26 replies to this topic

[Rekoro]

Hi there,

I´ve one simpel question for all my fellow warriors. Do you guys think that the elo system is working at all ?
So here is why i created this topic.

Almost every match in either my or the enemy team are Players that deal less than 100-200 dmg. And i´m not talking about lightmechs caught out of position, im talking about heavy and assault mechs in all kind of loadouts. I know things can go sideways and you end up with stats like that, but this is far beyond that. Every match i played so far, there were atleast 1-3 ppl with that kind of stats. This in mind i really doubt that the elo system works at all.
You have ppl dealing 400+ dmg and some around the 200 mark wich is totaly fine. and than you got these "noobs" that straight out die with <100 dmg done. IF the elo system would work these ppl wouldnt be in that match. And its not only about the damage, the loadouts i see are mindblowing as well. So what are your thoughts about this ?

[neviu]

ELO?
what is ELO?

[KinLuu]

Even the most skilled player goes down without much damage dealt if caught out of position by a good tagteam of jumpsnipers.

[MadPanda]

Had a bunch of 8-0/1/2 today. Do I think its working? Well since I was on the winning side I had some fun, but it doesn't seem to be working at all.

[Roadbeer]

Damage is a meaningless indicator of skill.
A skilled Boomcat player with 120 damage got 3 head shots and 3 kills was far more productive than the 900 damage Atlas Megaplatform player who literally stripped every piece of armor off of 3 heavies without killing them, only to end up being killed by said Boomcat.

As far as Elo, y'all need to stop obsessing over it. If you're +/- 10% from a 50/50 win/loss ratio, it's working just fine and matching you against the players it should be.

And sometimes, you're just in the wrong place at the wrong time. Someone's gotta die first.

Edited by Roadbeer, 20 April 2013 - 01:24 PM.

[Rekoro]

KinLuu, on 20 April 2013 - 01:18 PM, said:

Even the most skilled player goes down without much damage dealt if caught out of position by a good tagteam of jumpsnipers.

As i said, this is not happening once at a time, i encounter this almost every match and its crossteam which result in up to 4 ppl with that kind of stats. i highly doubt that this is a coincidence

@Road
The ppl im talking about, DIE with no kills and close to 0 dmg done. so its not like your example

Edited by Rekoro, 20 April 2013 - 01:29 PM.

[Vassago Rain]

Roadbeer, on 20 April 2013 - 01:23 PM, said:

Damage is a meaningless indicator of skill.
A skilled Boomcat player with 120 damage got 3 head shots and 3 kills was far more productive than the 900 damage Atlas Megaplatform player who literally stripped every piece of armor off of 3 heavies without killing them, only to end up being killed by said Boomcat.

As far as Elo, y'all need to stop obsessing over it. If you're +/- 10% from a 50/50 win/loss ratio, it's working just fine and matching you against the players it should be.

And sometimes, you're just in the wrong place at the wrong time. Someone's gotta die first.

Damage is a very good indicator of skill. It tells people you're doing something.

[Roadbeer]

Vassago Rain, on 20 April 2013 - 01:26 PM, said:

Damage is a very good indicator of skill. It tells people you're doing something.

Only if by "doing something' you mean you're exhibiting so much skill you're drawing your name on the enemy mech with your lasers.

Pinpoint firing kills, just pulling the trigger and doing damage does not equate to skill.

[Vassago Rain]

Roadbeer, on 20 April 2013 - 01:58 PM, said:

Only if by "doing something' you mean you're exhibiting so much skill you're drawing your name on the enemy mech with your lasers.

Pinpoint firing kills, just pulling the trigger and doing damage does not equate to skill.

If one guy draws emotes on the enemy atlas for 500 damage, he's a much better player than the player with 80 damage. Furthermore, he's doing something.

[Roadbeer]

Vassago Rain, on 20 April 2013 - 02:00 PM, said:

If one guy draws emotes on the enemy atlas for 500 damage, he's a much better player than the player with 80 damage. Furthermore, he's doing something.

This I don't argue with... unless the guy with 80 damage did it in a Boomcat and got to head-shots.

[Rekoro]

And that is what i was talking about. Players who underperform badly. Meaning they dont belong into the match. I know its not only about dmg but its an indicator, and when ~4 ppl die every match without doing anything besides ~90dmg i ld not call that an coincidence

[verybad]

Vassago Rain, on 20 April 2013 - 02:00 PM, said:

If one guy draws emotes on the enemy atlas for 500 damage, he's a much better player than the player with 80 damage. Furthermore, he's doing something.

Not if those 80 damage points are all on the head or rear central torso of the targets.

I would say that damage is ONE way of looking at skill, so are kills, asists, and some of the other things that are tallied.

The best way of checking skill however is not through any of those, it's watching a player play after you're dead and saying..."damn you're good!"

Skill is a subjective system, i've seen players in Spiders do stuff that is amazing, just DPS in an armor heavy build will get you damage

An Atlas is fat, just it's toughness and easy to put on enough weapons that cause enough damage to not be last in the list doesn't make the player skillful.

