Jump to content

Elo Worthless


298 replies to this topic

#101 Grits N Gravy

    Member

  • PipPipPipPipPipPip
  • 287 posts

Posted 13 November 2013 - 04:39 PM

View PostMischiefSC, on 13 November 2013 - 03:41 PM, said:

For that to work though you'd have to wipe and restart Elo - which is highly unlikely to happen.


You can just convert the scores run a global K factor of 30-40 for a period then go to much a lower K factor. Or use the old system while you build up new data then roll out the new system once you are properly seeded.

View PostMischiefSC, on 13 November 2013 - 03:41 PM, said:

You're still going to get the same sort of situation with premades as well - it's not about premades getting to drop against less skilled players to inflate their score it's that a player in a premade team plays better and has a higher score. So while you may accurately play at a 2000 Elo while in your 4man team your skill is more like 1600 when pugging.

Make sense? It's not about accuracy while in a premade. It's that while you're in the premade you play at one skill level and while you pug you play at another. If you primarily premade it will skew your value when you pug significantly.


There is less total swing in Elo scores in a Gaussian implementation, coupled with a lower K factor. The points just dont trade hands as quickly. So if someone spends most of their time pugging and ends up 2000, it takes much more time for their rating to drop while they solo.

While they do solo, they will most likely be placed on a team with a 4 man, fighting another team with a 4 man. As a result of our narrower match maker. Which, results in more premade vs premade as they are more isolated from the solo drops. Even while soloing dropping our castaway 4 man dropper is more likely to see even matches.

This also reduces Elo drift. As players who usually quit do so with less Elo points than they started with.

#102 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 13 November 2013 - 07:01 PM

View PostGrits N Gravy, on 13 November 2013 - 04:39 PM, said:


You can just convert the scores run a global K factor of 30-40 for a period then go to much a lower K factor. Or use the old system while you build up new data then roll out the new system once you are properly seeded.


I'd be a big proponent of this. I'd understood that they were using a 5 point base scale, not 50, which I considered perfectly reasonable. Slow, but reasonable. 50 points scaled up and down for relative value is too much swing. I know that I will hit 20+ matches some days and have over 5500 matches total. I should be nested almost immovably in my Elo target but instead, if it's 50 points per match, I could be swinging 1,000 points in a DAY. I've had 16 match losing streaks before. The difference between an 80 point and an 800 point swing is night and day. Over 200 matches 16 matches can easily get evened out but if I'm swinging 500-1000 points in Elo in a day.... well, the concurrent experience for the player is pretty wacky.

View PostGrits N Gravy, on 13 November 2013 - 04:39 PM, said:

There is less total swing in Elo scores in a Gaussian implementation, coupled with a lower K factor. The points just dont trade hands as quickly. So if someone spends most of their time pugging and ends up 2000, it takes much more time for their rating to drop while they solo.

While they do solo, they will most likely be placed on a team with a 4 man, fighting another team with a 4 man. As a result of our narrower match maker. Which, results in more premade vs premade as they are more isolated from the solo drops. Even while soloing dropping our castaway 4 man dropper is more likely to see even matches.

This also reduces Elo drift. As players who usually quit do so with less Elo points than they started with.


The issue though is that a player who's earned a 2,000 Elo with his premade but pugs at an 1600 value is going to skew matchmaker results. If he's dropping with mostly premade teams what's going to happen is his team is going to be off by 400 points in value however you cut it. In fact this gets *worse* with a gaussian implementation - almost all the pugs on both sides will be people who earned their high Elo in premade fashion and are now pugging.

The problem here is that premades are treated as a single Elo amalgam made up of the players amalgam plus a modifier meant to represent the benefits of playing in a premade.

So suppose you've got 2 4mans on each team plus 4 pugs and the match estimated Elo score is 2000.

