167 replies to this topic

[Ghogiel]

Diego Angelus, on 09 November 2013 - 01:26 PM, said:

ELO stat is laughable stat in mwo because it can only be true if it affects team not pugs. ELO is designed for chess so you get the picture how accurate it is in mwo when you are put in team of 12 people and base your rating around that. Best way to balance pugs is weight, how well does someone do in each mech (K/D,assists....) and number of matches played those are stats that MM should use.

Yes Elo was devised to rank chess players. Your assumption that it isn't applicable in team games is incorrect. It has found use is many team games including sports and video games.

[Diego Angelus]

Ghogiel, on 09 November 2013 - 01:30 PM, said:

Yes Elo was devised to rank chess players. Your assumption that it isn't applicable in team games is incorrect. It has found use is many team games including sports and video games.

Yes but in team games elo is used for whole team not individual player.

[Zarlaren]

Kills / Death 118 / 612
Wins / Losses 383 / 553

Time after time i'm being torn apart even when I try my best as a intermidate novice player. The thing is i'm being put in games where there is players that played since whenever and has 1000times more skill them I do. I'm looking at a 35%/65% win/loss chance here no matter what mech I use or loadouts I get cored in seconds when I get 200m range. Most of my kills are from the Jenner/Victor on a good day. And of course there is some games where the enemy cracks open a can of whoop butt and mass slaughters all my team without trying very hard in a perfect victory.

I'm being put in matches where there is elite players and I get gunned down before I can even do anything when shot at. Most of the mechs are made of paper and the Atlas is as solid as cardboard.

Edited by Zarla, 09 November 2013 - 03:03 PM.

[Adiuvo]

Zarla, on 09 November 2013 - 03:01 PM, said:

Kills / Death 118 / 612
Wins / Losses 383 / 553

Time after time i'm being torn apart even when I try my best as a intermidate novice player. The thing is i'm being put in games where there is players that played since whenever and has 1000times more skill them I do. I'm looking at a 35%/65% win/loss chance here no matter what mech I use or loadouts I get cored in seconds when I get 200m range. Most of my kills are from the Jenner/Victor on a good day. And of course there is some games where the enemy cracks open a can of whoop butt and mass slaughters all my team without trying very hard in a perfect victory.

I'm being put in matches where there is elite players and I get gunned down before I can even do anything when shot at. Most of the mechs are made of paper and the Atlas is as solid as cardboard.

What builds are you typically using?

[Mudhutwarrior]

Zarla, on 09 November 2013 - 03:01 PM, said:

Kills / Death 118 / 612
Wins / Losses 383 / 553

Time after time i'm being torn apart even when I try my best as a intermidate novice player. The thing is i'm being put in games where there is players that played since whenever and has 1000times more skill them I do. I'm looking at a 35%/65% win/loss chance here no matter what mech I use or loadouts I get cored in seconds when I get 200m range. Most of my kills are from the Jenner/Victor on a good day. And of course there is some games where the enemy cracks open a can of whoop butt and mass slaughters all my team without trying very hard in a perfect victory.

I'm being put in matches where there is elite players and I get gunned down before I can even do anything when shot at. Most of the mechs are made of paper and the Atlas is as solid as cardboard.

Yep, I have over 8000 matches pug only. I drop with both top players and new players and thats never changed. I really think the whole elo/mm is just a smokescreen to cover a severe lack of players. Today its 36 losses in a row and the majority stomps in 4 minutes or less.
Now of course the true beleivers will be along to disect your builds or tell you its your playstyle or some other such thing but we know differnt and anyone reading about MWO on any forum but this one knows too. This will all be censored soon.

[Zarlaren]

Adiuvo, on 09 November 2013 - 03:06 PM, said:

What builds are you typically using?

Right now I'm using Sarahs Jenner and a Catapult K2.

Jenner- 4 medium lasers and ssrms
K2- I'm still trying to skill it up so I can elite Jester when I get it. So K2 build is no too hot important cause it's purpose it to elite the Jester which I'll be getting soon.
Victor I sold some time ago I will revisit the Victor variants soon to elite it.
Atlas I sold wasn't too happy with it.

