Jump to content

Is This Part Of The Steamroller Problem?


26 replies to this topic

#1 JC Daxion

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Liquid Metal
  • Liquid Metal
  • 5,230 posts

Posted 17 May 2013 - 06:05 PM

I wonder if part of the problem with the completely one sided matches is. could the match maker be working to fast?

It takes barely a min to get a game.. How can it be really selecting people of your level that fast? are there really 5k people jumping in the que at the same time? just waiting to go, so it can make decent matches?

I find that unlikely as i often see people in the same match, match after match.. and some are incredibly good.. at the same time i see people saying it is their first day, in the same match..

Often i see 4 man teams show up in matches over a night, and all incredibly good players, so it is pretty obvious they are grouped and on TS, or voice.

what if the Que slowed down a bit, and it took a few mins, and stuck you in your matches lobby while you waited for more people at your level. would also give time for people to talk a little, make strats, and just give a little more social aspect to the game.

I have been on both sides of the steam roller, and honestly neither are really that much fun.. I wonder what can be done to make matches more competitive


I know improved match making is coming.. So it had me thinking, what about a Singles only game.. by that i mean, no pre grouping allowed. Give people options to play against other pre-made groups as well. I have to think there aer more than a few 4 mans going on a busy night.. just as there must be enough singles to fill ques as well.

#2 Redshift2k5

    Welcoming Committee

  • PipPipPipPipPipPipPipPipPipPipPip
  • Stone Cold
  • Stone Cold
  • 11,975 posts
  • LocationNewfoundland

Posted 18 May 2013 - 04:39 AM

There is a balance between finding an even match and finding a match in a timely fashion. The matcher actually gets LESS PICKY the longer you are in a queue. If it instantly finds a match that's a good sign! If it takes a little longer then it is still looking, and getting less picky the longer it searches.

You need to realize though that two eveny matched teams can still easily cause a lopsided match. A single tactical error, or bad luck, can leave a team down two players, making the match now 8 vs 6, and making it even easier for the team with a larger force to focus fire and bring it to an 8 vs 5, 7 vs 4, and so on. These can still be exciting and fair matches, but the fact is once a team has secured 2-4 kills, their chances of an 8-0 match become very high. Does not mean the matchmaker was poorly matched.

The devs said less than 2 percent of matches are falling into "poor matches" by the matchmaker since the last matchmaker chance. So, 98 percent of the time, you are seeing what you should be seeing, which includes 8-0 matches

A lobby still has it's uses, but it shouldn't be necessary to get an even match. A lobby system for socializing, for deciding upon the next map, and strategizing, is being planned.

#3 Hauser

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 976 posts

Posted 18 May 2013 - 08:26 AM

View PostRedshift2k5, on 18 May 2013 - 04:39 AM, said:

There is a balance between finding an even match and finding a match in a timely fashion. The matcher actually gets LESS PICKY the longer you are in a queue. If it instantly finds a match that's a good sign! If it takes a little longer then it is still looking, and getting less picky the longer it searches.


Nope. There is nothing you can infer from the waiting time. If its short there might be allot of people playing and you can be put into a match quickly. Or you might be the desperate pick of a match maker that has been waiting for a long time.

#4 Padic

    Member

  • PipPipPipPipPipPip
  • 391 posts
  • LocationColorado

Posted 18 May 2013 - 08:33 AM

View PostRedshift2k5, on 18 May 2013 - 04:39 AM, said:

You need to realize though that two evenly matched teams can still easily cause a lopsided match. A single tactical error, or bad luck, can leave a team down two players, making the match now 8 vs 6, and making it even easier for the team with a larger force to focus fire and bring it to an 8 vs 5, 7 vs 4, and so on. These can still be exciting and fair matches, but the fact is once a team has secured 2-4 kills, their chances of an 8-0 match become very high. Does not mean the matchmaker was poorly matched.


Thank you thank you thank you thank you.

This. A match with a lopsided score does not conclusively mean that the match had lopsided players - hell, it doesn't even necessarily mean that the match was lopsided. I've seen really hard fought games with extremely lopsided scores .

