83 replies to this topic

[Lily from animove]

Heffay, on 26 January 2015 - 07:25 AM, said:

Which is why winning is how you determine whether or not you're good, since winning is the only thing that matters.

not entirely,

the question is what should elo represent. the skill of a player? or just his W/L ratio amongst the X last games? because we need for proprt matchmaking include skill as a valid variable. when 2 entirely different skilled guys group up for a session of 20 games, they will very likely start to group at the same elo by the mechanic they play, while one may constantly die pointless and non win contributing, and the other exceeds in his performance.

when they suddenly get givided by playing alone. One will have a hard time by having a wrong elo. Getting put together with poeple of possibly sky high elos compared to his real skill.

A good system would be a rejudgement of the scoresystem, that measures different acts better against each other. And then making score a base of elo + probably having the win increase the score by lets say 50 or any other number..


This way, people who support of focus on objectives may not get the short stick handled yb score judgement. and a win/los has impact on your elo but still is not the only deciding factor. yet the system can kinda differ between the " I cound't carry hard enough" losers and the "I got carried winners". because they are the edgy groups that in the current system totally get judged wrong.

[Joseph Mallan]

Davers, on 26 January 2015 - 07:24 AM, said:

Winning is more important that getting kills. 2/3 game modes can be won without killing a single opponent (theoretically at least). Elo doesn't look at just one match- it looks at your entire history. If you are getting 4 kills and 6 assists every match, you will win a lot of games and your Elo would go up. But most people have good matches and bad mtches- and a lot of those factors are still within a player's control. For example, I've been running my YLW a lot lately- I use an XL. The YLW is not a 'top tier' mech and while I have great matches, I also have poor matches where I get dual gaussed from behind and die early. My mech choice effects my Elo, as it should, because mech building is part of the game as well. That is why good players prefer to run good mechs and good builds. People can claim that Elo doesn't work, but I have more good games than bad (win or lose) so I have to say Elo is working for me.

But if I are not the one Standing on teh objective... I am still being carried for the win. The point is in the doing, not the attendance.

[Heffay]

Lily from animove, on 26 January 2015 - 07:33 AM, said:

not entirely,

the question is what should elo represent. the skill of a player? or just his W/L ratio amongst the X last games? because we need for proprt matchmaking include skill as a valid variable. when 2 entirely different skilled guys group up for a session of 20 games, they will very likely start to group at the same elo by the mechanic they play, while one may constantly die pointless and non win contributing, and the other exceeds in his performance.

when they suddenly get givided by playing alone. One will have a hard time by having a wrong elo. Getting put together with poeple of possibly sky high elos compared to his real skill.

A good system would be a rejudgement of the scoresystem, that measures different acts better against each other. And then making score a base of elo + probably having the win increase the score by lets say 50 or any other number..


This way, people who support of focus on objectives may not get the short stick handled yb score judgement. and a win/los has impact on your elo but still is not the only deciding factor. yet the system can kinda differ between the " I cound't carry hard enough" losers and the "I got carried winners". because they are the edgy groups that in the current system totally get judged wrong.

Elo does represent the skill of the player. It takes *all* factors into account, such as mech, ability, computer, experience, yadda yadda yadda. It normalizes everything based on your ability to use all those factors into accomplishing the main objective: Winning the match.

[Joseph Mallan]

Heffay, on 26 January 2015 - 07:33 AM, said:

Statistically, this is almost impossible. With a win/loss rate of 50% based on a well balanced Elo score, the odds of having 333 more wins than average by being carried is... yeah. No. 2^333? The math escapes me at the moment.

But not Impossible.

I did say I exaggerated... you did see that right.

Heffay, on 26 January 2015 - 07:35 AM, said:

Elo does represent the skill of the player. It takes *all* factors into account, such as mech, ability, computer, experience, yadda yadda yadda. It normalizes everything based on your ability to use all those factors into accomplishing the main objective: Winning the match.

It does not! It would be nice if it did...

[RedDragon]

Heffay, on 26 January 2015 - 07:35 AM, said:

Elo does represent the skill of the player. It takes *all* factors into account, such as mech, ability, computer, experience, yadda yadda yadda. It normalizes everything based on your ability to use all those factors into accomplishing the main objective: Winning the match.

Then riddle me this: How is it (theoretically) possible to be the worst pilot in the world and still have a high Elo because of being always carried by the winning team?

[Heffay]

Joseph Mallan, on 26 January 2015 - 07:36 AM, said:

It does not! It would be nice if it did...

