58 replies to this topic

[DaZur]

giganova, on 19 March 2014 - 08:03 AM, said:

Stop projecting. You claiming it to be simple math was a straw man argument and the burden of proof lies on you, good sir.

It is plainly simple... For every mech lost on your side, the opposing team nets the equivalent numerical advantage as well as applied advantage of firepower. In abstract, once your team reached what I affectionately call the "tipping point" the applied force strength of the dominant team simply overwhelms the other team...

I don't care if your team is comprised of seasoned top-tier veterans or mouth-breathing knuckle-draggers... This phenomenon is predictable and repeatable. Granted, in one case it's much more likely to happen within the first 3 min. of a match...

It's not flipping magic or anything.

[Varent]

TygerLily, on 19 March 2014 - 07:53 AM, said:

I'd take supertight matchmaking over faster games and even Community Warfare...

Trust me its not what you think it is. Failing to find a match 3+times and sitting thre in que for about 15 minutes gets taxing. And even then that wont give you perfect match making. You are literally talking about 20-30 minute ques for perfect matchmaking. There is a reason that no shooter has perfect matchmaking. Look at any of the others out there, There is always deviation and always a few stomps. Its part of the game im afraid unless you want to be sitting around forever.

[Aresye]

I pick neither. As said in my original post, I don't necessarily have a problem with the way the matchmaker is programmed, and more that I have a problem with the fact that there's a matchmaker in the first place. Newer players had no problem adapting to mechwarrior 3 and 4 online. Counter-Strike is horribly unfriendly to new players yet remained the most popular online shooter for god knows how long.

Back when we only had tonnage balancing, I still had my fair share of stomps, but at least I knew that I was bested by better players, and there wasn't an automated system designed to stack the odds against me and predetermine my fate before my map even loads.

If anything, I vote for the keep it simple stupid approach. If there's one thing I've learned from the military, it's that the more you micromanage, the less productive you are as a whole. What PGI is doing in terms of balancing the game is essentially micromanaging players themselves, and not actually fixing balance issues at the basic level like they should. It's like fixing a helicopter's main transmission by only letting lighter crewmembers fly on it, and that's the exact approach that PGI takes for balancing, by taking unbroken mechanics and tweaking them to nullify broken mechanics. In the end, the mechanic is still broken, and players have even more restrictions on how they choose to play. That's why the meta is what it is. The meta builds themselves aren't what's broken. They're just simply the best builds to overcome broken mechanics in other areas, such as hit detection and tonnage mismatches.

It's simply an unwinable approach. It takes more time to implement, which costs more money for the developers. Essentially they're working harder, spending more, and continuously breaking their own previous fixes in a never ending cycle.

[Veranova]

Willard Phule, on 19 March 2014 - 07:08 AM, said:

Right. Ok. And I get it. And I understand that a "simple" solution is something we're NEVER going to see. Because every solution even considered by PGI must, by definition, be over complicated and have the potential to break the game worse than it is right now.

But.

Just for a second, imagine how the matchmaker would work WITHOUT the whole "prediction" factor.

It puts two teams together with relatively even Elos.

The game gets played.

The winning team's Elo is adjusted upward based on whatever the hell it is they're basing it on now.

The losing team's Elo is adjusted downward the same way.

Instead of having a system where your Elo goes up or down slowly, you will have a much more dynamic change and separation of the different skill levels.

Look, ask any of the Derps out there. They don't want to be stuck in a match where everyone just rolls over their face any more than we want to be stuck with a bunch of lemmings. There's a learning curve involved here.

Take away the prediction, separate the skill levels quicker, things move on.

Granted, this doesn't address tonnnage, weight classes, ecm, etc...but, for crying out loud, this is something they could do NOW..without a major overhaul or patch. And it would definitely calm the masses before they give us a half-assed, broken launch module in April.

I actually think it would be funner, because over-all we might end up across a greater range of ELO's throughout a week of play.
HOWEVER, it would seriously affect the number of roflstomps, since good and bad players would be doing exactly that, moving between ELO's all over the shop.

ELO ideally needs to represent your skill level, and allowing players to move between lots of levels across a week would mean that it's not representing their skill level.

There are rules in the math to stop your ELO tanking or bloating easily, except in win/loss streaks, but the"prediction" does help to stabilise the system as far as I can work out.

Edited by Veranova, 19 March 2014 - 08:30 AM.