#5 Appogee

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 10,966 posts
  • LocationOn planet Tukayyid, celebrating victory

Posted 18 May 2013 - 08:36 AM

I agree that one-sided matches are the single-biggest turnoff for the game as it stands.

It is very disheartening to see your noob team make all the usual noob mistakes and get slaughtered in the first 2 minutes, and be left with absolutely no chance of getting a kill, let alone a decent score.

I've just registered to join a clan to try and escape morale-destroying steamrollering night after night.

#6 Redshift2k5

    Welcoming Committee

  • PipPipPipPipPipPipPipPipPipPipPip
  • Stone Cold
  • Stone Cold
  • 11,975 posts
  • LocationNewfoundland

Posted 18 May 2013 - 08:42 AM

View PostAppogee, on 18 May 2013 - 08:36 AM, said:

I agree that one-sided matches are the single-biggest turnoff for the game as it stands.

It is very disheartening to see your noob team make all the usual noob mistakes and get slaughtered in the first 2 minutes, and be left with absolutely no chance of getting a kill, let alone a decent score.

I've just registered to join a clan to try and escape morale-destroying steamrollering night after night.


If your noob team is killed 8-0, doesn't mean the other team was any more skilled, just lucky.

Quote

In my last post, I shared a statistic: That 74% of our matches were within “tolerable” weight balance. When we started matching with the blue curve, that number spiked to 88%. Beyond that, less than 2% of our matches now fall into the “horrendously bad” category we described earlier – that’s down from the over-7-percent number it was before. The difference in Elo for most matches produced is also is less than half of what it used to be.


Less than 2% of matches are falling into a failed matchmaker. http://mwomercs.com/...19#entry2265319

#7 Appogee

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 10,966 posts
  • LocationOn planet Tukayyid, celebrating victory

Posted 18 May 2013 - 08:43 AM

View PostRedshift2k5, on 18 May 2013 - 04:39 AM, said:

You need to realize though that two eveny matched teams can still easily cause a lopsided match...
Yes, that can happen. But I don't believe that's the main reason for steamrolling.

Night after night I see clearly lop-sided matches - especially premade veterans vs a bunch of randoms. It's easy to detect when the match-making screen comes up and you see a bunch of founders in the same faction all ready to go, while it slowly adds additional single names to your team. Then the game starts, the premade team clearly works to focus fire.

This is a failure of matchmaking, not a failure of the solo players.


View PostRedshift2k5, on 18 May 2013 - 04:39 AM, said:

The devs said less than 2 percent of matches are falling into "poor matches" by the matchmaker since the last matchmaker chance.

In which case, they need to revise their definition of ''poor match''. There is no way 98% of games start out relatively evenly matched at present.

Edited by Appogee, 18 May 2013 - 08:48 AM.


#8 Tarl Cabot

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Tai-sho
  • Tai-sho
  • 7,763 posts
  • LocationImperial City, Luthien - Draconis Combine

Posted 18 May 2013 - 02:35 PM

View PostAppogee, on 18 May 2013 - 08:43 AM, said:

Yes, that can happen. But I don't believe that's the main reason for steamrolling.

Night after night I see clearly lop-sided matches - especially premade veterans vs a bunch of randoms. It's easy to detect when the match-making screen comes up and you see a bunch of founders in the same faction all ready to go, while it slowly adds additional single names to your team. Then the game starts, the premade team clearly works to focus fire.

This is a failure of matchmaking, not a failure of the solo players.



In which case, they need to revise their definition of ''poor match''. There is no way 98% of games start out relatively evenly matched at present.


Answer this question for the bold part. Have you even seen a drop where your team was all filled out and the other side only had one or two players and slowly filled out?

I haven't. If there were time, I wonder what everyone else sees, on both sides.

#9 trollocaustic

    Member

  • PipPipPipPipPipPip
  • 312 posts

Posted 18 May 2013 - 02:53 PM

It's actually a snowball effect, as even one mech lost means you'll fall prey to Lanchester's Square Law as they'll have more guns to tear you down, that's why focus fire and moving in groups is so important, no matter what the lighters (Conspiracy theorists) say.