*sigh*

It does Mr Mallan. Your individual stats mean nothing if you lose the battle. And if you consistently have phenomenal stats, odds are it will *influence* the battle and help you win, which will lead to your Elo rising. If you have a 2200 Elo score, odds are you're going to be a pretty hard hitter in the game, because that does help you win. But you could have the same stats as someone with an 1800 Elo score. The difference? The 2200 player pays attention to objects a *lot* more than the 1800 rated person.

[Lily from animove]

Heffay, on 26 January 2015 - 07:35 AM, said:

Elo does represent the skill of the player. It takes *all* factors into account, such as mech, ability, computer, experience, yadda yadda yadda. It normalizes everything based on your ability to use all those factors into accomplishing the main objective: Winning the match.

and thast the issue of the "bad matchmaking" elo is used to make matches not skills, and so since teams are not the same all the time, but get judged on team base, they simply get assorted horribly wrong sometimes.

simplified examply.

3 stong palyers and 3 weak ones get opposed to 3 strong and weak ones 6x in a row.

now they all end up with the same elo, and keep on the same level, because they both win 3 times. (maybe they played group queue)

Now the MM scramlbles the teams, putting the 6 baddies vs the 6 good ones, because suddenly event and pug queue. Outcome? horrible matchup.
and thats why elo should not only take W/L into account. becaus this is not leading to quality matchups (especially in the PUG queue where still the majority of players are playing.)

[Heffay]

RedDragon, on 26 January 2015 - 07:40 AM, said:

Then riddle me this: How is it (theoretically) possible to be the worst pilot in the world and still have a high Elo because of being always carried by the winning team?

Because luck is a real thing, and can be factored into your Elo score.

But with a 50% chance of winning a well balanced Elo match, what are the odds that you'll win significantly more matches than you lose if you get carried all the time?

[Davers]

Joseph Mallan, on 26 January 2015 - 07:34 AM, said:

But if I are not the one Standing on teh objective... I am still being carried for the win. The point is in the doing, not the attendance.

You can not be consistently 'carried for the win' enough to become a high Elo player nor to develop the 2.0 win ratio the OP is talking about. If you are not doing then you will lose far more than you will win.

[Lily from animove]

Davers, on 26 January 2015 - 07:46 AM, said:

You can not be consistently 'carried for the win' enough to become a high Elo player nor to develop the 2.0 win ratio the OP is talking about. If you are not doing then you will lose far more than you will win.

group queue may disagree.

[Davers]

Lily from animove, on 26 January 2015 - 07:49 AM, said:

group queue may disagree.

I would argue that you do not have a large enough pool size of data to determine that.

[Joseph Mallan]

Davers, on 26 January 2015 - 07:46 AM, said:

You can not be consistently 'carried for the win' enough to become a high Elo player nor to develop the 2.0 win ratio the OP is talking about. If you are not doing then you will lose far more than you will win.

Back in the day I got carried to a near 3.0 Win rate. Had a crappy Laptop, with horrid frame rates but good ping.

I switched to PUG Alts to test Elo. Recreated my Mechs that worked well, and when dropping in Group, got massacred while the team went on to win. One may not be able to be carried as far as I knowingly exaggerated, but you can be carried above your personal level of competence.

Edited by Joseph Mallan, 26 January 2015 - 07:52 AM.

[Heffay]

Joseph Mallan, on 26 January 2015 - 07:52 AM, said:

Back in the day I got carried to a near 3.0 Win rate. Had a crappy Laptop, with horrid frame rates but good ping.

I switched to PUG Alts to test Elo. Recreated my Mechs that worked well, and when dropping in Group, got massacred while the team went on to win. One may not be able to be carried as far as I knowingly exaggerated, but you can be carried above your personal level of competence.

So, what you're saying is that Elo self-corrected?

Yes, that is one of the benefits of it.

[Davers]

Joseph Mallan, on 26 January 2015 - 07:52 AM, said:

Back in the day I got carried to a near 3.0 Win rate. Had a crappy Laptop, with horrid frame rates but good ping.

I switched to PUG Alts to test Elo. Recreated my Mechs that worked well, and when dropping in Group, got massacred while the team went on to win. One may not be able to be carried as far as I knowingly exaggerated, but you can be carried above your personal level of competence.

Yes, rolling pugs with a team on VOIP can artificially inflate your Elo, but that is a case of outside circumstances that the MM could not account for. But Elo is self correcting, and it may only have taken 10-15 games to place you where you should be. Or are you still getting massacred every match?

