325 replies to this topic

[FaceRipt]

IamSalvation, on 23 May 2013 - 06:17 AM, said:

just the last days i brought 2 old CB Veterans and one new player into the game.

All quit because of the Premade Stuff.

Is it just me or did the poster say he tried to make his own 4-man, which did'nt seem to yield positive results, so all other 4-mans are exploiters.

I started this game solo and have since brought 3 more friends into this game that have stayed, several others could not take the 25 death trap matches and quit. Sorry your friends bailed keep bringing more and some will surely stick around.

This game Advertises for you to run in groups so, 4-mans= exploit i think not.

i know teamwork sucks right.

[Ghogiel]

DeaconW, on 31 May 2013 - 11:28 PM, said:

ELO is broken. Your argument is invalid (see, I can play your silly little game!). You could have vastly different performance in PUG or premade which would affect the average ELO, but not give a true measure of your performance for each specific case. But I admire your religious devotion to a single-player concept developed for chess that has been transported to a multiplayer environment where the pieces are all variable and the board changes constantly.



Such well reasoned arguments...you should be on the debate team! Congratulations! You have just demonstrated that you are not worth my time. Welcome to my ignore file. Your chair is the small blue one in the corner...

Blasphemer! You are shamed.... shaaaaaamed I tell you.

Elo, holy is thou name. Stand in the presence of Elo and confront it's divinity or ****. Your ignorance to its ways are irrelevant to it's judgement and it cares not about your fallacies and slander of either it's prophets or itself.

[Ngamok]

Ezekeel666, on 31 May 2013 - 01:46 PM, said:

The point is not that you cannot queue up solo, but that when you do the matchmaker will put you into a group with other solo player that is fighting an enemy group also consisting of solo players. Or if some small premade groups are mixed in that this happens equally on both opposing sides to ensure balance. In the same way, full premades are put against other full premades to give both sides a challenging match which improves your skill and team work which is exactly what the high-skilled competitive players are craving for. Due to the currently hardly working matchmaker MW:O is losing potential players on both ends.

You missed the point completely. You said any successful PvP game does not do it and I showed you ever successful PvP game that does do it. MWO is no different currently than dropping solo into a TDM CoD match where one side can have all the premades and one side has all the solo people.

[Taemien]

Not sure how this thread isn't locked or jettisoned yet.. but since it isn't, I'll ask this question:

What happened to the Solo player? In MechWarrior 3 and 4 they held their own against premades. They didn't complain everyone on the other team had a clan tag. They didn't throw a fit. They launched in the game and kicked butt. Why can't the solo player do the same thing here? In MW3 and MW4, they knew before they even hit launch they were going against a premade, they could see the clan tags in front of players' names in the lobby in MW3, they could see the unit name under the player name in the MW4 ready screen.

Don't say we didn't have VOIP back then, we did. It was BattleCom, Roger Wilco, and others. So the advantages premades have now, they had back then too. But still, what happened with the solo player?

[Panzerman03]

Ghogiel, on 31 May 2013 - 05:07 PM, said:

You have not mentioned how Elo actually works only weight class match making. What it does is give a rating to players, which is used to determine match ups based on past performance, assigns a probability rating on W/L for each match up and adjusts their rating based on outcome.

The odds are already predetermined based on how the player performs in previous matches. What that means is that it already takes into account the players behavior. PUG or premade it doesn't matter.

The first paragraph you wrote there is the same exact one I have read before many times, the one used to rationalize how PUGs playing against premades is somehow unfair. Since Elo rendered that idea moot, it also renders this one moot for exactly the same reasons.

More garbage. Your argument is only true if these players only play in sync dropped double 4-man teams. ELO gives ratings based on whether or not the outcome of a match agrees with predictions the system made when it created the match. If you sync drop an 8 man full of players that usually don't get to play with such a massive advantage, you not only get an unfair matchup, but also an invalid rating adjustment for both teams after you pug stomp the daylights out of your opponents.

The system does not account for the sync-droppers' behavior because they almost certainly won't behave the same in the vast majority of their matches where they're failing to successfully sync drop (either through lack of trying or just not pulling it off).

It's actually probably the case that without ELO you'd sometimes see random 8-man pugs that could tackle a double 4-man sync, but with ELO in place it's basically guaranteed to be an unfair match-up because you're using 16 ratings based on normal behavior when 8 of them on one team are gaming the system unfairly.

All that aside, it's also against the rules and it's pathetically unsportsmanlike.

