Jump to content

Matchmaker Is Resulting In Unbalanced Matches


23 replies to this topic

#1 Suko

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 1,226 posts
  • LocationPacific Northwest

Posted 02 March 2013 - 04:21 PM

My buddy and I have been having absolutely awful game balancing in almost all of our matches over the last week. It's usually just him and me (2 man team) and about 4/5 of our games are ending in complete stomps on our team. Like, we MIGHT get 2 kills and they wipe the rest of us out. I'm usually at the top of my team's leaderboard, but this is getting REALLY frustrating. I never had this issue before Elo.

Isn't Elo supposed to make the games more balanced and less one-sided? I've been finding it to be completely the opposite. Just stomp after stomp. I'm not even going to address the terrible tonnage mismatch, as that's something I wrote about in another topic.

PGI, Is this working as intended or is there a known bug in the system?

Edited by ShadowVFX, 02 March 2013 - 04:22 PM.


#2 Redshift2k5

    Welcoming Committee

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

Posted 02 March 2013 - 04:30 PM

The majority of my matches have been much more even and on the whole much more enjoyable.

#3 Accused

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 989 posts

Posted 02 March 2013 - 04:31 PM

Working somewhat as intended.

They are placing players with higher elo with players with less elo. It does take weight into account, but the engine will expand to cover a greater variety of variables very quickly to ensure there is a timely match. At least that's my understanding.

With that said I just played in a all light versus heavy match. We got stomped.

#4 Sagamore

    Member

  • PipPipPipPipPipPipPip
  • The Blood Bound
  • The Blood Bound
  • 930 posts
  • LocationCanada

Posted 02 March 2013 - 05:00 PM

View PostAccused, on 02 March 2013 - 04:31 PM, said:


With that said I just played in a all light versus heavy match. We got stomped.


I played similar match except we were on a conquest map and the lighter team won. We were all assaults and heavies!

#5 Felicitatem Parco

    Professor of Memetics

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 13,522 posts
  • LocationIs Being Obscured By ECM

Posted 02 March 2013 - 05:06 PM

If you are concerned over the "imbalances" caused by the rollout of Matchmaking Phase III (Ranked Scoring), then please review the following threads in the Command Chair:

http://mwomercs.com/...79-matchmaking/ <- The original Matchmaking post
http://mwomercs.com/...-making-update/ <- The Matchmaking Update

i will quote the Update here:


View PostMatthew Craig, on 21 February 2013 - 12:20 PM, said:

I wanted to just help clarify some points about the new match making system as there seems to be some confusion on these points:

Is the match maker taking weight class into account?

Yes the version currently live on production does take weight class into account. 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.

These curves will likely see some tuning for both skill and weight as we monitor the data coming in and work to ensure more and more matches are created within good thresholds. This tuning will happen over the course of upcoming patches.

I'm still getting matches that seem unbalanced.

It will take time for the ratings to stabilize please bear in mind that this will likely happen over weeks not days. Also it's important to clarify that the system is only using your match history since we announced that we enabled it for pre-seeding. If you've been with us since closed beta your skill rating is essentially a few weeks old and only based on what you've done in that time. This is the only way to be fair to everyone.

Also 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.

How does the match maker compose a teams Elo rating, is it average rating or closest to a target?

It's closest to a target value, so the match maker starts trying to make a match for an Elo of say 1300 and will pull in players to those teams closest to those values; however, as mentioned earlier within growing thresholds and those curves will be tuned. Currently it may be a bit 'sloppy' about how it's filling those buckets but over time it will be tuned to be much more precise.

We need to do this carefully over time as generally the cost of precision is time to find a match we want to slowly find a very nice balance between time to find a match and the number of matches that are correctly composed.

Can we expect match making to be tuned?

Yes as just stated we will monitor and tune over time to ensure it is working optimally and as everyone plays more games things will naturally settle out over time.

If your getting consistently hard or easy matches please bear with it while the system adjusts and know that we are monitoring and planning adjustments to ensure everyone gets fair matches.



These weight imbalances are to be expected for the next few weeks. If you would like to provide feedback to the developers regarding your Matchmaking experiences during this Phase III rollout, then please post in the Official Matchmaking Feedback thread.

It should all settle out, though, in due time.

Edited by Prosperity Park, 02 March 2013 - 05:06 PM.


