would go back to 8v8. The amount of firepower that a full team of 12 can unload is just atrocious. If a couple of dudes get caught in the open, they're reduced to dust in seconds. And then the ball starts rolling.

I could be wrong about this, but I feel like there's a greater margin for error in 8v8, and it also makes it easier for individual players to carry harder. You're not just a lemming in the great Nascar, with the option to either participate in a horrible strategy or to die alone later.

It's why I'm so excited for 4v4. It'll be a great test.

I agree MM is currently bad, but it used to be so much worse. So, so much worse.

to me there's a smaller margin for error in 8v8
lose 4 mechs and you're down half your team as opposed to 1/3 of your team.
8v8 lose 1 assault mech on your team and it's harder to overcome in comparison with having 3-5 assault mechs on your team.
Smaller match sizes just decrease the time it takes to get to the end point because there's less mechs to destroy on the map. The problems that still persist today existed when we did 8v8.

[MoonUnitBeta]

Ultimax, on 23 February 2016 - 10:19 AM, said:

I wouldn't care about what T3 guys are doing but Russ has specifically said the MM is still a bit wide for faster matches and feels more people prefer that.

I'd rather wait personally for tighter skill distribution, and this would beneficial for all tiers from a game enjoyment standpoint.

Still can't pass blame on T3 though. Argument is invalid until we actually have Public tier ratings shown on the scoreboard. I'm sure you will see T1 players dip below what you would even expect from a T3 player.
As T3 player, we get good matches we get bad matches, but I'm not naive enough to say that the only reason we have good games is because our team is matched with T1 players, and bad games is when matched with T4 or 5's.... That'd be just stupid of me to say that. So before I spout non-sense like that, I'm going to wait until we actually can see the composition of the team first.

It's also been confirmed by Russ himself that everyone may eventually reach T1. He implied that because people will "get better" but he was not confirming that T1 is reserved for only the best. If everyone is consistently good enough, you'll reach T1. It has marginally very little to do about how good you actually are. Eventually everyone will hit that ceiling if they play long enough, and we know that Tier rating is largely a "games played" metric. How can you even use PSR to gauge the quality of matches when PSR is inherently flawed like that? It's all we got. That's the only reason. But it's hardly accurate enough to throw around accusations that T3's are the problem to your "crappy" games.

PSR is the problem. Tier ratings from 1-5 are the problem. The moment you say "Tier", that's the moment the problem starts.

And if you're confused by that, PSR doesn't identify how people play - who cover fires who, who covers sides to push people back while your team makes a move, who shoots down UAV's, who targets, who does X and Y and Z ad nauseum. Match score is supposed to do that, but that's a joke in itself.

PSR is vague, I don't believe anyone can really brag about being in T1 because it means absolutely nothing. Yeah, you can thank PGI for the not-so-special ELO rating.

Edited by MoonUnitBeta, 23 February 2016 - 02:39 PM.

[Damia Savon]

Monkey Lover, on 23 February 2016 - 10:22 AM, said:

Half the "solo players" in tier2-1 got there by spamming lrms and camp sniping. Good teams will push when they have the numbers and the other team will get focus fired. I dont see rolls stopping any time soon.

The other half got there playing peekaboo with laser vomit alphas. Yawn

[Remains Intact]

Matchmaker is limited in the fact that it doesn't account for when the t1 players in the game are not playing seriously, or not playing strong "carry these bads that got matched with me" mechs. So if one team's allotment of "t1" players are all playing commandos, and the other teams "t1" players are all playing tryhard mechs, then that's a major disparity of contribution.

[Steel Claws]

Funny, most of the games I've been in lately have been fairly close. I haven't seem many rolls. Generally there are less than 4 mechs left on the winning team.

[Bilbo]

Sandpit, on 23 February 2016 - 02:24 PM, said:

to me there's a smaller margin for error in 8v8
lose 4 mechs and you're down half your team as opposed to 1/3 of your team.
8v8 lose 1 assault mech on your team and it's harder to overcome in comparison with having 3-5 assault mechs on your team.
Smaller match sizes just decrease the time it takes to get to the end point because there's less mechs to destroy on the map. The problems that still persist today existed when we did 8v8.

As long as you have some limitation on the number of each class allowed, the loss of an assault isn't that much of a problem. The only real problem with 8v8 was the lack of that restriction.

[Alistair Winter]

Sandpit, on 23 February 2016 - 02:24 PM, said:

to me there's a smaller margin for error in 8v8
lose 4 mechs and you're down half your team as opposed to 1/3 of your team.
8v8 lose 1 assault mech on your team and it's harder to overcome in comparison with having 3-5 assault mechs on your team.
Smaller match sizes just decrease the time it takes to get to the end point because there's less mechs to destroy on the map. The problems that still persist today existed when we did 8v8.