[Steel Claws]

Roland, on 23 May 2013 - 08:38 AM, said:

One thing, aside from the grouping issues, that I think needs to be said....

Most of the time, when a team gets roflstomped, it's not because one team had a group and the other didn't.

Most of the time, when you look at the scoreboard afterwards, it's because you had players on the losing team who were driving freaking HUGE mechs, and did freaking NOTHING with them.

If you take an atlas, and do 20 points of damage before you die, then you are a huge reason for your team losing. You should not be driving that mech. You need to take a smaller mech, and learn to play better, before you concentrate so much power into your hands. With great power comes great responsibility and all that jazz.

Can't agree more. Let me first say that I play mostly solo. Once in a while I will join some of my team members but their interest has wained lately.

This is a TEAM game but I really doubt your seeing that many 4 man teams. Two or three, yeah those are more common. Right now there are damn few four man teams playing. Certainly not enough to get paired up very often.

I'd also really like to knwo how someone who is playing the game in a manner that the creator wants is an exploiter?

Edited by Steel Claws, 01 June 2013 - 05:48 PM.

[Ghogiel]

Panzerman03, on 01 June 2013 - 10:31 AM, said:

More garbage. Your argument is only true if these players only play in sync dropped double 4-man teams. ELO gives ratings based on whether or not the outcome of a match agrees with predictions the system made when it created the match. If you sync drop an 8 man full of players that usually don't get to play with such a massive advantage, you not only get an unfair matchup, but also an invalid rating adjustment for both teams after you pug stomp the daylights out of your opponents.

No. Assuming both teams play to win, the adjustments made after the match for both those teams is fair.

What you are describing is 16 players who are adjusting their Elo rating. It seems you are hung up on the fact that the ratings are in flux. This is how it actually works though. The idea that going into the match the ratings and prediction are always less accurate is correct, what is important is that at the end of the match Elo ratings get adjusted depending on the outcome.

Players are always going into matches with the odds stacked in their favor or against them> you can also do things that can give them advantages, getting better hardware, getting better ping, getting a good nights sleep, getting better mech friends, getting more mech friends, have a strong cup of coffee, blast AC/DC while playing big stompy robits etc. All of these things can affect how well you play. Elo just doesn't care if you do or have any of these.


And no, it doesn't matter if players only play in sync drops or only use their gaming desktop to play instead of their crappy laptop. There is no caveat to playing as a PUG, 2-4 man premades, un/intentional sync drop or any mix there of. Elo only cares about your wins and losses, not about who you play with.

Quote

The system does not account for the sync-droppers' behavior because they almost certainly won't behave the same in the vast majority of their matches where they're failing to successfully sync drop (either through lack of trying or just not pulling it off).

It actually does. The results of every match are recorded and your rating is adjusted, thus their behavior in those matches is accounted for. Player behavior such as playing a mix of PUGs and in premades is only relevant to your enjoyment if the game.

Quote

It's actually probably the case that without ELO you'd sometimes see random 8-man pugs that could tackle a double 4-man sync, but with ELO in place it's basically guaranteed to be an unfair match-up because you're using 16 ratings based on normal behavior when 8 of them on one team are gaming the system unfairly.

Again, you seem to be hung up on the fact that players are not always static in their Elo rating. Elo ratings are adjusted is what matters.

3 people and I can start new accounts, buy some mechs and rofl stomp noobs in trial mechs at low Elo ratings for 500 matches, that doesn't mean that those games are unfair match ups. What it means is my Elo rating needed adjusting.

[Mynder]

The answer can't be to banish premades. The answer should be to make random groups able to work as a team, i.e. provide the necessary tools such as voice-comm IN-GAME.

Even for those who don't want to participate, don't have a headset/mic, it would still mean they can hear their team mates and REACT to what is said. Somebody calling out vulnerable/exposed targets usually makes a tremendous difference to how a battle plays out.

A quick "don't walk up the ridge at <coordinate>, they're all lined up behind it" can keep your team from getting butchered.

In my book, this is ESSENTIAL in a team game and constitutes the biggest advantage premades have over PUGs. This should be a very high priority, as it will close that gap considerably.

[Keifomofutu]

If someone spends half their time pubbing and half their time in premades then their Elo will be wrong for every single game they play. They'll be overrated when pubbing and underrated when playing in premades.

Elo doesn't work, white knight it all you want. It's a system designed to balance a single player against a single player and both have the exact same loadout every time.(chess)

It fails at balancing 16 players with various mechs and loadouts and various levels of teams and teamwork.

Edited by Keifomofutu, 02 June 2013 - 08:01 AM.