#6 Suko

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 1,226 posts
  • LocationPacific Northwest

Posted 02 March 2013 - 10:54 PM

Prosperity, thank you for giving me this info. However, while I do have a problem with the weight imbalance, this is not the issue of this post. My issue here has been that that vast majority of my games when my buddy and I go out as a 2 man team are total stomps. He and I generally get 1-3 place, but our team can rarely get 2 kills before everyone else is gone. It just seems like the difference in skill (and often times the weight) is totally out of whack. I'll try to keep patient about all this, but I thought it has been about 2 weeks since initial Elo went live, and these kinds of imbalances are crazy.

#7 Primetimex

    Member

  • PipPipPipPipPipPip
  • The Bludgeon
  • The Bludgeon
  • 353 posts

Posted 03 March 2013 - 04:59 AM

I'm with ShadowVFX - whatever this new ELO/MM is it is complete RUBBISH. Most of my matches - I'm fighting hard and BLAM, only 4 minutes later I find myself the last one standing and often there has only been one enemy kill and the rest is untouched. The rest of the matches, everyone's off doing their own thing, eg ridge rushing, rushing to open areas resulting in KFC time for them and so forth.

I have noticed a lot of matches now that no one is communicating much like they used to and I don't see the players I usually play with, in short it is rubbish and now the only way to win is to play in 4-mans and/or sync drops.

I'm just about to quit this game for awhile until this whole debacle's fixed.

#8 Onyx

    Member

  • PipPip
  • Bridesmaid
  • Bridesmaid
  • 42 posts
  • LocationIn the land of Twilight, under the Moon.

Posted 03 March 2013 - 05:18 AM

My matches are....random. Very random.

It's hard to tell if ECM is in play or if matches have severe weight discrepancies whether there's an imbalance due to Elo or due to team stacking with mechs that are best described as overpowered, so I can't safely comment whether Elo is actually working right or not.

For the most part, I'd say it's_ hit and miss. Sometimes it's great, other times it's terrible. The biggest differentiator in running into a slippery slope situation tends to be ECM or Splatapult stacking, however. My Elo seems to be yoyoing a lot presently. Winning game after game until it throws me constantly up against stacked enemy teams where they have stuff that almost guarantees loss after loss regardless of skill, then it drops back down to where it's faceroll easy. It's rather aggravating.

The few matches where it's even are, in fact, very fun. Just wish it didn't constantly hit a point where you get teams stacked with D-DCs and Raven 3Ls when my team has none of these.

As a note, this profanity filter is very, very bad...

Edited by Onyx, 03 March 2013 - 05:27 AM.


#9 Asmosis

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 2,118 posts

Posted 03 March 2013 - 05:23 AM

we've had ELO what a week now? going to take a couple more for it to balance out. Even now though, the matches are a lot more fun.

It occurs to me a lot of the people grumbling about ELO are former pugstompers. HMM.

Edited by Asmosis, 03 March 2013 - 05:26 AM.


#10 Jestun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,270 posts

Posted 03 March 2013 - 05:23 AM

Until everyone in the queue has played enough matches to have a perfectly accurate ELO, there will continue to be imbalances.

Things should improve over time, both through additional data on each person queueing and potential improvements to the matchmaker.

#11 Tolkien

    Member

  • PipPipPipPipPipPipPipPip
  • Giant Helper
  • Giant Helper
  • 1,118 posts

Posted 03 March 2013 - 05:31 AM

View PostJestun, on 03 March 2013 - 05:23 AM, said:

Until everyone in the queue has played enough matches to have a perfectly accurate ELO, there will continue to be imbalances.

Things should improve over time, both through additional data on each person queueing and potential improvements to the matchmaker.



Just to play devil's advocate I have been tracking my win/loss and Kill/death prior to phase 3 and now during to see if it would push win/loss towards 50/50 as we would expect (if we were getting balanced matches for our skill level).

In the approximately 200 games I've played since the intro of phase 3 my win loss has climbed from 62% to 64%. This means that for me at least the matchmaker is less effective than it was in phase 2. Since this is a historical average too, it means that my current win loss since phase 3 is much higher than 64%! I'd crunch the numbers but I'm in a hurry so let's call it 66-70% which is way out of whack for a competitive game.

I'm currently on the happy side of the imbalance, but this means there are people out there suffering due to the system. Moreover it means that whatever is being done in the name of balance is currently less effective (at least for me) than just straight tonnage/class matching.

Edited by Tolkien, 03 March 2013 - 05:34 AM.


#12 Jestun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,270 posts

Posted 03 March 2013 - 05:35 AM

Also, I recall seeing (but have no link to) confirmation that ELO is weight-specific. i.e. your ELO in an assault is not the same as your ELO in a light.

This will obviously mean that anyone who plays multiple weight classes will take even longer to get an accurate ELO.