Team 1:

4man amalgam Elo - 1900
#1 actual score 1700
#2 actual score 1900
#3 actual score 1500
#4 actual score 1700
average 1700
premade bonus 200
amalgam Elo 1900

2nd 4man amalgam Elo - 2100
#5 actual score 1900
#6 actual score 1700
#7 actual score 2100
#8 actual score 1900
average 1900
premade bonus 200
amalgam Elo 2100

Pug # 9 Elo - 2200 - normally premades, without his team his value is 1800
Pug #10 Elo - 1800 - normally premades, without his team is value is 1600
Pug #11 Elo - 2000 - normally premades, without his team is value is 1900
Pug #12 Elo - 2000 - god tier pug who never premades *HIGHLY UNLIKELY*

Matchmaker Elo calculation:
Premade 1 1900 + Premade 2 2100 + Pugs 2200+1800+2000+2000 = 12,000/6 = 2,000.
Actual Elo value because of the pugs who are mis-calculated = 1883

Almost a 200 point swing.

Make sense? People perform better in teams than they do solo. Different skill set even, you know who you're with and how they'll act. You move in an organized group, you call targets on coms and you focus fire which multiplies your effectiveness. When pugging you have none of those benefits. In fact many of the builds that you use in a premade are only moderately effective when pugging. Ask any poptart or sniper or LRM boat. Without support your success plummets. Most rock solid 4mans will tell you that when they pug it's far, far more difficult.

The Gaussian model will skew higher Elo matches where you pretty much need to be dropping 4mans to get that level of success disproportionately because of this. Every pug in those matches is pretty much guaranteed to be rated higher than his skill represents. Even more bothersome is when these people get pulled to fill lower/higher Elo matches and a 'gimp' is pulled to offset the advantage they represent. Remember, not all premades are the same. There are plenty of premades that are just friends getting together without comms. Most fall in that description and they'll flesh out just about every band of Elo and require pugs to fill in the matches and inevitably to get weight matching and the like it'll have to reach into the +/-1 range to pull people. So it'll pull +1 pugs who normally premade who may be rated as much as 200 or more points above their actual skill while pugging, then it'll pull someone -1 who's legitimately 200 points below the Elo mean for the match and that team will end up 400 points out of variance because one guy is ranked incorrectly due to his difference in premade vs pug performance.

Are you getting what I mean? The difference for a lot of people between premade and pug performance can be hundreds of points of Elo performance. The higher the Elo the more likely this is. When the matchmaker does need to pull further +/- scale to fill a match it's going to be pugs being pulled to fill a match in accurate weight classing so when it pulls one or two widely miscalculated players it's going to effectively double the miscalculation when it pulls someone from down scale to offset them and balance the team total.

It's an inherent problem that magnifies as Elo rises and has a literally exponential statistical effect on match results. Everyone in that match gets their results skewed because the Elo prediction is actually off by 10 or even 20%. While I agree that if it were isolated cases a lower k-value would help balance these out but most premades pug at least some of the time. That means that above a certain Elo band close to 100% of the pugs are going to be incorrectly ranked. Even if only 33% of any matches population is pugs that's going to throw Elo estimates off by almost the same % because the tighter Elo scale grouping that's such a boon for the middle 80% of players essentially tosses the top 10% to the wolves.

If pug Elo is separate and the highest Elo scores are essentially only possible in premade teams it's still going to involve having to pull pugs to fill matches in higher Elo games but at least Elos ability to accurately calculate values for each team will be preserved, ensuring that the resulting Elo impact on everyone is as fair as possible. The top tier matches are always going to have balance issues but at least this way you're giving the matchmaker the most accurate possible data from which to predict win/loss and award or remove points accordingly.

