55 replies to this topic

[AwDangit]

MadBede, on 13 January 2022 - 06:10 PM, said:

I love Mechs but i hate BS this game become. Time to move away from mess MWO become.
-snipers in every match, you cant move in open, you cant peak
-extreme damage alpha everyone can bring to the match, zero room for error. Or you swarm with team or you get rekt asap
-12-0 0-12 all day long, no balance in teams
-lights running rampart
-assault only camps, they move they get fkd by snipers, poptarts loaded to brim with heavy weapons
-new map just for fast mechs, you can sneak around and farm from behind, great idea, give lights more fun, spit on slow mechs
-models hit boxes from hell, one mech can eat dumb amount of punishment (all small designs) biggies no matter where to shot you rekt them, hihi fun to play King Crab or Marauder or Anihillator , 100 tomes of joke.
-MG no heat, no cd on reload just keep finger om trigger, sure sure
-LRM turned to useless poo, try go with LRM, someone will MG your mech in 5 sec , no problem
-ECM, Stealth spam one one team when other got none..fun 12-0 inc
PGI went for mech packs to farm cash, balance went to hell, game is just boring.
Hope one day Devs will open eyes and do something.

Join an active unit.

Log out and back in.

Stop letting a game ruin your happiness.

Or just leave.

[dubstep albatross]

Apteko, on 16 January 2022 - 05:39 AM, said:

What I didn't think of back then is that peak of resulting curve can be shifted from 6-12 and there is no single reason for curve to be symmetrical - after first few pickoffs winning team has pretty damn good chance to crush their opponents.
And I guess, this proportion between roflstomps and close games can be explained with the fact that basically any random imperfection lowers the chance of game being close. And there is nothing that increases this chance if we assume that matchmaking works fine and tries to create balanced initial state, which can't be improved by random factors/relatively small player base/e.t.c. e.t.c. (due to it being balanced), but always can get worse.

I have put some thought into this overall topic and I agree. I would say that there is a certain gravity to runaway games that the routes to a "close" game are comparatively few. I attribute this to fundamental game mechanics, the stochastic nature of the match, and myriad unaccounted for variables in the matchmaking process.

It is interesting to think about "the first few pickoffs". How did the match arrive in that state? Is it randomness or is it a result of skill disparity? Is it a symptom of the underlying impending defeat (i.e., the defeat is going to happen because of other variables and this is how it initiates) or is it a contributing cause (i.e., because the one team now has less armor/firepower/eyes, the defeat has been initiated)? Naturally things are more grey than black-and-white and, intuitively, each match may have a different mix. Still, it is interesting to consider.

[Horseman]

dubstep albatross, on 16 January 2022 - 08:19 AM, said:

It is interesting to think about "the first few pickoffs". How did the match arrive in that state? Is it randomness or is it a result of skill disparity? Is it a symptom of the underlying impending defeat (i.e., the defeat is going to happen because of other variables and this is how it initiates) or is it a contributing cause (i.e., because the one team now has less armor/firepower/eyes, the defeat has been initiated)? Naturally things are more grey than black-and-white and, intuitively, each match may have a different mix. Still, it is interesting to consider.

Chicken and egg. The kill imbalance *can* happen because one team is focusing fire better, but when it happens the team that loses more mechs has less firepower to remove enemy assets with.

[Humpday]

I haven't played in years looks like a bunch of cool things happened with quirks. Though I very likely tier 100 skill level by now lol.

[dubstep albatross]

Horseman, on 17 January 2022 - 04:54 AM, said:

Chicken and egg. The kill imbalance *can* happen because one team is focusing fire better, but when it happens the team that loses more mechs has less firepower to remove enemy assets with.

I agree that there is a clear synergy to those two intertwined aspects. I also think that the mix of those two aspects can be vastly different game-to-game. I would say that often there is no clear way to know, in any given match, what came first. Looking at available data can sometimes give hints. Too bad there is not more match data available.

I have considered recording games and using OCR to extract data regarding the sequence/flow of battle. First on that list would to create a timeline of kill notifications.

[LordNothing]

you know around these parts we have a saying, if you don't like the weather, wait 15 minutes. mwo is a lot like that. when pgi spoon-feeds the community with occasional free stuff event, or a patch, or any new content no matter how meh it is, the community inevitably has this huge kneejerk reaction where they act like total idiots over the tiniest of change. it creates short periods of confusion where everyone forgets how to play the game. wait a week and things go back to the normal stagnation. just take a break when the gameplay is derp.

[DaZur]

Horseman, on 17 January 2022 - 04:54 AM, said:

Chicken and egg. The kill imbalance *can* happen because one team is focusing fire better, but when it happens the team that loses more mechs has less firepower to remove enemy assets with.

dubstep albatross, on 17 January 2022 - 08:50 AM, said:

I agree that there is a clear synergy to those two intertwined aspects. I also think that the mix of those two aspects can be vastly different game-to-game. I would say that often there is no clear way to know, in any given match, what came first. Looking at available data can sometimes give hints. Too bad there is not more match data available.