[Percimes]

MischiefSC, on 09 November 2013 - 09:13 AM, said:

I find that strange as his experience doesn't match with mine at all. If his experience is true then why do I only see new players once in a dozen or two dozen games?

I get the desire for people to want to feel that the matchmaker is responsible, that if they're not winning it's not because they made a mistake or over-extended themselves or left cover or chased a kill or went right on Canyon or some other nub mistake. I get that.

It doesn't make it true.

Elo is pretty accurate and it's based off of how well the player influences matches to drive a win for their team. That's it. Not your ability to swoop in and grab the kills that your allies weakened before dying, not your ability to stay alive by avoiding combat until the very end, etc. In fact a lot of the behaviors that benefit KDR (playing very cautiously, nabbing the easy kills) actually drive win/loss DOWN - if you're not in 12mans or seriously competitive play then playing cautiously is probably shafting your teammates, especially when the other team plays aggressive. You're not drawing attention from anyone and this allows the other team to focus their fire on fewer potential targets, increasing the damage your teammates take as individuals and causing that cascade of failure where 2 or 3 people on a team get focused down and then they get rolled.

At this point though it's been debated here a lot. All the details are out there. The MM is designed to keep the game challenging and that's uncomfortable for some people. They don't want a challenge, they want to feel like they're a winner.

The problem of course is that in any system where that's actually challenge oriented and not just handed out not everyone appreciates the effort winning entails.

John MatriX82's location is Italy, so he's most likely not playing during the US primetime. I usually play during the east coast afternoon, so I see many german and russians players. I've been able to play during primetime last week (and will for the next week too) and so far I can see a difference, if not in match outcomes, at last in team composition. More premades (on both sides) during primetime.

I've been noting my wins and loses for the last 3 weeks (got 420ish so far) so next week I should have a 2 weeks from the afternoon vs 2 weeks of primetime with 500 to 600 matches. Except for the first 120 or so I even have the mech weight classes.

[Homeless Bill]

I think wins and losses are simply the easiest way to calculate it and actually the most indicative of skill over time. Whereas damage, kills, assists, and every other hard metric can be very useful in certain circumstances, they can also be useless in others. Lots of damage could mean you tore everyone up and carried or are a terrible shot and needed 300 damage to kill each 'mech. Kills can be all you or a lucky steal.

Likewise, other things are quite useful to your team but could never be quantified. Suppression and distractions are often overlooked, but have immeasurable value in many situations. Even if you're hunted down while doing very little damage, if you are the reason the main brawl was an 11v7, you largely contributed to a win.

After thousands of matches, there's only one common factor. My win-loss ratio stays pretty constant and stays higher on my try-hard 'mechs. After about 1000 matches I don't think there is a better indication of usefulness for your team. There may be "worse" (in terms of aiming or however you want to define a pilot's skill) pilots with higher Elos, but they're also more "useful" in the sense that they cause their team to win more.

[Victor Morson]

Wispsy, on 08 November 2013 - 03:23 PM, said:

Well some of us manage to win an awful lot more then we lose...I guess that it is not really our contribution but just random matchmaking luck?

The latter. I've been in more than enough games where I have 3-4 times over the damage of the next highest person and more than a few there I have more damage than 2 lances combined, and we still lost. Frequently.

You can sometimes make the difference but ultimately if you've got 3 newbies and 2 Leroy Jenkins on your team, GG. If they have a premade or two on top of that?

Adiuvo, on 08 November 2013 - 03:30 PM, said:

Assuming you're not using a useless mech, you can always carry your matches. Like Wispsy, I win a lot more than I lose even when solo dropping. It requires more effort than if I was grouped, but it's still possible and I mainly play lights. Play an assault mech and the whole thing is easier.

I'd say this is silly but honestly it's more delusional if you think you can take credit for your team's loss/victory in a PUG. Well, perhaps if they're listening to you and you're trying to command them maybe you can then, but my point is that "having a great 'mech" and even getting great shots and spreading your damage ultimately will not impact your w/l in the match in the end, unless you're lucky enough to have dropped with at least a few people of equal talent, and even luckier to have avoided gimmick/grief/complete newbies getting on your team with the messed ELO.

Plus even if you two were totally right, and that you can totally one-man Rambo the entire enemy team... you just gave a win stat to 11 other people who may well be terrible, further coming full circle to the original problem anyway.