#10 Blacke

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 238 posts
  • Location3rd Takata Lancers

Posted 18 May 2013 - 04:56 PM

Most of my matches lately have been much more even with it coming down tot he alst 2 or 3 mechs.

#11 TygerLily

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 2,150 posts

Posted 18 May 2013 - 05:43 PM

View PostHauser, on 18 May 2013 - 08:26 AM, said:


There is nothing you can infer from the waiting time.


Actually, he's not inferring but referring to a dev's own update here.

Specifically the quote

Quote

The way the match maker works it tries to group players within certain thresholds and grows those thresholds over time to ensure a match is made within currently a 2 minute time period.


And then:

Quote

currently after 2 minutes you may get an unbalanced match as the match maker increases the thresholds dramatically to ensure you get a match this is as designed to ensure that you don't wait longer than 2 minutes to find a match.


#12 Just wanna play

    Member

  • PipPipPipPipPipPipPipPip
  • 1,520 posts
  • LocationInside the Womb of a Great Turtle

Posted 18 May 2013 - 07:25 PM

View Posttrollocaustic, on 18 May 2013 - 02:53 PM, said:

It's actually a snowball effect, as even one mech lost means you'll fall prey to Lanchester's Square Law as they'll have more guns to tear you down, that's why focus fire and moving in groups is so important, no matter what the lighters (Conspiracy theorists) say.

and when you have someone disconnect (especially an assault) it just screws you over and matchmaking cant do anything in that case

few simple suggestions for people trying to take "advantage" of matchmaking

first off, understand how its matched based on weight (those 2% horrible matchs they mentioned are whenoney team has the weight advantage equivalent to 2 assault mechs, but you can still possibly win, but most pugs dont take advantage of their mechs to do so)
ok ever been in a match where the enemy team seems to be mostly mediums and bigger but no assualts and your team is mostly lights and 1 or 2 assaults??? your team has a very important responsibility in this case, keeping the assault mech alive!!! most of your team mates are lighter to allow that assualt mech to be on your team, so while their weight is mostly spread out, yours is focus in that assault, and so he ccan beat anyone on the other team (if hes competent that is) 1 on 1, but he is als more likely to lose when out numbered because hes also slower then anyone on the other team, and once you lose him your at a major disadvantage even if you've already killed one of them because hes more important then any single mech on either side

and even if he is noobish, that doesnt mean leave him to die and try to cap for the win, a little advice goes a long way, and again, he can easily help you kill anyone that gives you any trouble, especially if hes in a trial because then you dont have to worry about him being a good for nothing boat, remind him to let the fastest mechs get the party going and strip mechs armor off first, THEN let the big guy come in and finish them off, keep in mind an assault needs his armor a lot more then a light does so its best for the assualt mech to be fairly conservative in this case, overall remind him not to go charging in and try to stay back, its very effective to slowly lure out enemy mechs, first get the fastest lights chasing you then lead them to the killing ground where your assault mech is, and3since you are faster in then them most of them wont be catching up, then work your way up to bigger and bigger mechs on the enemy team


now then, if your on the opposite team, try very hard not to get caught alone and try to kill their assault mech even at the cost of your own life because hes worth more then you, and remember FOCUS FIRE, unlike your team their tonnage is so focused that them losing one player could be equal to you lossing 2, and or since their other mechs are lighter its possible for them to die a lot easier as well

think of each mechs tonnage as a seperate amount of health contributing to total team tonnage (and total team health) and total team tonnage is almost always nearly the same, or almost the same, but different teams have it spread out differently
the death of bigger mechs count as a bigger loss in health then smaller mechs

you can kill lights faster then assualts so killing them is like shooting an ac2 at their health when killing them quickly, but that requires certain condition and the ability to constantly shoot, while killing their assault is like shooting an ac/20 at their health bar but the low reload time means you should torso twist (disengage or not have other enemies around in the first place) away to keep damage to a minimum until your ready to take another chunk of their health out, doesn't do the damage often but does a lot of it when it does do damage, so the key is to only meet them at a rate at which your "fire rate" (speed at which you kill people) isn't whats slowing you down from lowering their team health to zero and wasting time while takeing dmg

singling them out (bigger mechs in specifics) constantly will allow you to use the "ac/20" efficiently, but when their are lots of easier to kill targets its best to be an ac/2 (kind of comparable to a pack of lights constantly chipping away at enemies, then the ac20 (in this case the assault) finishes off the enemy with its higher burst damage weapon)

is that easily understood or should i redo it??? my bad if its kind of hard to completely understand

Edited by Just wanna play, 18 May 2013 - 07:30 PM.