My experience is the magic number is three... It takes roughly a differential of three mechs to shift the composite team strength enough to tip the preverbal scales. That said, it's not hard and fast and I've been on and witnessed cherry-picked matches where this doesn't hold up.

Usually when I see my team is up by three, I know I can play a lot looser and the team generally pushes harder resulting in team wins.

Edited by DaZur, 17 January 2022 - 10:01 AM.

[Relishcakes]

Apteko, on 16 January 2022 - 05:39 AM, said:

> I built another bar chart

Oh, thanks, that is awesome and informative.
What I wanted to point out in my first comment was normal distribution of results so even with perfect matchmaking most matches can't be close ones. And "quarter" was taken straight from "feels about right" zone, so it is not very surprising that my approximation happen to be quite imprecise.

What I didn't think of back then is that peak of resulting curve can be shifted from 6-12 and there is no single reason for curve to be symmetrical - after first few pickoffs winning team has pretty damn good chance to crush their opponents.
And I guess, this proportion between roflstomps and close games can be explained with the fact that basically any random imperfection lowers the chance of game being close. And there is nothing that increases this chance if we assume that matchmaking works fine and tries to create balanced initial state, which can't be improved by random factors/relatively small player base/e.t.c. e.t.c. (due to it being balanced), but always can get worse.

Entirely serious question, Whats up with this community and making charts/graphs? i see more charts here than anywhere else.

[CFC Conky]

Relishcakes, on 18 January 2022 - 03:15 PM, said:

Entirely serious question, Whats up with this community and making charts/graphs? i see more charts here than anywhere else.

I had a chart that explains why but I can’t find it at the moment.
Good hunting,
CFC Conky

[justcallme A S H]

The funniest part of the "oh but mah 12-0 every match tho"

Even in matches that are pretty close skill wise, complete wipes will happen. Just have a look at how many comp matches in 8v8 have been complete 8-0s over the years with opponents of very similar skill. It is very common.

And people think QP is some magical beast and it won't happen there? Not to mention as dubstep showed, it's blown out of proportion (again, surprise surprise )

People seriously need to get a grip on reality. Stop blowing things out of proportion. Record and provide some data - As others do - then you will see you are purely overreacting and completely incorrectly at that.

Edited by justcallme A S H, 18 January 2022 - 05:50 PM.

[LordNothing]

people only seem to remember the bad matches, when it goes the other way the match is equally forgettable, but because you won you got nothing to complain about on the forums. there are a lot of games in the 4-12 where at least its not a stomp. the fun games are the ones that are hard fought by both sides. the worst games are the ones with < 12 kills combined total of both teams, because that meant everyone sat on their hands for 15 minutes or it ended in a troll camp 0 to 0. that's what a bad match really looks like.

Edited by LordNothing, 18 January 2022 - 06:03 PM.

[Vexistential]

LordNothing, on 18 January 2022 - 06:02 PM, said:

people only seem to remember the bad matches, when it goes the other way the match is equally forgettable, but because you won you got nothing to complain about on the forums. there are a lot of games in the 4-12 where at least its not a stomp. the fun games are the ones that are hard fought by both sides. the worst games are the ones with < 12 kills combined total of both teams, because that meant everyone sat on their hands for 15 minutes or it ended in a troll camp 0 to 0. that's what a bad match really looks like.

That is confimation bias. Confirmation bias is certain to occur. But when we have a matchmaker that is designed to create a bias by separating tiers, and that bias can be crossed, it is even more likely to be noticed. Not that I'm saying we shouldn't have such a thing like a matchmaker that just tries to get matches made, cause it does a good job, all things considered.

It's just that it should be acknowledged that tier thresholds, and the means to cross over those thresholds are difficult for players to understand. What can be expected from one game to another can vary wildly. Especially if you are bouncing from tier 3 and tier 4 (Like myself). Matchmaker is certainly working! But, if I could figure out how to only play against the crappy half of my opponents... wow, that would likely benefit my stats.

[B O O M E R A N G]

I would recommend playing an ECM Atlas and trying to make those brawls happen baby

[Bassault]

MadBede, on 13 January 2022 - 06:10 PM, said:

-snipers in every match, you cant move in open, you cant peak

Stop rotating into sniper sightlines. Stop NASCAR'ing. Know the sniper spots so you can avoid fighting there.

MadBede, on 13 January 2022 - 06:10 PM, said:

-extreme damage alpha everyone can bring to the match, zero room for error. Or you swarm with team or you get rekt asap

Torso twist or push them. High alpha mechs have poor dps. So don't play peek games with them or push them and torso twist.

MadBede, on 13 January 2022 - 06:10 PM, said:

-12-0 0-12 all day long, no balance in teams

At least it ends quickly. Try grouping with others and coordinate with eachother to prevent stomps.

MadBede, on 13 January 2022 - 06:10 PM, said:

-lights running rampart

Shoot them, they die when you shoot them.