Not to mention the fact that ELO will never be perfectly accurate as it involves 16 different people in the match and repeating the same match multiple times will not necessarily give the same result (too many variables).

View PostTolkien, on 03 March 2013 - 05:31 AM, said:

In the approximately 200 games I've played since the intro of phase 3 my win loss has climbed from 62% to 64%. This means that for me at least the matchmaker is less effective than it was in phase 2. Since this is a historical average too, it means that my current win loss since phase 3 is much higher than 64%! I'd crunch the numbers but I'm in a hurry so let's call it 66-70% which is way out of whack for a competitive game.


This could be an issue with the matchmaker, or your initial ELO may have just been too low therefore you are playing against "weaker" opponents than before.

*shrug*

#13 Onyx

    Member

  • PipPip
  • Bridesmaid
  • Bridesmaid
  • 42 posts
  • LocationIn the land of Twilight, under the Moon.

Posted 03 March 2013 - 05:46 AM

View PostTolkien, on 03 March 2013 - 05:31 AM, said:

*snip*

Do remember that Elo doesn't guarantee a trend towards a 50% win rate. In fact, if you're better than average, it will guarantee a trend towards 60%, or 70%, or 80%.

When you're a good or bad player, you will constantly have an upward trend in Elo and win rate as you win game after game against people around your current elo. When you finally hit your true skill level, you'll start winning and losing. Once you hit this point, you'll start to decay back to 50%, but then you'll drop down to where you're paired against progressively weaker teams. It's at this point that you start an upward trend again as you start to maintain a slight jitter around your proper, true Elo rating/win rate. Just because your win rate is going up doesn't mean the balance is working improperly, it just means it's not to where it should be quite yet.

#14 Tolkien

    Member

  • PipPipPipPipPipPipPipPip
  • Giant Helper
  • Giant Helper
  • 1,118 posts

Posted 03 March 2013 - 05:51 AM

View PostOnyx, on 03 March 2013 - 05:46 AM, said:

Do remember that Elo doesn't guarantee a trend towards a 50% win rate. In fact, if you're better than average, it will guarantee a trend towards 60%, or 70%, or 80%.

When you're a good or bad player, you will constantly have an upward trend in Elo and win rate as you win game after game against people around your current elo. When you finally hit your true skill level, you'll start winning and losing. Once you hit this point, you'll start to decay back to 50%, but then you'll drop down to where you're paired against progressively weaker teams. It's at this point that you start an upward trend again as you start to maintain a slight jitter around your proper, true Elo rating/win rate. Just because your win rate is going up doesn't mean the balance is working improperly, it just means it's not to where it should be quite yet.


You're describing an under damped system hunting (like a badly designed HVAC system that lets the temperature go up and down by 4 degrees around the set point).

A properly implemented Elo system will have variation due to this being a random process, but it should not diverge and it should not oscillate!



View PostJestun, on 03 March 2013 - 05:35 AM, said:

Also, I recall seeing (but have no link to) confirmation that ELO is weight-specific. i.e. your ELO in an assault is not the same as your ELO in a light.
....
Not to mention the fact that ELO will never be perfectly accurate as it involves 16 different people in the match and repeating the same match multiple times will not necessarily give the same result (too many variables).
...
This could be an issue with the matchmaker, or your initial ELO may have just been too low therefore you are playing against "weaker" opponents than before.

*shrug*


I know I have read that your Elo for solo play is different than your Elo when you are in a group, but I cannot recall reading about it differing for varied weight classes, though that would make some sense too.

Regarding it possibly matching me against weaker opponents than before, we all were seeded at 1300 rating so it should have quickly sorted the good from the bad (if properly implemented).

What has your win-loss done since the Elo patch?

Edited by Tolkien, 03 March 2013 - 05:54 AM.


#15 Heffay

    Rum Runner

  • PipPipPipPipPipPipPipPipPipPip
  • The Referee
  • The Referee
  • 6,458 posts
  • LocationPHX

Posted 03 March 2013 - 05:53 AM

View PostOnyx, on 03 March 2013 - 05:46 AM, said:

Do remember that Elo doesn't guarantee a trend towards a 50% win rate. In fact, if you're better than average, it will guarantee a trend towards 60%, or 70%, or 80%.


That's not right. That will only happen if the majority of your matches are against people with lower Elos because the system can't find enough people in your range to match you up equally. It will only happen at the very outer edges of the rankings, or if there aren't a lot of people playing.

You can be in the top 10%, and if there are enough people playing you will still have a roughly 50% win rate.