[Ghogiel]

Please tell us more about these bald assertions you just made.

[Panzerman03]

Ghogiel, your argument is largely missing the point. The entire point to a hidden ELO system like we have is to create fair/balanced/even match-ups in order to facilitate player enjoyment of the game. There is no other reason for it to exist - it's not a public ranking system or a ladder. If you tank your ELO, roll a smurf account, or game the matchmaker in any other way (including sync dropping) you actually are creating brutally unfair match-ups. Are you suggesting that if 7 of my unit buddies and I rolled new accounts, autolost 25 games then bought optimized mechs and sync dropped that the following match would be fair?

You and I don't disagree that ELO doesn't care why you win or lose, but I'd argue that this is one of the reasons it's possible to game the system - a sync drop curbstomp can be hugely enjoyable for the team that did it and the MMR adjustment with which they "pay" for their behavior is no different than any other win or loss that can happen for myriad reasons.

I get that the ELO system is no guarantee of a fair or even match, and that MMRs are constantly in flux and adjusting but you also can't deny that it goes a long way towards creating balanced match-ups. Sync dropping (and again, I'm not talking about 2-4 mans) is deliberately avoiding the match pool where 8-man teams are supposed to play in order to subvert the ELO system to get an easy win.

Edit: Also, thanks for the rational, civil discussion on this.

Edited by Panzerman03, 02 June 2013 - 11:52 AM.

[DeaconW]

Panzerman03, on 02 June 2013 - 11:50 AM, said:

Ghogiel, your argument is largely missing the point.

That is an understatement! I think he's just trolling...hard to believe anyone can really be that obtuse. The other thing he is completely ignoring is that the MM will *ignore ELO* at some point to get a match, thereby negating the point of ELO altogether!

[blinkin]

DeaconW, on 02 June 2013 - 01:12 PM, said:

That is an understatement! I think he's just trolling...hard to believe anyone can really be that obtuse. The other thing he is completely ignoring is that the MM will *ignore ELO* at some point to get a match, thereby negating the point of ELO altogether!

that isn't really a failing of ELO though. if there are only 16 people searching for a match when you are searching the system can either put you with those people or tell you that you are SOL.

i think a large part of the issues described seems to be coming from the fact that the pool of players is just too small for balanced matches to be likely.

edit: also people get grumpy when they are not winning ALL OF THE TIME and a decent system will have you win about 50% of the time and your KD should hover right around 1.00

Edited by blinkin, 02 June 2013 - 01:58 PM.

[Shakespeare]

blinkin, on 02 June 2013 - 01:56 PM, said:

that isn't really a failing of ELO though. if there are only 16 people searching for a match when you are searching the system can either put you with those people or tell you that you are SOL.

i think a large part of the issues described seems to be coming from the fact that the pool of players is just too small for balanced matches to be likely.

edit: also people get grumpy when they are not winning ALL OF THE TIME and a decent system will have you win about 50% of the time and your KD should hover right around 1.00

Not to quibble, but K/D is more about playstyle and chassis preference - there's a reason ELO uses W/L as a metric, as a good light pilot isn't necessarily the one doing all the killing. But I agree with you in principle: it's instinct to try and interpret streaks of wins or losses as attributable to outside forces, but a good matchmaker should keep you with the sort of people that can make for a good game, instead of a stomping. Much as I think the MM needs work, particularly in the weight distribution department, it has kept my wins at around 1.2 per. That said, 8-mans definitely boosts my numbers vs my KDR, since I can die early or without getting a kill and still win, and PUG in medium mechs surely drags my w/l down.

So, statistically, I don't think its doing a bad job, as compared to nothing or pure random. I do, however, think there's a lot to be done in order to get good, fun matches as the norm.

[krash27]

CapperDeluxe, on 23 May 2013 - 07:05 AM, said:

i occasionally play solo as well, because sometimes the friends aren't available at that time. I don't think anything different of it other than I need to be able to coordinate over text chat (or just try to "read" the motions of my team). Thing is that its focus fire that tends to bring down the solo player compared to the team player. As such solo players need to be more observant of which mechs their team is firing at and go for them as well.





Exactly! The only time I'll use the "close" remark in that scenario is if the other team was a jerk about the loss.

So what your saying is you roll in their with your 4 man and roflstomp the pugs and they are jerks??

blinkin, on 02 June 2013 - 01:56 PM, said:

that isn't really a failing of ELO though. if there are only 16 people searching for a match when you are searching the system can either put you with those people or tell you that you are SOL.

i think a large part of the issues described seems to be coming from the fact that the pool of players is just too small for balanced matches to be likely.

edit: also people get grumpy when they are not winning ALL OF THE TIME and a decent system will have you win about 50% of the time and your KD should hover right around 1.00

Hmm I wonder if this small player pool is a result of being roflstomped and people leaving?

