76 replies to this topic

[SgtKinCaiD]

Same thing as before : for one good match, and i'm not talking about W/L but rather not being teamed up with total incompetents, i get 3 sore matchs : steamrolling the ennemy team or getting steamrolled, being put with a AS7-D-DC only armed with an AC-10 and 2LL and no ECM, half of the team doing less than 100 dmg ...

[Kinilan]

Your matchmaker is still broken. 8/0 8/1 matches are just as common if not more so. Games are still launching with 1-2 players missing on a side and pre-mades still ruin the experience. This is no better then it was before.

[Splinters]

It's getting stranger now as we get 1-2 disco's per match and we have some strange ELO-balanced teams.

Today I was in a rather unfortunate match of being the only assault on my team vs 6 assaults on the OPFOR. Where does MM figure out that an equivalent ELO assault can be offset with a Hunchback?

-S

[Xenon Megablast]

There is definitely going something wrong with Elo.

I'm a rather new pilot, started playing on 13. Apr 2013. For my first almost two weeks - or in numbers - about ~700 matches, I never even had to think about the deeper intricacies of matchmaking, as my matches were mostly ending with the one, or the other side, after lot of fighting, being left with 1-2 mechs still alive. Which you would, at any rate I think, call a "good fight".

There were no noticeable issues, we were winning, we were losing, things were working as one would normally expect.

Now, for the past few days, or say, the last ~100 matches, as something presumably changed in my Elo rating dramatically at the point of learning to pilot properly (and my kill ratio getting way up), two very exclusive things started to happen:

For about 90% of time I'm ending up placed in a team that gets *completely* slaughtered without making even a single kill, with me usually being left as the last mech alive (note that I usually move out and fight as one of the first mechs, I'm not exactly hiding in the base). This happens over and over, easily for 5-6 matches in a row.

It's always entire teams getting wiped, rarely managing to make a sole kill. People completely ignoring any tactical data I'm relaying, ignoring even the basic objectives, people aimlessly wandering around the battlefield, or just shooting randomly each other. As now almost a common situation, during a conquest map, I end up in a team that as a whole moves to the center of the map and for the whole duration (~10 minutes) refuses to cap a single point, despite repeated pleas, leaving even the opposite team equally puzzled (such matches end without a single shot fired, the other team simply circling around the massed static blob, capping the whole map instead).

Or durring assaults, I rarely even manage to make a single kill (without pulling a lone mech somewhere aside of the main battle), as people around me just blindly run into enemy fire, guns blazing, with half of the team exploding in the first minute of the match, the scattered rest soon to follow.

Now for the remaining 10% of the cases, mirrorring this situation, I end up in a team, that again completely wipes the other one without taking a single loss (or up to one, once in a while). In these cases, I usually end up with a score of 3 kills + 5 assists, or even at the "worst" case, with 8 assists.

And there is nothing in between.

This basically makes the game unplayable, leaving it to just grinding CB and XP for additional mechs and unlocks, but hardly having any fun while doing it.

Please, whatever is going on here, fix this. And let me again play with the "normal" people that I've been playing with for the initial two weeks.

[Simbacca]

I find that 60% of the matches tend to be lopsided.

Either due to players or mechs.

Was on Caustic Valley, just a few minutes ago. Foe had 3 Assaults, heavies, 1 light. We had no assaults, 2 lights, some mediums and heavies. We lost badly.

The use of Win/Loss ratio modifier is a poor measure, as in theory, a player could actually do nothing the whole match yet still win.

The Kill/Death ratio would be a much better measure for a modifier.

[Xenon Megablast]

Simbacca, on 27 April 2013 - 06:36 PM, said:

The use of Win/Loss ratio modifier is a poor measure, as in theory, a player could actually do nothing the whole match yet still win.

The Kill/Death ratio would be a much better measure for a modifier.

That sounds like an extremely bad idea for light pilots. The role of a light is not to engage a group of 8 dug up assaults head on, but scout, cap points and bases, and provide ecm or light fire support when possible. Based on your K/D approach, most lights pilots would end up classified as bad players, that "do nothing the whole match, yet still win".

[Kageru Ikazuchi]

@ Xenon Megablast: Your Elo score is different for each weight class ... did you change mechs recently?

Also, the player base changes on the weekend, depending on the time of day, matches can be either more erratic or better.

All that said, a very significant portion of the player base--including almost all new pilots--is in the middle of the Elo score range.

Not saying it's right ... I think they still have quite a bit of tweaking to do before the system is working the way they want, where both experienced players and new players have a more or less positive experience.

[Xenon Megablast]

Kageru Ikazuchi, on 27 April 2013 - 09:10 PM, said:

@ Xenon Megablast: Your Elo score is different for each weight class ... did you change mechs recently?

That's interesting, I did exactly that! This might be a very valid point, I wasn't aware of different Elo between classes.

Yet that still raises a question, why did I experience the opposite when playing lights? From day one, I was put in teams that rather frequently discussed their tactics, relayed enemy positions, at points regrouped or pulled back when appropriate, were covering damaged teammates... Which kind of points to those not being rookie players (which retrospectively brings up the question, why was I being placed there in the first place, but not that I can complain). And now since moving to another class, I constantly land in games that end in utter massacres... There must be something wrong.