#103 nehebkau

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,386 posts
  • LocationIn a water-rights dispute with a Beaver

Posted 14 November 2013 - 07:57 AM

View PostMischiefSC, on 13 November 2013 - 07:01 PM, said:

<<< Read his post its big and informative>>>


Ok, I believe I have an understanding of what is going on. This is what I think our simple solution is:

1. Add a premade bonus to people playing in pre-made groups to artificially inflate their ELO for matches / match-making. (To make pre-mades face tougher solo pugs)
2. Any ELO changes are calculated using that artificially elevated ELO (To reduce effects of pre-made victories skewing individual ELO scores)
3. soloing players use their regular ELO values. (To keep routine pre-made players from entering ELO hell if they solo)

What will happen, I believe, is that pre-mades will be matched with higher solo ELOs and comparible pre-made players. The ELO point swings will NOT be as great for players in pre-mades since a victory will not result in huge individual ELO increases. When a usual grouping player is soloing they should have a more reflective ELO value.

What I mean is that if I were pre-made and had a victory which, currently, would result in a 100 point ELO change, artificially increasing my ELO based on the group size would have changed that ELO increase to, say 10 with the victory. Why? Factoring in the group-size ELO bonus indicated that the odds of me winning the match were signifigantly higher.

PGI's first step, then, is to track the victories players have when pugging vs when they are in a pre-made group to determine how much grouping affects the win/loss ratio.

#104 Kunae

    Member

  • PipPipPipPipPipPipPipPipPip
  • 4,303 posts

Posted 14 November 2013 - 07:59 AM

The range right now is sooooo huge, anyways, it doesn't really matter. We're talking a 600-900pt spread in the pool.

#105 Burke IV

    Member

  • PipPipPipPipPipPipPipPip
  • Guardian
  • Guardian
  • 1,230 posts

Posted 14 November 2013 - 09:16 AM

There is no ELO they told you this so there would be something to post over. The match maker puts us together based on loadouts. This is why it will group all the ECM on one team and all the LRMs on the other. Its set up to make your life miserable if you drop with certain mechs. Im tellin ya

Edited by Burke IV, 14 November 2013 - 09:17 AM.


#106 Appogee

    Member

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

Posted 14 November 2013 - 09:26 AM

Tonnages dont matter.

What matters is when I and a few decent players are expected to carry a team of noobs to victory against a decent enemy team.

If I happen to be levelling up some kind of sub-optimal Mech (eg. Jenner K, any Locust) then we're inevitably screwed.as the noobs die quickly leaving a few of us to take on a dozen barely damaged enemies.

Edited by Appogee, 14 November 2013 - 10:16 AM.


#107 Joseph Mallan

    ForumWarrior

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

Posted 14 November 2013 - 09:29 AM

View PostAppogee, on 14 November 2013 - 09:26 AM, said:

Tonnages dont matter.

What matters is when I - and a few decent players - are expected to carry a team of most noobs and somehow win against a decent enemy team.

If I happen to be levelling up some kind of non-optimal Mech (eg. Jenner K, any Locus) then we're inevitably screwed.as the noobs die quickly leaving a few of us to take on a dozen barely damaged enemies.

Why do you feel yo are expected to carry the team? Do your part. Win or lose, at least YOU ha a good game. When I PUG I don't expect anyone to carry me. I do my pat or I fail to do my part. How arrogant is it to think you are expected to do anything but have fun killing the enemy?

#108 Kunae

    Member

  • PipPipPipPipPipPipPipPipPip
  • 4,303 posts

Posted 14 November 2013 - 09:31 AM

View PostJoseph Mallan, on 14 November 2013 - 09:29 AM, said:

Why do you feel yo are expected to carry the team? Do your part. Win or lose, at least YOU ha a good game. When I PUG I don't expect anyone to carry me. I do my pat or I fail to do my part. How arrogant is it to think you are expected to do anything but have fun killing the enemy?

To be fair, PGI expects it by having the MM balance out high Elo players with multiple low Elo players on the same team.

#109 Joseph Mallan

    ForumWarrior

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

Posted 14 November 2013 - 09:38 AM

View PostKunae, on 14 November 2013 - 09:31 AM, said:

To be fair, PGI expects it by having the MM balance out high Elo players with multiple low Elo players on the same team.