[TygerLily]

Varent, on 19 March 2014 - 08:25 AM, said:

Trust me its not what you think it is. Failing to find a match 3+times and sitting thre in que for about 15 minutes gets taxing. And even then that wont give you perfect match making. You are literally talking about 20-30 minute ques for perfect matchmaking. There is a reason that no shooter has perfect matchmaking. Look at any of the others out there, There is always deviation and always a few stomps. Its part of the game im afraid unless you want to be sitting around forever.

I know how it feels...And I wouldn't say I'm asking for "perfect matchmaking" but "supertight". Like you said, the "deviation" of "a few stomps" is unavoidable but I think many peoples experience is that it's not "a few"...but quite a bit more, actually.

What's more taxing, waiting several minutes for a good game? Or getting smashed repeatedly over the course of several minutes?

[Rokuzachi]

Yea, the LRMs turned out to not be that big a deal. I thought it was a problem at first, but not so much now. Especially in a week when the new-shiny wears off.

It basically distracted everyone, lol. Maybe that was their plan

I'm still far more annoyed at; base turrets, one-sided matches, hit reg, getting stuck in the world/falling through it, non-LRM weapon balance, the role of medium mechs in the current state of the game, et cetera.

Could care less about LRMs if those things were being fixed.

[Varent]

TygerLily, on 19 March 2014 - 08:34 AM, said:

I know how it feels...And I wouldn't say I'm asking for "perfect matchmaking" but "supertight". Like you said, the "deviation" of "a few stomps" is unavoidable but I think many peoples experience is that it's not "a few"...but quite a bit more, actually.

What's more taxing, waiting several minutes for a good game? Or getting smashed repeatedly over the course of several minutes?

How about using smart tactics and becoming a better player so you have more of an overall effect in game. Or as a team droping with 3 others and using tactics.

I have never experienced a shooting game to date that has had the type of matchmaking people want. That said if I had to choose id choose short time frames and having good players in the matches that push me to be better. People get discouraged to much jsut because they lose instead of seeing it as an opportunity to get better.

[Rokuzachi]

DaZur, on 19 March 2014 - 08:14 AM, said:

It is plainly simple... For every mech lost on your side, the opposing team nets the equivalent numerical advantage as well as applied advantage of firepower. In abstract, once your team reached what I affectionately call the "tipping point" the applied force strength of the dominant team simply overwhelms the other team...

I don't care if your team is comprised of seasoned top-tier veterans or mouth-breathing knuckle-draggers... This phenomenon is predictable and repeatable. Granted, in one case it's much more likely to happen within the first 3 min. of a match...

It's not flipping magic or anything.

I might not have enough battles to say this with certainty, but the 'tipping point' in MWO feels much sooner than in other games of a similar nature.

The amount of victories snatched from the jaws of defeat that I've experienced in this game are extremely low. Maybe I just suck, but I'm also much less able to come back against the odds personally in this game than others.

[TygerLily]

Varent, on 19 March 2014 - 08:54 AM, said:

How about using smart tactics and becoming a better player so you have more of an overall effect in game. Or as a team droping with 3 others and using tactics.

I have never experienced a shooting game to date that has had the type of matchmaking people want. That said if I had to choose id choose short time frames and having good players in the matches that push me to be better. People get discouraged to much jsut because they lose instead of seeing it as an opportunity to get better.

I think you're asking for two opposite things:
"short time frame": lower threshold, larger room for gaps of skill.
"having good players": higher threshold, longer wait, but less gap of skill.

I actually drop with 3 or 4 most nights...and we do fairly well in our 12-man matches. We have good skilled players, lots of experience, and know some good tactics. I think Svidro was right up above: it seems like MM builds one good ELO team, then tries to build another, but ends up incrementally lowering the threshold to find players. That skill gap on one team means regardless of you own skill or good tactics, there are more "rookie" mistakes being made on one team than another (for example, separation: even the most elite 4-man can only do so much vs 12 grouped mechs, of any skill level).

Svidro's explanation also follows the procedure for MM that was described by the devs. I think it's just a big off kilter...

[xXBagheeraXx]

So much this....I would rather have a MM simply made up by weight, not elo....Ill take a moron in an atlas anyday over having a team full of lights vs DDC lance from hell on the enemy team..yeah our lights might be the spawn of kerensky himself, but when you are out weighed, you are simply OUTWEIGHED. The ELO thing cuts both ways.