When fewer players are involved, it's more about individual skill and less about teamwork. That's why teamwork is so critical in 12v12 and it's part of the reason why roflstomps are increasingly common as you increase group size. Large, disorganized teams are much more likely to get outmaneuvered and fight ineffectively than smaller groups. In a 2v2 or 4v4 match, the individual player has a better chance to make an impact. It's a lot easier to carry 7 others to victory than to carry 11 others to victory.

Sure, if you lose 4 of 8 mechs, you're screwed. But that's half your force. No matter how many players are involved, losing half your force is a bad thing.

[Sandpit]

Bilbo, on 23 February 2016 - 04:32 PM, said:

As long as you have some limitation on the number of each class allowed, the loss of an assault isn't that much of a problem. The only real problem with 8v8 was the lack of that restriction.

Don't agree. 12 mechs means a longer TTK and it's easier to overcome losses when you have more on your side to begin with.

Now from a personal standpoint, yes I agree that an individual player had more impact on 8v8 but that's simply because there weren't as many mechs to kill. I can kill 4 mechs in an 8v8 and that's half the enemy force. 4 kills in 12v12 and that's only 1/3 even though I just did extremely well on an individual basis.

It doesn't change the way the game handles TTK and attrition though.

Alistair Winter, on 23 February 2016 - 04:39 PM, said:

When fewer players are involved, it's more about individual skill and less about teamwork.

see above lol

and just a sidenote?

That's kinda the point since we're playing a TEAM game and not an individual game

[Wattila]

Can understand where OP is coming from. However, PSR is meant to do two things only: Shorten the queue time, and separate experienced players from beginners. Your skill defines the rate at which you climb, a skill rating it is not. Besides, it's not really possible to find balanced matchups because of all the variables involved (piloting a meta mech can boost your "skill"). Anyway, a lot of the MM problems stem from a small matchmaking pool, nothing you can do about it.

[Bilbo]

Sandpit, on 23 February 2016 - 04:44 PM, said:

Don't agree. 12 mechs means a longer TTK and it's easier to overcome losses when you have more on your side to begin with.

...

Twelve mechs greatly reduces TTK. It may have increased match time, but it certainly didn't increase TTK.

[Revis Volek]

Bilbo, on 23 February 2016 - 04:56 PM, said:

Twelve mechs greatly reduces TTK. It may have increased match time, but it certainly didn't increase TTK.

I have to agree with this...to an extent. In the High level of play i think TTK went down on avg yes, more mech shooting one target means faster kills. But in the avg PUG match i would say it went up because of the lack of focus fire and the fact that there are just 4 more mechs worth of hits points for the avg player to shoot at now.

I dont think we can say 12v12 died ONE thing, I think it changed the scenes of both sides differently...


ANd on that note i would love to see the COMP and high level stuff be 8v8 only and leave the 12 man teams to CW and SOLO play.

Edited by Revis Volek, 23 February 2016 - 05:05 PM.

[Mcgral18]

Alistair Winter, on 23 February 2016 - 04:39 PM, said:

Sure, if you lose 4 of 8 mechs, you're screwed. But that's half your force. No matter how many players are involved, losing half your force is a bad thing.

I can kill 3-4 mechs without much issue, given the opportunity.
Killing 6-8 is a much harder task, while possible, will not happen in half the matches.

Both correspond to the same percentage of the team, but the effort is so much greater in 12 VS 12, and the chants to RNGeesus.


I could carry in a theoretical 8 man Scenario without much effort, but the 12 man becomes Override+shoot as much as possible, because I cannot sustain the firepower (due to heat) or armour to kill a half dozen mechs, because one of our lances faffed off and did nothing contributing to victory.

[Revis Volek]

Mcgral18, on 23 February 2016 - 05:56 PM, said:

I can kill 3-4 mechs without much issue, given the opportunity.
Killing 6-8 is a much harder task, while possible, will not happen in half the matches.

Both correspond to the same percentage of the team, but the effort is so much greater in 12 VS 12, and the chants to RNGeesus.


I could carry in a theoretical 8 man Scenario without much effort, but the 12 man becomes Override+shoot as much as possible, because I cannot sustain the firepower (due to heat) or armour to kill a half dozen mechs, because one of our lances faffed off and did nothing contributing to victory.

3-4 mechs is almost a typical carry match i see out of many from time to time...

5 and more starts stepping into some unknown super leet territory if you do it consistently id say.

[Fleeb the Mad]

It might be a bit of fridge logic, but well matched teams can still have patterns of lopsides victories or defeats.

In a game with both teams having finite resources it's not realistic to expect that those sorts of things can't happen frequently. There's no way to recover once an enemy team has built up a large advantage. In my experience it's actually pretty rare that a game is a turkey shoot for one side from beginning to end.

