16 replies to this topic

[Amberite]

Not aware that it is visible anywhere atm, why not let us see our effective "score". There will be some e-p waving, but it will also allow us to check matchmaking from our end.

[Tennex]

no can't happen

[FREDtheDEAD]

Can I suggest that instead of an arbitrary number for ELO, there's just four settings - new, veteran, champion and legendary. That should make the whole matchmaking system a lot easier to balance.

[Amberite]

Tennex, on 07 June 2013 - 03:56 AM, said:

no can't happen

Perhaps if you elaborate upon that somewhat, it'll be easier to see your point of view?

[Damocles]

Amberite, on 07 June 2013 - 05:05 AM, said:

Perhaps if you elaborate upon that somewhat, it'll be easier to see your point of view?

Its the PGI response

[Albert Meyburgh]

all that elo does, is trend a player towards a 50/50 win/loss ratio. It doesn't equal skill, it's just a correlation at best. so the way you can check matchmaking, is to track you win/loss ratio on your end over time.

[Tennex]

elo is just a system used to place players in equal matches.\

now if they ever decide to do ladders and leagues (similar to league of legends, starcraft 2), itd be nice to know if what league we were in. but never the ELO.

Edited by Tennex, 07 June 2013 - 09:55 AM.

[Gritash]

if you are any good with spreadsheets you can have fun with the stats pages

[Amberite]

The reason I asked this was due to the number of times I'm getting dropped with groups composed of relative newcomers, vs what is blatantly organised (full voicecomms) teams. If we can identify erroneous matching, perhaps we could eliminate the problem.

[MadTulip]

srccoder, on 07 June 2013 - 09:31 AM, said:

all that elo does, is trend a player towards a 50/50 win/loss ratio. It doesn't equal skill, it's just a correlation at best. so the way you can check matchmaking, is to track you win/loss ratio on your end over time.

thats nonesense. false information. yes,it trends you to 50% win but 50% chance to win against players of the same elo level. against players of lower elo you have a higher chance of winning. against players of higher elo you have a lower chance of winning. what else would be a better measure of "skill" but a statistic about the result ? something with high correlation describes the object under observation well with the tested object so thats exactly the info the op intended to have, no?

Read here

ofc it doesnt "equal" skill as apples dont equal food. its probably the best correlation you can get, yes. tracking your own win/loss ratio is not needed. if the elo system works its 50% with large enough games and players obviously.

id like to see elo also. if its not important then there is no reason not to show it right ?

you can maybe argue that youre game is not balanced and therefore elo only correlates to mech and equipment used AND skill. but well that the other story

Edited by MadTulip, 07 June 2013 - 02:24 PM.

[Chemie]

srccoder, on 07 June 2013 - 09:31 AM, said:

all that elo does, is trend a player towards a 50/50 win/loss ratio. It doesn't equal skill, it's just a correlation at best. so the way you can check matchmaking, is to track you win/loss ratio on your end over time.

First, liking your own post is kind of bad?

Anyway, since my W/L is around 3:1 since ELO was implemented, what exactly does this say?

[Chavette]

Xajorkith, on 07 June 2013 - 04:53 AM, said:

Can I suggest that instead of an arbitrary number for ELO, there's just four settings - new, veteran, champion and legendary. That should make the whole matchmaking system a lot easier to balance.

This is the best middle road. No elo farming or showing off, but the basics are visible.

[Amberite]

A note on ELO, W/L and K/D.

I spent a fair amount of time playtesting this game, using really cruddy builds sometimes to test theories and bugs. I've ended up with a fairly crappy K/D and W/L due to this. ELO doesnt appear to have helped redress this by matching me with less able players. In fact, it appear the majority of matches seem to land me with Uber-team-o'doom on the opposite side.

So...until I can see some evidence of ELO affecting my matchmaking I'm going to assume it isn't operating properly. There isn't really a solid rationale I can see for not showing something of the mechanisms governing how we get matched...unless of course ELO really is the heap of manure it appears to be.

and, as I've said before, without the tools to test this crud properly, we aren't really beta testing......we're just playing a part-done game.

[Asmosis]

There will be actual ranking type stuff later on with CW and so on, plus the longer term events that get run.

If you care about rankings, compete in those.

As for knowing where you are in Elo, try to figure out who the "top" players are and when you start seeing them on a regular basis, you know you made it ! (if you see me on a regular basis, you gots a while to go yet)

Edited by Asmosis, 11 June 2013 - 04:33 PM.

[Seddrik]

ELO = an ARTIFICIAL system whereby you are FORCED to lose after winning.

It is to baby new players and punish successful players.

If you are stupid enough to win several rounds, then play solo... you get uber messed over. Forced into horrible groups back to back... because the system expects you to carry new players and trial mech people and unorganized people who wander off solo and die. 99 out of 100 times theres no teamwork out side of a group... so why limit group making to 4 mans so you have to play with randoms?

This is the first game I have ever played where grouping was force altered by your win/losses AND limited to only half the available players on a side. All other games I have played lets people play randomly, or in groups up to the full team - period. And it made sense to play in groups because teamwork = wins. This system does neither since 8 mans are so limited no one wants to play them and when you do play solo you don't even get true random (or even just mech weight balanced) matches and again 4 mans limits the group to half randoms (at best) who could care less about teamwork.

Please take the chains off this game and let people play without elo and without 8v8 only restrictions.

A tough game is fine, interesting, challenging. And you can over come challenges with skill & teamwork.

A biased system is not fine, but pushes people up who fail and pushes people down who succeed. Stop rewarding failure and punishing success please. You want to help new players? Fine. I'm not against that in a starter package or whatever limited time help. But don't punish the rest of us to do that.

Edited by Seddrik, 12 June 2013 - 06:09 AM.

[Amberite]

Following on to Seddrik:

I have to agree. I can't recall playing a game before that purposefuly pushes you towards losing in response to you winning.

Matching you with -equally- skilled players would be far better, and that's what most games endeavour to do.

[Seddrik]

Personal experience.

If you play with a team and win often. Don't play solo.

Elo will make you rage with the teams it gives you.