Relying on the points accumulated in a game to tell you who is and is not skillful if like watching sports through Excel. It just don't work that way.

Edited by verybad, 20 April 2013 - 02:13 PM.

[Roadbeer]

Rekoro, on 20 April 2013 - 02:09 PM, said:

And that is what i was talking about. Players who underperform badly. Meaning they dont belong into the match. I know its not only about dmg but its an indicator, and when ~4 ppl die every match without doing anything besides ~90dmg i ld not call that an coincidence

Well, since you've barely been in the game 3 weeks, I'd say you probably haven't reached the threshold of having an accurate Elo ranking established yet.

verybad, on 20 April 2013 - 02:11 PM, said:

Not if those 80 damage points are all on the head or rear central torso of the targets.

I would say that damage is ONE way of looking at skill, so are kills, asists, and some of the other things that are tallied.

The best way of checking skill however is not through any of those, it's watching a player play after you're dead and saying..."damn you're good!"

Skill is a subjective system, i've seen players in Spiders do stuff that is amazing, just DPS in an armor heavy build will get you damage

An Atlas is fat, just it's toughness and easy to put on enough weapons that cause enough damage to not be last in the list doesn't make the player skillful.

Relying on the points accumulated in a game to tell you who is and is not skillful if like watching sports through Excel. It just don't work that way.

QFT

Edited by Roadbeer, 20 April 2013 - 02:23 PM.

[Willie Sauerland]

I read this entire thread.
This was approximately 5 minutes of my life that I won't get back.
To be honest, I would completely care if this were an actual issue,
except I am just too busy not giving a ****.
As your KDR approaches 1, you will find out ELO is working.
And some KDR's are so high, it will take a while but will get there.

[Rekoro]

Than pls tell me how i can spectate DEAD ppl to judge what they are capable of .

Ok since some of you dont get what im saying. I AM NOT talking about players who died for the team in someway so they end up with bad overall stats (kills,assists,dmg) or bad loadouts/ aiming skills. Im talking about the players who just die. plain and simple. Also im not talking about light or medium mechs because most of the time its an heavy or assault mech which performs badly. And no this was allready happening before everyone jumped on the ppc´s.

@Road I´ve played 500+ matches so far, if this isnt enough for the system than it is broken, its as simple as that.

@willie

You are srsly talking kdr ? How is KDR related to elo ? its not like you can steal kills

Edited by Rekoro, 20 April 2013 - 03:04 PM.

[Roadbeer]

Rekoro, on 20 April 2013 - 02:57 PM, said:

Than pls tell me how i can spectate DEAD ppl to judge what they are capable of .

Ok since some of you dont get what im saying. I AM NOT talking about players who died for the team in someway so they end up with bad overall stats (kills,assists,dmg) or bad loadouts/ aiming skills. Im talking about the players who just die. plain and simple. Also im not talking about light or medium mechs because most of the time its an heavy or assault mech which performs badly.

@Road I´ve played 500+ matches so far, if this isnt enough for the system that it is broken, its as simple as that.

500 matches on an account created March 30th?

[krash27]

ELO seems to be working for the most part, its the Match Maker that has sh*t the bed.

[Puppeteerxerxes]

This link on matchmaking phase 4 may answer your question OP. I was wondering why after getting to where I thought I was a bit better than average that I saw a LOT of trial mechs with new players. If you look at the distribution if you are a player with a lot of matches in the average range of ELO you will see a lot of new players (the lighter colored curve overlaid on the dark blue bell curve.) Not sure how or why the new players are in that hopper rather than just starting off low, I've not taken time to really read up on how it ELO as a system is devised or how exactly the dev's are implementing it.

http://mwomercs.com/...view=getnewpost

As for the damage as indicator of skill, after seeing the arguing on forums about the placement of damage etc. can't we all agree that it is a very useful way to see how someone does, but can be misleading. I just figure we probably have more interesting topics to argue...

[Rekoro]

Puppeteerxerxes, on 20 April 2013 - 03:10 PM, said:

As for the damage as indicator of skill, after seeing the arguing on forums about the placement of damage etc. can't we all agree that it is a very useful way to see how someone does, but can be misleading.

That is exactly why i said im not talking about the ppl who actually did something good for the team

[Vassago Rain]

verybad, on 20 April 2013 - 02:11 PM, said:

Not if those 80 damage points are all on the head or rear central torso of the targets.

I would say that damage is ONE way of looking at skill, so are kills, asists, and some of the other things that are tallied.

The best way of checking skill however is not through any of those, it's watching a player play after you're dead and saying..."damn you're good!"

Skill is a subjective system, i've seen players in Spiders do stuff that is amazing, just DPS in an armor heavy build will get you damage

An Atlas is fat, just it's toughness and easy to put on enough weapons that cause enough damage to not be last in the list doesn't make the player skillful.

Relying on the points accumulated in a game to tell you who is and is not skillful if like watching sports through Excel. It just don't work that way.

In the actual world, this doesn't happen, just like no one gets only tetrises.