[RedDragon]

Heffay, on 26 January 2015 - 07:42 AM, said:

Because luck is a real thing, and can be factored into your Elo score.

But with a 50% chance of winning a well balanced Elo match, what are the odds that you'll win significantly more matches than you lose if you get carried all the time?

Tell that to my W/L ratio of 0.65
I am by no means one of the best pilots here, but I have lots of experience (pilot for 15 years now), a (well) positive K/D ratio, tend to be the one capturing instead of pursuing kills and am generally a team player. I won't say that I carry my team (sometimes it happens, often it does not), but I am quite optimistic that I can carry my 1/12th share of the game. And yet I am losing game after game. Even if I am the worst pilot on earth, shouldn't I at last get into an Elo bracket where my W/L ratio is about 1?

[Heffay]

RedDragon, on 26 January 2015 - 08:01 AM, said:

Tell that to my W/L ratio of 0.65
I am by no means one of the best pilots here, but I have lots of experience (pilot for 15 years now), a (well) positive K/D ratio, tend to be the one capturing instead of pursuing kills and am generally a team player. I won't say that I carry my team (sometimes it happens, often it does not), but I am quite optimistic that I can carry my 1/12th share of the game. And yet I am losing game after game. Even if I am the worst pilot on earth, shouldn't I at last get into an Elo bracket where my W/L ratio is about 1?

It's tough to figure out what's going on in your case. Would need to see stats, maybe video of how you play, etc. Or maybe you don't have that many games under your belt? Or you're horribly unlucky (commercial pilot? oh man...)?

[Davers]

RedDragon, on 26 January 2015 - 08:01 AM, said:

Tell that to my W/L ratio of 0.65
I am by no means one of the best pilots here, but I have lots of experience (pilot for 15 years now), a (well) positive K/D ratio, tend to be the one capturing instead of pursuing kills and am generally a team player. I won't say that I carry my team (sometimes it happens, often it does not), but I am quite optimistic that I can carry my 1/12th share of the game. And yet I am losing game after game. Even if I am the worst pilot on earth, shouldn't I at last get into an Elo bracket where my W/L ratio is about 1?

Don't know enough about your play style or what game modes you are playing.

[Macksheen]

Single stats are rarely good.

I'd look at as much info as you can - from time to time I dump my stats page into a google sheet and make some composites and different stats to look at (damage / match, damage / kill) in addition to the provided win/loss and kill / death.

These are useful only in trying to understand how you may perform in one mech or another yourself; they are likely all really challenged in trying to figure out how you may perform as compared to others - so I don't try to even think of the stats in that way (too many unknowns to make comparisons).

I think you can objectively understand that "I am better in Mech A than Mech B", just not with any realistic clarity understand "I'm better than most other players in Mech A".

[RedDragon]

Heffay, on 26 January 2015 - 08:05 AM, said:

It's tough to figure out what's going on in your case. Would need to see stats, maybe video of how you play, etc. Or maybe you don't have that many games under your belt? Or you're horribly unlucky (commercial pilot? oh man...)?

Over 1500 matches since Closed Beta (including a place in the top 20 of one of the leaderboard challenges) should be enough to know how the game works
The point is: I am not that bad a pilot. But I can by no means influence a match that far as to win despite getting 5 new players on my team while the enemy doesn't. So I may be the one performing best on my team and we still lose. But that loss didn't hinge on me "not carrying hard enough" (which is a *** thing to say anyhow - you shouldn't be forced to carry more than your 1/12th share on a regular basis) but because there were too few valuable players on the team OR the enemy had more good players. So yeah, great, I did a good job, but still my Elo goes down. How is that fair?

[Karl Streiger]

RedDragon, on 26 January 2015 - 08:15 AM, said:

Over 1500 matches since Closed Beta (including a place in the top 20 of one of the leaderboard challenges) should be enough to know how the game works

problem is team work.
don't get me wrong - as a team player that usually drops in premade teams you (i) usually have extreme problems to be of worth in a drop that consists mainly out of single players.
They behave different - i don't say they are bad or they can't team play but their team play has another "quality" we (i) can't understand.

Anyhow - ELO is the only value that works. simple because its a derivate of all player statistics. it includes his ability to maneuver, shoot, react and act on enemy movement ....

Its a worse value - but there is no better.
The only thing you can modify is the "formel" for the MM - instead of average the ELO should get a specific weight - a 2000 point player + 600 point player != 2 players with 1300

Edited by Karl Streiger, 26 January 2015 - 08:23 AM.