Edited by Victor Morson, 09 November 2013 - 05:25 PM.

[Victor Morson]

Wispsy, on 08 November 2013 - 05:16 PM, said:

So...I must be really lucky to go 70 wins in a row without my performance affecting the outcome. Then to do it again...the gods just love me?

No but I'd suspect you're dropping with a premade if you're having any kind of consistent record. 4 people can dictate the round.

And honestly I in no way buy you have 70 wins in a row if you're dropping solo. That is a just plain silly and ludicrous claim with not much backup and is crossing into blatant lie territory.

EDIT: In fact Wispy, I'm outright calling you on this. I know you are a good pilot but the odds of not getting ludicrously broken weight, disco'ed 'mechs on your team, etc. in pugs makes your claim have worse odds than the lottery. Even in a 4 man premade 70 in a row without a MM disaster is close to impossible. So yes, I suppose I am calling you a liar on this one.

Edited by Victor Morson, 09 November 2013 - 05:37 PM.

[Deathlike]

MischiefSC, on 09 November 2013 - 09:13 AM, said:

I get the desire for people to want to feel that the matchmaker is responsible, that if they're not winning it's not because they made a mistake or over-extended themselves or left cover or chased a kill or went right on Canyon or some other nub mistake. I get that.

It doesn't make it true.

MM is responsible for a lot of bad lopsided matches. Then again, it doesn't always dictate guaranteed success or failure either. Matchmaking could be better either way. It is easy to attribute how you win or lose to a multitude of factors though.

Quote

Elo is pretty accurate and it's based off of how well the player influences matches to drive a win for their team. That's it. Not your ability to swoop in and grab the kills that your allies weakened before dying, not your ability to stay alive by avoiding combat until the very end, etc. In fact a lot of the behaviors that benefit KDR (playing very cautiously, nabbing the easy kills) actually drive win/loss DOWN - if you're not in 12mans or seriously competitive play then playing cautiously is probably shafting your teammates, especially when the other team plays aggressive. You're not drawing attention from anyone and this allows the other team to focus their fire on fewer potential targets, increasing the damage your teammates take as individuals and causing that cascade of failure where 2 or 3 people on a team get focused down and then they get rolled.

Yes, the little things matter sometimes. It can be the difference. Although, in some matches, it makes no difference. It's a strange dichotomy.

Quote

At this point though it's been debated here a lot. All the details are out there. The MM is designed to keep the game challenging and that's uncomfortable for some people. They don't want a challenge, they want to feel like they're a winner.

I think most people want it to be competitive in the sense that the people you face are generally the people at about your level. It's completely subjective, but on the other hand, I wish it was kinda quantifiable. There's a lot of feel to how the MM trolls everyone equally though.

Quote

The problem of course is that in any system where that's actually challenge oriented and not just handed out not everyone appreciates the effort winning entails.

Sometimes winning is easy. Other days it's not. It is the luck of the draw. I kinda wish that matches were close and competitive more often than not. That doesn't always happen here, but I guess we can keep chugging along for a bit more.

The best we can do here is do as much as we possibly can to tip it our way when the game is "relatively close".

Edited by Deathlike, 09 November 2013 - 05:32 PM.

[Victor Morson]

Praetor Shepard, on 08 November 2013 - 09:59 PM, said:

What about adding to Elo? Maybe some sort of Matchscore addition that could further raise or decrease the gains/losses in Elo score outside of winning and losing?

And for more on Elo:
http://mwomercs.com/...ost__p__1626065

Matchscore would be a great stat to read from instead, definitely.

The guy with 800 damage on the losing team and capped three points should get rewarded, not punished for a bad team.

[Grits N Gravy]

Homeless Bill, on 09 November 2013 - 05:09 PM, said:

I think wins and losses are simply the easiest way to calculate it and actually the most indicative of skill over time.

That's the beauty of Elo scoring systems, they quantify the quality of win/loss rates. IE you get more points for betting a higher ranked, (better), opponent and lose.