[DaZur]

Rokuzachi, on 19 March 2014 - 08:58 AM, said:

I might not have enough battles to say this with certainty, but the 'tipping point' in MWO feels much sooner than in other games of a similar nature.

The amount of victories snatched from the jaws of defeat that I've experienced in this game are extremely low. Maybe I just suck, but I'm also much less able to come back against the odds personally in this game than others.

Completely based on personal "feel"... IMHO the tipping-point occurs around a differential of 4 mechs. There is defiantly +/- effect dependent upon the makeup of the remaining mechs. (i.e... A remaining Highlander has a stronger influence on the ebb/flow than say a Trebuchet or Quickdraw).

I've both been on and been the recipient of tide-turns with a greater differential but those are few and far between...

[giganova]

DaZur, on 19 March 2014 - 08:14 AM, said:

It is plainly simple... For every mech lost on your side, the opposing team nets the equivalent numerical advantage as well as applied advantage of firepower. In abstract, once your team reached what I affectionately call the "tipping point" the applied force strength of the dominant team simply overwhelms the other team...

I don't care if your team is comprised of seasoned top-tier veterans or mouth-breathing knuckle-draggers... This phenomenon is predictable and repeatable. Granted, in one case it's much more likely to happen within the first 3 min. of a match...

It's not flipping magic or anything.

Yep, you were definitely projecting. You fail to consider the "flipping magic" that is the ELO and matchmaking processes. The phenomena is by no means predictable or repeatable, but rather the opposite - it is quite stochastic. I recommend brushing up on some probability theory before relying on anecdotal evidence.

DaZur, on 19 March 2014 - 07:36 AM, said:

"Stomps" have little to do with the MM or the Elo aside from the proficiency of the team composite.

Sounds pretty contradictory and dogmatic to me...

Edited by giganova, 19 March 2014 - 09:28 AM.

[Varent]

TygerLily, on 19 March 2014 - 09:06 AM, said:

I think you're asking for two opposite things:
"short time frame": lower threshold, larger room for gaps of skill.
"having good players": higher threshold, longer wait, but less gap of skill.

I actually drop with 3 or 4 most nights...and we do fairly well in our 12-man matches. We have good skilled players, lots of experience, and know some good tactics. I think Svidro was right up above: it seems like MM builds one good ELO team, then tries to build another, but ends up incrementally lowering the threshold to find players. That skill gap on one team means regardless of you own skill or good tactics, there are more "rookie" mistakes being made on one team than another (for example, separation: even the most elite 4-man can only do so much vs 12 grouped mechs, of any skill level).

Svidro's explanation also follows the procedure for MM that was described by the devs. I think it's just a big off kilter...

Im actually not. You do realize the only way you get better is being challenged right? If your seeing nothing but stomps consistently maybe some of these players should consider that the problem might be on there end.

If I dont carry and none of my team carries, I expect to lose and get stomped.

If I carry and am doing my job, I expect to win the majority of those. And I do.

[TygerLily]

Varent, on 19 March 2014 - 09:30 AM, said:

Im actually not. You do realize the only way you get better is being challenged right? If your seeing nothing but stomps consistently maybe some of these players should consider that the problem might be on there end.

If I dont carry and none of my team carries, I expect to lose and get stomped.

If I carry and am doing my job, I expect to win the majority of those. And I do.

I agree and I'm not actually taking issue with how players get better. The threshold exists to allow a small variation of ELO difference...so no matter what, you'll likely be playing against least a few people better than you. That's good, etc.

"If I carry and am doing my job, I expect to win the majority of those. And I do." That seems correct to me...but I'm arguing that there's only so much you can carry. And if MM is allowing too much of a skill gap, then then even the most elite 4-man team will be annihilated by 12 average players if the 4-man is matched with inexperience players (who scatter, etc.)

[Orbit Rain]

Varent, on 19 March 2014 - 08:54 AM, said:

How about using smart tactics and becoming a better player so you have more of an overall effect in game. Or as a team droping with 3 others and using tactics.

How about not making invalid assumptions? Tyger's in the higher end of players. When you're in the higher end, you *know* you're often getting pug-grouped with bad/new/cr@ppy players to average out the elo (as has been said many times before) to some form of "balance" with the other team. It's not just the bad players that are getting bad games, it's the good players too. (also, said many times before)

None of us are getting paid to fix this problem. PGI will have to figure it out to make the monies...or not (as they are), and continue to *not* make the money they could be making. If your answer is "carry harder" you aren't fixing the problem.