It doesn't even necessarily matter if you're trading evenly at the onset for damage. I've seen time and again 'stomp' teams that have a whole pile of shredded mechs that just didn't get killed during the fight, because they had the opportunity to stay in the back lines and let other teammates spread damage around. Once a 2 or 3 mech advantage builds then some part of the remaining force can be overwhelmed and it will collapse quickly under focus fire. All those beat up mechs can still shoot and an outnumbered enemy can't focus on them.

After a point the remaining people on the down team just can't survive long enough under fire to make a meaningful difference. In theory with equal teams of competent players, an early game mistake from one or two people would doom one side every time. On maps where terrain and positioning allow those teams to leverage their numerical advantage the collapse would happen quickly. The only chance you generally have of recovering from a 2 or 3 mech disadvantage is generally if the enemy team collectively does something dumb, which will happen less when they're both stuffed with experienced players.

[Sandpit]

Wattila, on 23 February 2016 - 04:51 PM, said:

Shorten the queue time, and separate experienced players from beginners. Your skill defines the rate at which you climb, a skill rating it is not.

That's an interesting take on it. I never thought of it like that before but it's pretty accurate. Skill defines the speed at which you climb the PSR tier system.

Essentially what PGI has said about climbing through the tier system is that everyone has the potential to get to Tier 1, meaning there's not set number of players who can be in Tier 1 like SC does with their upper tiers. As players move down another player moves up into that spot.

With this PSR system there's no need for a player to move down for another to advance into Tier 1. That's what I think Russ means by "everyone can eventually get there"

Now standing between Tier 1 and Tier 5 are a lot of games and other players. The other part of that is yes, all players can potentially make it to Tier 1, but realistically it's not going to happen because individual skill does play into the equation. Very skillful players will climb through those tiers really quickly.

It took me all of about 10-15 games or so to get from Tier 5 to Tier 3.
I've seen some players say it took them more than that to get out of Tier 5.

That's where individual skill comes into play. Even though I strongly believe that KDR and W/L can't be the main or only factors to determine a player's "skill" in this game, it does play into that.

Typically, the longer a player plays a game, the better they get by default. Simple things like eye to hand coordination fo rbeing able to hit keys without looking at the keyboard, knowing the shortcut commands to key actions, simple basic strategies, all from repetition. SO yes, eventually a player is going to move up through the tier system by default and that's what PGI means when they talk about all players being able to hit Tier 1.

Realistically that's not going to happen, but it COULD happen and there's no cap on players being able to advance to Tier 1 and as players gain more experience in the game eventually they're going to hit a new natural skill level that's going to assist their climb until they hit the next plateau.

This can take players months if not years. I'm a decent player, nothing special, I've managed to break top 20 on past leaderboards, and I do ok and I've been chugging through Tier 3 for about.... 4-5 months now. That should give everyone at least an inkling of an idea as to the realistic time frames for some players to advance from a Tier 5 to Tier 1.

That's why I hate it when players try to oversimplify the PSR system. True, it's not a true "ranking" system, but it IS a system that is much deeper and more complex than a simple experience bar.

You're not going to magically hit game number XX and become Tier 1. If it were an exp bar that's what would happen.

Bilbo, on 23 February 2016 - 04:56 PM, said:

Twelve mechs greatly reduces TTK. It may have increased match time, but it certainly didn't increase TTK.

sure it does
The only time it doesn't is if a mech gets focused down by more than 8 enemy mechs.

[MrMadguy]

Kuritaclan, on 23 February 2016 - 01:16 PM, said:

Your system only raise the bar. It isn't a better system than this what we have now. You are still wrong. You are planting your seed everywhere. Humm. And the first attempt doesn't convinced. http://mwomercs.com/...-make-psr-fair/

What do we need to fix in current PSR system at first?
1) It's too W/L dependent. And when you carry or are being carried - your W/L stops representing your personal skill and it messes up matchmaking, cuz it's your personal skill rating, that is used to match you with other players. Look at my recent stats.

200dmg and 600XP per match. Survival rate is just 17%. I usually die fast - I'm first one, who dies, very often. I contribute NOTHING towards victory. But somehow my overall W/L stays the same and I keep advancing:


Also, this happens very often:

Conclusion? I'm being carried. And this can't be stopped due to PSR being too W/L dependant.
2) PSR is too biased towards increasing. It's way too easy to advance and it's way too hard to descend. Your BEST match should be the following, if you want to stabilize:


It's WAY TOO LOW. First - stabilization match score should be increased. Second - you should have equal opportunities to both advance and descend.

My system accomplishes both goals. 325MS - is good MS for average match. May be even a little bit high. Yeah, system still isn't perfect. It still doesn't take into account, that different 'Mechs, you have, may have different performance - having per-'Mech PSR modifier would fix this problem. But it's still better, than current crappy PSR XP bar.

Please note, that XP is doubled due to DoubleXP event on this screenshots!



P.S. Why should I explain it IN EVERY THREAD? My sig has link to initial thread. Why can't you just read it?

Edited by MrMadguy, 24 February 2016 - 12:00 AM.