Homeless Bill, on 09 November 2013 - 05:09 PM, said:

After thousands of matches, there's only one common factor. My win-loss ratio stays pretty constant and stays higher on my try-hard 'mechs. After about 1000 matches I don't think there is a better indication of usefulness for your team. There may be "worse" (in terms of aiming or however you want to define a pilot's skill) pilots with higher Elos, but they're also more "useful" in the sense that they cause their team to win more.

The best indicator for usefulness to a team is mech choice and build. Subsequently, your mech choice and load out will largely be best way to boost your Win/loss, Elo, KDR, ect. One of the problems with Elo is that it assumes all things are equal from the participants. IE in chess you have the same number of pieces. In MWO it's a little different because you can take a decent chassis and make it totally ineffective. Don't underestimate the power of stupid people in groups. Also because we are only dealing with 12 per side all it takes is 1 or two bad builds and you're at large numeric disadvantage on any metric.

Most of these issues would work themselves out if the game was on a much bigger scale, like teams of 50 to 100 scale. For instance, in Planetside 2 if one guy has a stock rifle, it won't stop a base take. In MWO if someone goes all small pulse lasers on a Battlemaster, it might ruin a match.

Edited by Grits N Gravy, 09 November 2013 - 05:43 PM.

[Diego Angelus]

Homeless Bill, on 09 November 2013 - 05:09 PM, said:

I think wins and losses are simply the easiest way to calculate it and actually the most indicative of skill over time. Whereas damage, kills, assists, and every other hard metric can be very useful in certain circumstances, they can also be useless in others. Lots of damage could mean you tore everyone up and carried or are a terrible shot and needed 300 damage to kill each 'mech. Kills can be all you or a lucky steal.

Likewise, other things are quite useful to your team but could never be quantified. Suppression and distractions are often overlooked, but have immeasurable value in many situations. Even if you're hunted down while doing very little damage, if you are the reason the main brawl was an 11v7, you largely contributed to a win.

After thousands of matches, there's only one common factor. My win-loss ratio stays pretty constant and stays higher on my try-hard 'mechs. After about 1000 matches I don't think there is a better indication of usefulness for your team. There may be "worse" (in terms of aiming or however you want to define a pilot's skill) pilots with higher Elos, but they're also more "useful" in the sense that they cause their team to win more.

W/l is not best if you need one stat to try and get a feel for player skill its K/D Ratio it gives a lot more information then elo or W/L. Lets take my example I have 176W/179L witch is average but my K/D says I suck 173/253. But if I go to mech stats i can see that mechs that i play best have around 1/1 K/D. If i have good win ration that doesn't mean I'm good player it just means I'm lucky because lets take my TBT-5J stat I have 21W/6L but K/D is 0.35 one of worst so tell me how is w/l as a stat good ?

[Wispsy]

Victor Morson, on 09 November 2013 - 05:28 PM, said:

No but I'd suspect you're dropping with a premade if you're having any kind of consistent record. 4 people can dictate the round.

And honestly I in no way buy you have 70 wins in a row if you're dropping solo. That is a just plain silly and ludicrous claim with not much backup and is crossing into blatant lie territory.

EDIT: In fact Wispy, I'm outright calling you on this. I know you are a good pilot but the odds of not getting ludicrously broken weight, disco'ed 'mechs on your team, etc. in pugs makes your claim have worse odds than the lottery. Even in a 4 man premade 70 in a row without a MM disaster is close to impossible. So yes, I suppose I am calling you a liar on this one.

Ofc most of the games were in a 4man. To be honest I have only broken 70 once, closest to that was just over 60. Point is it happened and I still win a significant amount more then I lose, even when solo. Of course I have bad games and hard games and really really challenging perfect play only games but I have a lot more that I can influence in enough of a way to win reliably. Just carry more, dodge harder. Instead of whining about how unfair it was, how about just looking inward and seeing what more you could have done to influence the fight, given the tools (team) at your disposal.

[Victor Morson]

Wispsy, on 09 November 2013 - 06:00 PM, said:

Ofc most of the games were in a 4man.

And this is the reason everything you have said to this point is entirely invalid about ELO. If you control a third of the team, then yes, you can make a massive increase in difference.

Having one winning streak with 4-mans I can believe, as unlikely as it is, but still - your initial argument seemed to take the rather definite posture of "See if you're super super good yourself, and you believe really really hard, you too can change the game!" style rhetoric, when in reality, that is entirely an unrealistic picture to paint.