[Varent]

TygerLily, on 19 March 2014 - 09:39 AM, said:

I agree and I'm not actually taking issue with how players get better. The threshold exists to allow a small variation of ELO difference...so no matter what, you'll likely be playing against least a few people better than you. That's good, etc.

"If I carry and am doing my job, I expect to win the majority of those. And I do." That seems correct to me...but I'm arguing that there's only so much you can carry. And if MM is allowing too much of a skill gap, then then even the most elite 4-man team will be annihilated by 12 average players if the 4-man is matched with inexperience players (who scatter, etc.)

How else will players Learn.

Honestly I have not experienced this constant stomp that people seem to feel exists. every once in awhile there is a bad game. Every once in awhile there is an amazing game. Stomps will always happen on occassion. Im not an amazing player, im not a bad player, Im defeinetly above average and I consistently try new things and work to improve.

If players are experiencing constant stomps they may want to try to determine what they are contributing to the overalll game and team themselves.

If they are on average only doing 200-300 damage then perhaps they need to look at what they are contributing and evaluating how much of the problem is coming from themselves.

Orbit Rain, on 19 March 2014 - 09:41 AM, said:

How about not making invalid assumptions? Tyger's in the higher end of players. When you're in the higher end, you *know* you're often getting pug-grouped with bad/new/cr@ppy players to average out the elo (as has been said many times before) to some form of "balance" with the other team. It's not just the bad players that are getting bad games, it's the good players too. (also, said many times before)

None of us are getting paid to fix this problem. PGI will have to figure it out to make the monies...or not (as they are), and continue to *not* make the money they could be making. If your answer is "carry harder" you aren't fixing the problem.

See above statement. You are never not going to have some stomps, thats just a fact. You cant expect every game to get close.

Ive not seen this huge string of stomps in a row. Perhaps evaluate yourself.

Ive played with Tyger and against them a few times in 12's and non.

This isnt a personal attack. This is a statement for those that are in stomps to perhaps evaluate them abit closer.

[DaZur]

giganova, on 19 March 2014 - 09:22 AM, said:

I recommend brushing up on some probability theory before relying on anecdotal evidence.

Sounds pretty contradictory and dogmatic to me...

​I acknowledge the composite skill of the team does have an indirect effect on how decisive the effect takes place... what more do you want? Absolutely... you can't predict the terms of the MM. But it's not the MM that manifests the stomps...

No probability model is necessary... It's a functional compounding equation with predictable outcome +/-.

This is a circular argument. You clearly refuse to acknowledge stomps are a function of compounding force strength and have it in your head to blame the Elo... I obviously don't have the breath to belabored this issue...

[giganova]

DaZur, on 19 March 2014 - 10:02 AM, said:

​I acknowledge the composite skill of the team does have an indirect effect on how decisive the effect takes place... what more do you want? Absolutely... you can't predict the terms of the MM. But it's not the MM that manifests the stomps...

No probability model is necessary... It's a functional compounding equation with predictable outcome +/-.

This is a circular argument. You clearly refuse to acknowledge stomps are a function of compounding force strength and have it in your head to blame the Elo... I obviously don't have the breath to belabored this issue...

I can in fact acknowledge that when a mech goes down, that mech's opposing team then has the statistical advantage. That is simple math. I never said anything to the contrary. What goes on before the match even begins, however, THAT is what I am bringing attention to and that is what you appear to be dodging by calling a circular argument. Have you not seen the "flow charts"? I would argue that a probability model is fairly useful in deciphering them, even necessary.

I require no belaboring on your part, if you can't make your argument in less than 5 minutes (or a few kilobytes of text, in this case), then you're probably wrong.

Edited by giganova, 19 March 2014 - 10:23 AM.

[DaZur]

giganova, on 19 March 2014 - 10:22 AM, said:

I require no belaboring on your part, if you can't make your argument in less than 5 minutes (or a few kilobytes of text, in this case), then you're probably wrong.

Now who's projecting?

[giganova]

DaZur, on 19 March 2014 - 10:29 AM, said:

Now who's projecting?

Zing? I'll have to invoke hypothesis non fingo for that one... Nice dodge again, BTW

Edited by giganova, 19 March 2014 - 10:37 AM.