MadBede, on 13 January 2022 - 06:10 PM, said:

-assault only camps, they move they get fkd by snipers, poptarts loaded to brim with heavy weapons

Sometimes your teammates are dumb, just like your opponents as well.

MadBede, on 13 January 2022 - 06:10 PM, said:

-new map just for fast mechs, you can sneak around and farm from behind, great idea, give lights more fun, spit on slow mechs

Somewhat true, the new map is too forgiving for jumpjet mechs. Just stay on the high ground though, you will have much success.

MadBede, on 13 January 2022 - 06:10 PM, said:

-models hit boxes from hell, one mech can eat dumb amount of punishment (all small designs) biggies no matter where to shot you rekt them, hihi fun to play King Crab or Marauder or Anihillator , 100 tomes of joke.

The annihilator has amazing hitboxes and survivability and the Marauder is pretty good too, they are 100 tons of greatness and they are very viable. I pilot Annihilators in quickplay and I do it in solo most of the time, and I manage to win a lot more than I lose. I don't even snipe either, I play ac10 lppc, hgauss medium pulses, or quad lbx10.

MadBede, on 13 January 2022 - 06:10 PM, said:

-LRM turned to useless poo, try go with LRM, someone will MG your mech in 5 sec , no problem

LRMs are effective if you know how to use them, look at DATA in his LRM fafnir dishing out 1k or more damage per game. If someone, I presume a light, is machinegunning you in 5 seconds, you need to be with your team and maybe pop a UAV or ask for help in your microphone.

MadBede, on 13 January 2022 - 06:10 PM, said:

-ECM, Stealth spam one one team when other got none..fun 12-0 inc

Stealth armor sucks. ECM is only a problem if you are blind or rely on LRMs. Even then, as soon as your team has a lock or if you have tag, your LRMs will work.

MadBede, on 13 January 2022 - 06:10 PM, said:

PGI went for mech packs to farm cash, balance went to hell, game is just boring.

The quickplay meta is as diverse as it ever has been.

Sounds like you're just upset because you don't understand how to play the game and overcome the obstacles infront of you.

[ScrapIron Prime]

Well... to be fair, the last time he logged on was in writing the post at the beginning of this topic. So point to him for that, assuming there's not a new alt account out there just ready to be all offended!

[LordNothing]

so a typical play 3 games and whine (to be fair op logged close to 1200). you cant judge the situation with so few data points. im almost to 30k games (jarls only seems to have logged 13k of them). things start to level out in a bigger data set. small numbers of games isn't even enough to get sorted by the psr system. it could take hundreds of games. id say the op had enough.

another thing to think about. the last batch was played during the holiday event season, when everyone plays like idiots in search of free loot. theres a huge spike about july around the summer lootbag event i believe. you really cant judge the game by how it plays at event time. gameplay is always *** during those.

Edited by LordNothing, 19 January 2022 - 11:22 AM.

[DaZur]

LordNothing, on 19 January 2022 - 11:21 AM, said:

You really cant judge the game by how it plays at event time. gameplay is always *** during those.

The game GUI and website should have a scrolling banner that says "Quality of gameplay will be impacted by this months event".

[lolbbq]

dubstep albatross, on 16 January 2022 - 08:19 AM, said:

I have put some thought into this overall topic and I agree. I would say that there is a certain gravity to runaway games that the routes to a "close" game are comparatively few. I attribute this to fundamental game mechanics, the stochastic nature of the match, and myriad unaccounted for variables in the matchmaking process.

It is interesting to think about "the first few pickoffs". How did the match arrive in that state? Is it randomness or is it a result of skill disparity? Is it a symptom of the underlying impending defeat (i.e., the defeat is going to happen because of other variables and this is how it initiates) or is it a contributing cause (i.e., because the one team now has less armor/firepower/eyes, the defeat has been initiated)? Naturally things are more grey than black-and-white and, intuitively, each match may have a different mix. Still, it is interesting to consider.

QP games are not 100 or even close to 80% efficient in terms of player-to-player encounter. A T1 player might bring a bad mech. A player might DC, overheat or forget to override. A bad pilot might unluckily run into two good enemy pilots or luckily stick around to support their two friendly pilots. Its not a matter of taking 12 players per side and running some variables to work out an outcome. Good pilots can make mistakes and bad pilots can perform well even outside of luck.

[Magnus Santini]

MadBede, on 13 January 2022 - 06:10 PM, said:

-lights running rampart

I assume you are referring to the big wall around HPG Manifold. I am glad you brought up the issue because whenever I try to run fast up there I inevitably go sailing off the edge at 167 and people need to know what to do in that situation. Shoot on your way down, "I meant to do that." Clean miss with 3 laser pulses? "Made you flinch." Remember, whatever mistake you make in a light, you still look cool if you just keep going. The only exception is faceplanting into a wall. That is embarrassing.

[LordNothing]

this thread is old, if he made good on his "im leaving" whine, you just bumped a 2 month old thread to argue with someone who is not here.