In fact, the fact you had a 4 man rolling teams 70 times in a row just goes to prove my point even more. Are you saying every one of the pilots you faced in 70 games deserved a crushing loss because they didn't have an equally effective 4-man to counter yours? Are you saying that entering with a crack 4-man and influencing who wins/looses entirely makes w/l any kinds of acceptable for ELO of the individual pilots?

And most importantly you understand that solo pilots, no matter how skilled and how much damage they get, lost ELO because they had a worse team?

Really man I have to say, while I do appreciate individual pilots being able to occasionally make impacts, this whole "If you're badass enough you can do anything!" kind of claims when this comes up are outright silly and this just continues to prove that.

Edited by Victor Morson, 09 November 2013 - 06:06 PM.

[MischiefSC]

If it was actually random win/loss would be 1.0 after a couple of hundred matches.

it isn't.

Then again we've been over this repeatedly.

800 damage means next to *nothing* in terms of a team winning or losing. 80 match score means *nothing* in terms of a team winning or losing. It means you got assists, or you use LB10X instead of AC10s and did plenty of damage to components to wrack up the score instead of putting the kill down. A guy who pop the right leg off every enemy mech and all but guarantees a win for his team might get a 40 match score since he did comparatively little damage and no kills.

The ECM spider who sticks with his team or uses ECM to disrupt enemy LRM fire does little damage but sows havoc on the enemy team, the two man suicide squad that drew half the enemy team to the tunnel and died with only 140 damage and a score of 30 absolutely swung the match in his teams favor.

Conversely the guy in the 2 PPC Cicada who lurks in the back and finishes every mech with a red CT he can find isn't contributing anything to the match that a stock locust couldn't do, only with more spine. He does get a high KDR though, right? Especially when he runs off and shuts down as soon as the match looks like his team is going to lose. He'd have a high KDR and probably a high average score since he rarely exposes himself to fire and survives most matches.

In the end, win/loss is what matters. It's all that really matters. Every single other factor, luck of the matchmaker, who you drop with/against, these are metrics that are the same for EVERYONE. For this mythical magical 'evil matchmaker bad luck' to function it would have to be driven by sentience and purposefully gimp only the people who come here to say the MM is evil and unfair.

All that balances out. What doesn't balance out is YOU. We are, each one of us, the only consistent variable in every match we play. Every single other variable balances out across every other player. Different builds, mechs, weapons, modules, maps, they all contribute in different ways both good and bad to our performance. If you play well and focus on helping your team win rather than getting a high score/more kills you absolutely WILL win more than you lose. If you hone your skill with a couple styles that work well then you'll kill more enemies and pull more than your 1/12th the burden in matches and thus statistically win more than people who don't pull as hard.

It's math. It's not even hard math. In every single match your team is divided in effort 12 ways. If you bring a bigger piece of success, be that in kills or damage or assists or spotting or tactics or positioning or drawing disproportionate enemy attention or whatever you'll result in a bigger over-all total. Sometimes it's enough to counter-balance any detriment on your team from the other 11, sometimes its not. This is absolutely balanced out by the times the other team gets left with a comparatively empty bucket. How much YOU BRING is what makes the statistical difference FOR YOU in the aggregate.

You want to win more? Bring more. Do better. Focus on how to win the match not how to raise your personal score. That's exactly why Elo works - it doesn't measure your score because you can always game that. It measures how consistently you help bring your team a victory. The only way to game win/loss is by winning more or losing more. Which, in turn, affects your score.

Elo works in MW:O. It's the right system to use for matchmaking in this sort of game and that's why other games of the same genre (like LoL) use something similar.

Victor Morson, on 09 November 2013 - 06:04 PM, said:

And this is the reason everything you have said to this point is entirely invalid about ELO. If you control a third of the team, then yes, you can make a massive increase in difference.

Having one winning streak with 4-mans I can believe, as unlikely as it is, but still - your initial argument seemed to take the rather definite posture of "See if you're super super good yourself, and you believe really really hard, you too can change the game!" style rhetoric, when in reality, that is entirely an unrealistic picture to paint.