As for the different player composition on weekends as you mentioned, I was thinking about that too, I'm aware that weekends are mostly the time of "casual gamers", experiences from other MMOs support that view. But even then, if ELO was working, right, I would expect about the same number of "weekend players" on both sides. Say, if you see five people randomly milling around and running into walls on your team, you don't expect the other team walking in, in perfect formation, lasers blazing, tearing everyone apart and ending the match within first minute...

Guess I'll have to see how (if) things change in the coming days, based on the new class Elo...

[SmaSol]

I'm playing (mostly) solo for a few days, because I'm too busy to keep up with the shape/requirements of my unit.
I'm sharing the opinion of most of the pre-posters - something is wrong in public matches.
It is either 2-8 or 8-2, but nothing in between.
I cannot explain the reason - I've had matches where my team was stomped 2-8. In the next match, I have 1-2 of the previous winners on my side and we get again stomped 2-8.

About the reason I can only speculate. I would assume there are a lot more teams of 4 than I could think.

[Gremlich Johns]

still seeing random tonnage disparities. Not all the time, but often enough to notice.

[aelfkins]

Huge disparities in tonnage and skill, since the last matchmaking change.
Winning or loosing 8-1 is not fun.

[Kageru Ikazuchi]

I think the biggest frustration for solo PUG players, particularly the new players, is matching up against a roster full of experienced 2-4 man groups ... I have dropped in several games over the past few days where the enemy side was already full, and my side is filling our roster one name at a time. Sometimes 2-3 games in a row against the same eight guys (with only two or three repeats on my side). Those are the games that most often end in a stomp.

I see three possible solutions to improve both new player and PUG experience ...

• New players start with a below-average Elo score (right now, they start at 1300, which according to the graph in the post from Omid is slightly above average, and right in the middle of the largest group of players).

and/or

• Groups are first matched against groups ... if two groups join the queue at the same time, they are matched against each other, and solo PUGs fill in the rest. If one group joins a match, the next group to join will be placed on the opposing team.

and/or

• Solo pugs only join a match if their Elo score is equal to or higher than the average score of an opposing group.

Since I don't know exactly how the matches are actually made, implementation may be harder or more difficult that I think, or there would be some unintended consequences.
Overall, my experience has been positive, and almost every time I get rolled by a clearly better team, I learn something. However, while some of the more experienced players complain about the new solo players not being able to contribute, it is vastly more frustrating to me when I can't contribute because my team is so outclassed in terms of either skill or coordination.

(I posted something similar in a different thread, but I think it most appropriately belongs here.)

[Gabriel Amarell]

danneskold, on 17 April 2013 - 09:25 AM, said:

From my perspective, as a newer pilot, i think on skill level it seems better - but the weight class issue is big. I have a thing for mediums - and I am trying to do something about that - but in the mean time, it is daunting on the light end of match vs 4-5 assaults and heavies....I dont have a very good role in that match up. I have moderate firepower, but not enough armor to last long enough, and am not as fast as a light, so I get stuck in limbo.

Seems to be in the right direction,but weight class match will be more impactful.

now - time to go buy At LEAST a heavy....

Mediums are really not powerful enough to mount heavy weapons, they are not armored enough to survive by armor and not fast/maneuverable to survive by speed/maneuverability, the other huge problem is ecm, as in mediums don't carry it which makes them completely impotent against ecm streak lights. SRM's made mediums effective. They were a hard hitting weapon small and light enough for a medium to carry, now that their effectiveness is negligible mediums are a joke. its a shame because it was my favorite weight class. When you nerf the *&^(* out of missiles what you get is the PPC/Gauss sniper thing thats going on now, because brawling is a joke. SRM's were the brawling weapon, but with them being worthless long range completely dominates.

I'm a good shot and even with Artemis SRM's only 60% of my SRM'*****. They do not hit at all at 270 meters, the max effective range is more like 150. Even at 150 they do no concentrate their damage. So lets see, only 60% of the missiles hit, they don't concentrate damage, they are rediculously short range, and they got a nerf, a nerf of 40%... (and "splash" does not begin to make up for the 1 pt. reduction in damage per missile.) Fix splash so the SRM'***** as hard as they should, or take it out and put the damage per missile back at 2.5. As long as SRM's are a joke brawling is dead and the sniper domination will continue.

Edited by Gabriel Amarell, 04 May 2013 - 08:14 AM.

[Mr Andersson]

Something is definitely wrong with the matchmaker. I want to bring up two points:

1. More than 50% of the matches I play end up either 8 - 0 or 0 - 8 (I only play assault mode). Close-fought matches are extremely rare.

2. In all games I enter the enemy team is always filled up while my team is not. I have observed this over the last 60 matches (with not a single exception). Most of the time my team eventually gets filled up as well, but sometimes we have to play 7, or even 6, against 8. Not a single time have I been in the team with more players.

[XSerjo]

Is there weight-balance or only mystery "average ELO per team"?
Look at this, perfect match:

Spoiler

7 vs 8, 0 assaults vs 4, 385 tons vs 540, 1 light vs 2.

Even single assault can't balance this, how does it works?

P.S.
I agree, all random matches looks like 0/2-8, 8-0/2 sequence, with tiny amount of "close" games.

[Rebas Kradd]

I am almost never seeing unbalanced matches or 8-0/8-1 victories.

[XSerjo]

Also syncdrops still occurs:
8 assaults vs 0, I don't even want to calculate weight mismatch. They also acted as one team, with obvioust result: 8-0.