#13 Wildstreak

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Civil Servant
  • Civil Servant
  • 5,154 posts

Posted 18 May 2013 - 07:49 PM

View PostRedshift2k5, on 18 May 2013 - 08:42 AM, said:

Less than 2% of matches are falling into a failed matchmaker. http://mwomercs.com/...19#entry2265319

I think the definition of FAILED needs to be more specific.

Been playing a lot of Conquest since my Hunchbacks do better there, the number of times I get put on a team with no Lights and stacked with Heavy and Assault is silly plus such a team does not win as easily as you think. I don't mean by Point loss either. I have even been on Conquest teams of no Lights, I am the only Medium and we get decimated.

#14 Hellen Wheels

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,326 posts
  • LocationDraconis March

Posted 18 May 2013 - 07:49 PM

View PostHauser, on 18 May 2013 - 08:26 AM, said:


There is nothing you can infer from the waiting time.

Oh, BS. the longer the wait, the fewer players available.

Period.

And the waits get longer, and longer, and longer.....

You must have rose-colored blinders on to ignore the plain fact that is inferred.

Fewer Players Available.

#15 Just wanna play

    Member

  • PipPipPipPipPipPipPipPip
  • 1,520 posts
  • LocationInside the Womb of a Great Turtle

Posted 18 May 2013 - 07:55 PM

Some more advise thats when not in game

not only should you pick weapons for their damage/tonnage ratio, but your mech itself should be a very good use of tonnage

for example, if you put a 360 engine (the max engine) in an atlas then put an engine in a phract that makes it just as fast, the atlas only has 6 more spare tonnage left to use on armor and weaponry then the phract, but weighs 30 tons more!

another example, if you put the max engine in the awesome 9m then make a phract 2x go just as fast, if they are both standard engines, the phract will actually have more spare tonnage for everything else then the awesome that weighs 10 tons more!

i also highly recommend support mechs being as small as possible while still being powerful, and also lighter then what they will be supporting (yes you lrm boat stalker) if you look at a catapult then think about how much bigger the guys that actually need the tonnage (brawlers and such) would be if you used the cat instead of the 85 ton dong tube, you might realize why lrm boat stalkers and other assualt boats (maybe not the awesome) are not that smart use since the 20 tons saving could be the difference between having an awesome in the game and having an atlas while you could still be only slightly worse while the brawlers on your team would be a lot better

overall the heavier the mech the more important the tonnage advantage, look at lights, its actually really smart to use a commando since it can easily do the job of other lights because its plenty fast like jenners and raven wit decent weaponry and you are taking one for the team to let others use bigger mechs (or your making the enemy team lighter as well), think of it this way, you could have a jenner with 23 damage and a top speed of 113kph and an awesome with 46 damage with 480 armor on your team or you could have a command as fast as the jenner with 20 (sometimes more then ravens and Jenner actually, as well as better range of motion when aiming) damage and a highlander with 53 damage and 560 armor on your team

and to those of you that give your atlas d-dc 2 lrm 15s 1 lrm 10 and 2 large laserz, wtf where you thinking, how did you waste so much tonnage???

View PostMerchant, on 18 May 2013 - 07:49 PM, said:

I think the definition of FAILED needs to be more specific.