That is awful high thinking of oneself.Well I am here to let you know you are o't expected to do a dam thing in a match we are in together. Now the pressure has been lifted from you shoulders.

#110 Kunae

    Member

  • PipPipPipPipPipPipPipPipPip
  • 4,303 posts

Posted 14 November 2013 - 09:41 AM

View PostJoseph Mallan, on 14 November 2013 - 09:38 AM, said:

That is awful high thinking of oneself.Well I am here to let you know you are o't expected to do a dam thing in a match we are in together. Now the pressure has been lifted from you shoulders.

It has nothing to do with ego, Joseph. It's observation.

#111 Randalf Yorgen

    Member

  • PipPipPipPipPipPipPipPip
  • 1,026 posts
  • Locationwith in 3m of the exposed Arcons rear ct

Posted 14 November 2013 - 09:49 AM

View PostJoseph Mallan, on 14 November 2013 - 09:38 AM, said:

That is awful high thinking of oneself.Well I am here to let you know you are o't expected to do a dam thing in a match we are in together. Now the pressure has been lifted from you shoulders.

Aww, you guys having a spat?

Had a game last night, conquest Canyon. we left the bridge end and did the Kappa, Sigma, Epsi route, 2 jaggers and a Thunderbolt. none of us in our prefered mechs. by the time we took Epsi the rest of our team folded and had only taken out two of the enemy. It ended in with the enemy winning by cap but they had 3 mechs remaining and we had 2... all beat to snot. So, tell me again about Elo please and exactly how it doesn't put pressure on the more experienced players. Sure if we had followed the rest of the team maybe we would have won, but then again maybe we would have been boxed in, blocked from manouvering or worse yet, killed by team mates or run smack dab into the enemy blob and been slaughtered while the rest of our team ran for cover. those who stayed and watched it till the end, everyone was saying it was a great game and quite entertaining. both teams wanted just 2 more minutes to see who would have been left standing.

#112 Joseph Mallan

    ForumWarrior

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

Posted 14 November 2013 - 09:52 AM

View PostKunae, on 14 November 2013 - 09:41 AM, said:

It has nothing to do with ego, Joseph. It's observation.

Its an Assumption derived from an egoistical perspective. You are put in a Match, you are expected to pull your weight, nobody else's. If you do that and lose you still did your part.

Edited by Joseph Mallan, 14 November 2013 - 09:52 AM.


#113 Appogee

    Member

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

Posted 14 November 2013 - 10:18 AM

View PostJoseph Mallan, on 14 November 2013 - 09:29 AM, said:

Win or lose, at least YOU ha a good game.
I DON'T have a good game. I find myself trying to take on a half dozen enemy Mechs against impossible odds, and then I die having won few CBills and XP, which just makes the game more grindy.

Then, when I've lost a long string of games, suddenly I get put back on the ''average'' team, where I stomp enemy teams with impunity, which is a different version of ''not a good game''.

What would be a ''good game'' would be a two teams of relatively evenly matched players, fighting it out with equal odds of team and personal success, so that our skill determined the outcome.


View PostJoseph Mallan, on 14 November 2013 - 09:29 AM, said:

How arrogant is it to think you are expected to do anything but have fun killing the enemy?
Congratulations to you and your smug one man army. It's a special person who finds losing repeatedly against impossible odds 'fun'.

Edited by Appogee, 14 November 2013 - 10:24 AM.


#114 Randalf Yorgen

    Member

  • PipPipPipPipPipPipPipPip
  • 1,026 posts
  • Locationwith in 3m of the exposed Arcons rear ct

Posted 14 November 2013 - 10:21 AM

View PostAppogee, on 14 November 2013 - 10:18 AM, said:

I DON'T have a good game. I find myself trying to take on a half dozen enemy Mechs and I die.

Congratulations to you and your smug one man army.[/size]

ohh, sound slike someone has an itch they just can't reach to scratch.