In fact, the fact you had a 4 man rolling teams 70 times in a row just goes to prove my point even more. Are you saying every one of the pilots you faced in 70 games deserved a crushing loss because they didn't have an equally effective 4-man to counter yours? Are you saying that entering with a crack 4-man and influencing who wins/looses entirely makes w/l any kinds of acceptable for ELO of the individual pilots?

And most importantly you understand that solo pilots, no matter how skilled and how much damage they get, lost ELO because they had a worse team?

Every one of those solo pilots was equally as likely to be on the same team as Wispy and his 3 friends as they are the other team and its less successful 4man. Statistically irrelevant to everyone who pugs.

For your point to work the MM would have to be sentient and evil. An interesting proposition but I admit one I find unlikely.

[Wispsy]

Victor Morson, on 09 November 2013 - 06:04 PM, said:

And this is the reason everything you have said to this point is entirely invalid about ELO. If you control a third of the team, then yes, you can make a massive increase in difference.

Having one winning streak with 4-mans I can believe, as unlikely as it is, but still - your initial argument seemed to take the rather definite posture of "See if you're super super good yourself, and you believe really really hard, you too can change the game!" style rhetoric, when in reality, that is entirely an unrealistic picture to paint.

In fact, the fact you had a 4 man rolling teams 70 times in a row just goes to prove my point even more. Are you saying every one of the pilots you faced in 70 games deserved a crushing loss because they didn't have an equally effective 4-man to counter yours? Are you saying that entering with a crack 4-man and influencing who wins/looses entirely makes w/l any kinds of acceptable for ELO of the individual pilots?

And most importantly you understand that solo pilots, no matter how skilled and how much damage they get, lost ELO because they had a worse team?

Really man I have to say, while I do appreciate individual pilots being able to occasionally make impacts, this whole "If you're badass enough you can do anything!" kind of claims when this comes up are outright silly and this just continues to prove that.

It was not in a single day/session. A number of those I played my Dragon solo, it was not all 4mans, just mostly.

[Victor Morson]

MischiefSC, on 09 November 2013 - 06:06 PM, said:

Every one of those solo pilots was equally as likely to be on the same team as Wispy and his 3 friends as they are the other team and its less successful 4man. Statistically irrelevant to everyone who pugs.

Except you just disproved your own point.

If a pug is statistically as likely to appear on Wispy's four man murdersquad, as against it, that means they are going to get a 1:1 w/l ratio of every game they play with Wispy if they lucked out and got a perfect 50/50 split going on and off his team.

Rendering ELO from wins moot. Because you just noted in this scenario it's pure luck if you end up on the team with the winning group.

MischiefSC, on 09 November 2013 - 06:06 PM, said:

For your point to work the MM would have to be sentient and evil. An interesting proposition but I admit one I find unlikely.

Or completely incompetently programmed.

Wispsy, on 09 November 2013 - 06:07 PM, said:

It was not in a single day/session. A number of those I played my Dragon solo, it was not all 4mans, just mostly.

This still seems to discount the "If you are good enough, you entirely dictate if your team wins" statements that started this quite a bit.

I think everyone here can agree ELO doesn't work at all right now, and again, this is the reason I continue to directly attribute. Win/loss in a team heavy game is simply not a stat worth tracking, outside of team-only modes.

Edited by Victor Morson, 09 November 2013 - 06:18 PM.

[Adiuvo]

Victor Morson, on 09 November 2013 - 05:21 PM, said:

I'd say this is silly but honestly it's more delusional if you think you can take credit for your team's loss/victory in a PUG. Well, perhaps if they're listening to you and you're trying to command them maybe you can then, but my point is that "having a great 'mech" and even getting great shots and spreading your damage ultimately will not impact your w/l in the match in the end, unless you're lucky enough to have dropped with at least a few people of equal talent, and even luckier to have avoided gimmick/grief/complete newbies getting on your team with the messed ELO.

Plus even if you two were totally right, and that you can totally one-man Rambo the entire enemy team... you just gave a win stat to 11 other people who may well be terrible, further coming full circle to the original problem anyway.

You're referencing a lot of things as based on luck when they're in reality skill based, and able to make an impact on matches.

How do you explain people who have high solo drop win rates? Is it just magical luck? If it was literally everybody would have a 1.0W/L.