Been playing a lot of Conquest since my Hunchbacks do better there, the number of times I get put on a team with no Lights and stacked with Heavy and Assault is silly plus such a team does not win as easily as you think. I don't mean by Point loss either. I have even been on Conquest teams of no Lights, I am the only Medium and we get decimated.

i thought it was pretty specific

"about 7% are what we would call "horrendously bad" (i.e., they carry the weight difference of one or two full assault Mechs between teams)."

so think total team tonnage being higher by ALTEAST 160 tons on one side

thats each person being on one side being 20 tons heavier then those on other team


(think entire team of awesome vs entire team of atlases)

#16 WVAnonymous

    Member

  • PipPipPipPipPipPipPipPip
  • Wrath
  • Wrath
  • 1,691 posts
  • LocationEvery world has a South Bay. That's where I am.

Posted 19 May 2013 - 12:17 AM

Guys like me are the variable the matchmaker can't fix. Tonight I have been working My Atlas DDC up to Master. Three games with 4 kills, a game with 3 kills. I also bought a Pretty Baby back when they put them on sale. 0.82 KDR, just finished Eliting it. My Awesome 8Q 0.5 KDR. 9M... 0.2 KDR :)

All of these builds are "the same" as far as Elo score goes: Assaults. But I do not, by any means, get the same performance out of every assault, even after they are maxed out. I can say similar things about Cicadas, Centurions, and Hunchbacks, and OMG the difference between a Dragon and a Catapult.

Per some of the comments earlier about if a team goes down by 2 or 3 quickly, that's usually conclusive. If someone as flaky as I am lands in the wrong bucket, that's all it takes.

Matchmaker does what it can, but it can't do it all.

#17 Hauser

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 976 posts

Posted 19 May 2013 - 02:32 AM

View PostTygerLily, on 18 May 2013 - 05:43 PM, said:

Actually, he's not inferring but referring to a dev's own update here.


I was commenting on the short wait time. You can't infer from that you'll have a good match. You could be the person a desperate matchmaker will take just to fill out the team. If you have a long waiting time it can be a bad matchup.

#18 WaddeHaddeDudeda

    Member

  • PipPipPipPipPipPipPipPip
  • 1,567 posts
  • LocationAllocation Relocation Dislocation

Posted 19 May 2013 - 02:46 AM

Hey PGI, here's an idea: f*** your matchmaker and finally give us lobbies or something similar!

#19 armyof1

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • 1,770 posts

Posted 19 May 2013 - 03:02 AM

View PostWVAnonymous, on 19 May 2013 - 12:17 AM, said:

Guys like me are the variable the matchmaker can't fix. Tonight I have been working My Atlas DDC up to Master. Three games with 4 kills, a game with 3 kills. I also bought a Pretty Baby back when they put them on sale. 0.82 KDR, just finished Eliting it. My Awesome 8Q 0.5 KDR. 9M... 0.2 KDR :)

All of these builds are "the same" as far as Elo score goes: Assaults. But I do not, by any means, get the same performance out of every assault, even after they are maxed out. I can say similar things about Cicadas, Centurions, and Hunchbacks, and OMG the difference between a Dragon and a Catapult.

Per some of the comments earlier about if a team goes down by 2 or 3 quickly, that's usually conclusive. If someone as flaky as I am lands in the wrong bucket, that's all it takes.

Matchmaker does what it can, but it can't do it all.


Elo for entire weightclasses is certainly a mess, since as you wrote you can have so varied success with different mechs in the same weightclass. There really should be a separate Elo for different mechs and ideally for different variants too, since I doubt anyone is having the same success with a Spider-K as they do with a Spider-D.

#20 Deathlike

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Littlest Helper
  • Littlest Helper
  • 29,240 posts
  • Location#NOToTaterBalance #BadBalanceOverlordIsBad

Posted 19 May 2013 - 07:16 AM

View Postarmyof1, on 19 May 2013 - 03:02 AM, said:


Elo for entire weightclasses is certainly a mess, since as you wrote you can have so varied success with different mechs in the same weightclass. There really should be a separate Elo for different mechs and ideally for different variants too, since I doubt anyone is having the same success with a Spider-K as they do with a Spider-D.


Imagine the ELO you would have if the 6MG Spider existed!





18 user(s) are reading this topic

0 members, 18 guests, 0 anonymous users