Can I ask, how is it that you find yourself against 6 enemy mechs? I mean if you're the last guy standing that's one thing but if it's early in the match then.....well...... you know

#115 Joseph Mallan

    ForumWarrior

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

Posted 14 November 2013 - 10:28 AM

View PostAppogee, on 14 November 2013 - 10:18 AM, said:

I DON'T have a good game. I find myself trying to take on a half dozen enemy Mechs against impossible odds, and then I die having won few CBills and XP, which just makes the game more grindy.

Congratulations to you and your smug one man army. It's a special who finds losing repeatedly against impossible odds 'fun'.[/size]

Not Smug at all. If the Team loses an I get 4 kills and 600 damage, I crack a cold one an know I had a good game. If the team wins an I did 75 damage an 0 kills an 0 assists I sucked. I don't worry how the team does if I am PUGgin cause I don't have control over that. I do my part, and base my fun on my results. Most games I do good, Some I am king of battle. Others I should not have hit Launch. If the Odds are impossible and I score a kill or a few assists, I had fun. Winning isn't everything and losing isn't the end of the world. 33 years of competing have taught me that lesson.

Edited by Joseph Mallan, 14 November 2013 - 10:29 AM.


#116 Appogee

    Member

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

Posted 14 November 2013 - 10:31 AM

View PostRandalf Yorgen, on 14 November 2013 - 10:21 AM, said:

Can I ask, how is it that you find yourself against 6 enemy mechs? I mean if you're the last guy standing that's one thing
Bingo.

I'm not saying I'm always the last guy standing, or that I don't occasionally make a stupid tactical mistake that leads to my own premature death. However, the most common experience is that half the team walks into the enemy guns and dies, achieving little damage. Gone. That leaves me and - if I'm lucky - a couple of other guys to now take on twice as many enemies or more. Sure, we will use cover, our skills and experience, to try to win. But most often, the sheer number of enemy guns amassed against us will win out.

The evidence is on the scoreboard at the end of these matches. The inexperienced players are dead having scored less than 100 damage - many of them were in trial Mechs. Then there will be three of us who combined accounted for 75% of the damage scored. If we're lucky we managed to kill a few Mechs between us before being overrun.

Edited by Appogee, 14 November 2013 - 10:33 AM.


#117 Joseph Mallan

    ForumWarrior

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

Posted 14 November 2013 - 10:33 AM

View PostRandalf Yorgen, on 14 November 2013 - 10:21 AM, said:

ohh, sound slike someone has an itch they just can't reach to scratch.

Can I ask, how is it that you find yourself against 6 enemy mechs? I mean if you're the last guy standing that's one thing but if it's early in the match then.....well...... you know

You've never turned a corner and Opps! There it is! If not, count yourself lucky!

#118 Kunae

    Member

  • PipPipPipPipPipPipPipPipPip
  • 4,303 posts

Posted 14 November 2013 - 10:34 AM

View PostJoseph Mallan, on 14 November 2013 - 09:52 AM, said:

Its an Assumption derived from an egoistical perspective. You are put in a Match, you are expected to pull your weight, nobody else's. If you do that and lose you still did your part.

That's the problem. The way PGI's MM works, the weight some people are expected to pull is thus much higher than others.

#119 Joseph Mallan

    ForumWarrior

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

Posted 14 November 2013 - 10:36 AM

View PostKunae, on 14 November 2013 - 10:34 AM, said:

That's the problem. The way PGI's MM works, the weight some people are expected to pull is thus much higher than others.

That is your perception alone. PGI did not expect you to do anything. You put yourself under to much unnecessary pressure!

#120 Nik Van Rhijn

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,905 posts
  • LocationLost

Posted 14 November 2013 - 10:39 AM

What I can't understand is why new players start with an average Elo. Surely it would be better for all if they started at a lower level?





5 user(s) are reading this topic

0 members, 5 guests, 0 anonymous users