Edited by krash27, 02 June 2013 - 02:13 PM.

[blinkin]

Shakespeare, on 02 June 2013 - 02:09 PM, said:

Not to quibble, but K/D is more about playstyle and chassis preference - there's a reason ELO uses W/L as a metric, as a good light pilot isn't necessarily the one doing all the killing. But I agree with you in principle: it's instinct to try and interpret streaks of wins or losses as attributable to outside forces, but a good matchmaker should keep you with the sort of people that can make for a good game, instead of a stomping. Much as I think the MM needs work, particularly in the weight distribution department, it has kept my wins at around 1.2 per. That said, 8-mans definitely boosts my numbers vs my KDR, since I can die early or without getting a kill and still win, and PUG in medium mechs surely drags my w/l down.

So, statistically, I don't think its doing a bad job, as compared to nothing or pure random. I do, however, think there's a lot to be done in order to get good, fun matches as the norm.

i remember hearing that ELO had separate scores per weight class, they should probably also consider a separate score for group play.

we just need to get away from all of this flawed "it isn't quite right so throw it out completely" style of thinking that infests the forums.

[DeaconW]

blinkin, on 02 June 2013 - 02:16 PM, said:

i remember hearing that ELO had separate scores per weight class, they should probably also consider a separate score for group play.

I agree.

[Shakespeare]

Funny, as much as I agree, I WOULD like to throw out any and all ELO calculation for 8-man (I assume there is one there). Frankly, there aren't enough of us playing in the evening on the east coast to keep everyone in matches with varied opponents. Last 2 nights I've played in 8s, we faced less than 5 teams total - and for some of the time, it was clearly just rotating between us and two other teams.

Need more 8-man!

[KharnZor]

Soy, on 31 May 2013 - 11:50 PM, said:

HOW THE **** IS THIS THREAD STILL HERE

Indeed. why


WHY

[Ghogiel]

Panzerman03, on 02 June 2013 - 11:50 AM, said:

Are you suggesting that if 7 of my unit buddies and I rolled new accounts, autolost 25 games then bought optimized mechs and sync dropped that the following match would be fair?

Statically, the match up would be fair yes. Well no match in this game is 'fair', ever. I am guessing it is as close to fair as can be worked out mathematically.

This issue, and any like it, isn't even Elo or matchmakings problem, it is stricter moderation, something that can deter players who just want to derp around rather than play as serious gundam. That is if you actually care about how other people play the game...

Quote

You and I don't disagree that ELO doesn't care why you win or lose, but I'd argue that this is one of the reasons it's possible to game the system - a sync drop curbstomp can be hugely enjoyable for the team that did it and the MMR adjustment with which they "pay" for their behavior is no different than any other win or loss that can happen for myriad reasons.

I get that the ELO system is no guarantee of a fair or even match, and that MMRs are constantly in flux and adjusting but you also can't deny that it goes a long way towards creating balanced match-ups. Sync dropping (and again, I'm not talking about 2-4 mans) is deliberately avoiding the match pool where 8-man teams are supposed to play in order to subvert the ELO system to get an easy win.

Edit: Also, thanks for the rational, civil discussion on this.

Even if a team were to sync drop, it is not flat out easy wins. It could be an easy win yeah. It could be the team that has the higher probability of winning. Or it could be a loss.

Avoiding the 8man queue if it's not 5-7 man syncs, sure, but statistically any sort of match up un/intentional sync match ups would be matched like any other match, and adjusted depending on outcome.

DeaconW, on 02 June 2013 - 01:12 PM, said:

That is an understatement! I think he's just trolling...hard to believe anyone can really be that obtuse. The other thing he is completely ignoring is that the MM will *ignore ELO* at some point to get a match, thereby negating the point of ELO altogether!

It doesn't ignore it, it expands the threshold to certain limits before it fails to find a match. The reason why that is done is so we don't have to wait around for 20mins to get in a match. They could restrict it so it never finds a match unless the Elo ratings are very close, but there was a lot of complaining about the 10mins looking at the hamster wheel so they tweak it.

Also note that the adjustment made to the Elo scores that happens if 2 teams are of a very different rating is either very tiny or a lot depending on who won.

Edited by Ghogiel, 02 June 2013 - 11:12 PM.