#16 Jestun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,270 posts

Posted 03 March 2013 - 05:57 AM

View PostTolkien, on 03 March 2013 - 05:51 AM, said:

What has your win-loss done since the Elo patch?


I don't track my win-loss. If I were playing competitively I imagine I would but it is a meaningless stat for me right now.

Besides, I'm much more casual and I doubt I have the correct ELO yet. Since my join date I've played less than 200 matches and those were split between solo & duo, as well as light & assault.

#17 Onyx

    Member

  • PipPip
  • Bridesmaid
  • Bridesmaid
  • 42 posts
  • LocationIn the land of Twilight, under the Moon.

Posted 03 March 2013 - 06:01 AM

View PostTolkien, on 03 March 2013 - 05:51 AM, said:


You're describing an under damped system hunting (like a badly designed HVAC system that lets the temperature go up and down by 4 degrees around the set point).

A properly implemented Elo system will have variation due to this being a random process, but it should not diverge and it should not oscillate!

Except that it, by nature, does. Especially during the normalization process where everyone's being seeded to their proper Elo rating. Every game you win or lose is oscilating your score, even if only slightly when you get to the end, and the end result is, if your Elo is above average, your win rate will be above average, and the inverse holds true as well. That initial placement and the few corrections that will take place over time guarantee your win rate to trend upwards or downwards if you're respectively a good or bad player.

The fact that this game has random team compositions and random mech builds even further throws a hamper on Elo ever achieving true stability where it can't oscillate enough to allow lose stability. And not all oscillations need be severe. The important thing is that it's not completely stable, but stable enough to place you near where you should be over time, and you'll just stick to 50% win rates, fall to 51% win rates, and rise back up to 50% over the course of hundreds of games. Even still, if you're above average, your win rate will compress once you're placed, but you'll still be an above average win rate player.

#18 Tolkien

    Member

  • PipPipPipPipPipPipPipPip
  • Giant Helper
  • Giant Helper
  • 1,118 posts

Posted 03 March 2013 - 06:09 AM

View PostOnyx, on 03 March 2013 - 06:01 AM, said:

... Every game you win or lose is oscilating your score, even if only slightly when you get to the end, and the end result is, if your Elo is above average, your win rate will be above average, and the inverse holds true as well. ...


This is a central misunderstanding in your thinking: As your Elo converges to a higher value you should be matched against more skilled players until your Elo stabilizes to match you against peers of equal skill. At this point your win loss would in theory converge to 50% as players of equal skill (all else being equal) have a 50-50 chance of winning a given game.

Suggesting that win/loss should converge to something other than 50-50 under an Elo system betrays a deep misunderstanding of what the system is doing.

http://en.wikipedia....o_rating_system

Sorry to be so coarse, but it's true. The only way my win/loss would constantly climb while my rating climbed over hundreds of games would be if I was one of the best players in the world. Trust me when I say I am not.

Edited by Tolkien, 03 March 2013 - 06:10 AM.


#19 Jestun

    Member

  • PipPipPipPipPipPipPipPip
  • 1,270 posts

Posted 03 March 2013 - 06:09 AM

View PostTolkien, on 03 March 2013 - 06:09 AM, said:


This is a central misunderstanding in your thinking: As your Elo converges to a higher value you should be matched against more skilled players until your Elo stabilizes to match you against peers of equal skill. At this point your win loss would in theory converge to 50% as players of equal skill (all else being equal) have a 50-50 chance of winning a given game.

Suggesting that win/loss should converge to something other than 50-50 under an Elo system betrays a deep misunderstanding of what the system is doing.

http://en.wikipedia....o_rating_system

Sorry to be so coarse, but it's true.


Of course, when it cannot find a perfectly balanced match it widens the criteria, so you are not necessarily fighting the appropriate people for your ELO...

#20 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 03 March 2013 - 06:48 AM

View PostAsmosis, on 03 March 2013 - 05:23 AM, said:

we've had ELO what a week now? going to take a couple more for it to balance out. Even now though, the matches are a lot more fun.

It occurs to me a lot of the people grumbling about ELO are former pugstompers. HMM.

I wouldn't call a 62% win percentage PUGstomping. In the last week I have lost 3% more games than normal. I know 3% isn't a big deal, but that is a LOT more losses than usual, so does trigger a WTF response. Which I guess is proportional to the, "Hey, we are winning more!" crowd.. Also if less players are playing 8 man... guess we found where they went ? :ph34r: :wacko:





4 user(s) are reading this topic

0 members, 4 guests, 0 anonymous users