19 replies to this topic

[SVK Puskin]

http://pokec.azet.sk...i9=325b10b31569

If the ELO trying to put players together with same or similar skills and the weight class is out than this system is pure fail of PGI. How could you PGI **** up this game like this? Bring back the weight class or weight limit ASAP.

[SVK Puskin]

If you want to see the images in full size than click on the image and than click on originál ( _ _ _ KB) - it is under the image, right side.

[WolvesX]

They got the worst elo system ever:

IT BALANCES TEAM BY ADDING **** PLAYERS INTO A HIGH ELO GROUP.

HOW STUPID IS THAT?

HOW STUPID MUST SOMEONE BE TO...

When you are high elo, you HAVE to play an assault to stay compeditive, because the MM will just IGNORE weight balance!

Edited by WolvesX, 15 June 2013 - 04:34 PM.

[Traigus]

Dropped with my team 300 tons light 2x today... out of 20 games.. ish. 100+ tonnes low maybe 1/2 the time.

it is my fault probably. I'm driving mediums.

[SVK Puskin]

WolvesX, on 15 June 2013 - 04:24 PM, said:

They got the worst elo system ever:

IT BALANCES TEAM BY ADDING **** PLAYERS INTO A HIGH ELO GROUP.

HOW STUPID IS THAT?

HOW STUPID MUST SOMEONE BE TO...

When you are high elo, you HAVE to play an assault to stay compeditive, because the MM will just IGNORE weight balance!

I do not know my ELO, do you know yours? The weight class balancing was took out to be readjusted but this is the main problem, many times i was outclassed and sometimes by a big difference. Outclassed team has usually no chance.

[WolvesX]

ENS Puskin, on 15 June 2013 - 05:22 PM, said:

I do not know my ELO, do you know yours? The weight class balancing was took out to be readjusted but this is the main problem, many times i was outclassed and sometimes by a big difference. Outclassed team has usually no chance.

I saw you playing and you got the tactic right and hit things = HIGH ELO

[Traigus]

ENS Puskin, on 15 June 2013 - 05:22 PM, said:

I do not know my ELO, do you know yours? The weight class balancing was took out to be readjusted but this is the main problem, many times i was outclassed and sometimes by a big difference. Outclassed team has usually no chance.

I'm pretty sure they put weight balancing back in, in the last couple of months (phase 3) just doesn't seem to count for much.

[WolvesX]

WolvesX, on 15 June 2013 - 05:50 PM, said:

I saw you playing and you got the tactic right and hit things = HIGH ELO

Most players can't hit a barn with a big bullseye on every side and a neon shield with a arrow stating "HIT ME".

So much for "convergence" is too stronk...

WolvesX, on 15 June 2013 - 05:54 PM, said:

Most players can't hit a barn with a big bullseye on every side and a neon shield with a arrow stating "HIT ME".

So much for "convergence" is too stronk...

QFT

Edited by WolvesX, 15 June 2013 - 05:57 PM.

[Deathlike]

Just so you know.. we're in phase four aka "Weight balancing w/ELO, favoring ELO over weights in the worst case scenario".

Edited by Deathlike, 15 June 2013 - 05:59 PM.

[Profiteer]

I dislike the ELO system.

I know PGI don't want to split the community up, but I feel the only way to balance the games out is to have 3 different divisions: cadets/soilders/vets, playing only against each other.

For me, playing with/against newb/casual players is a terrible experience. PUG games are a total lottery; occasionally I'll get a competent group, but more often than not I'll get players that suicide charge in single file. Last night I had a guy on my team ask how to respawn...

ELO is like the NFL trying to balance games out by putting semi-pro players on some sides.

Sport works by putting players of similar skill in divisions together, so you get somewhat even games. I don't see why online gaming should be any different.

[Deathlike]

I think the problem actually IS that ELO must be AVERAGED together.. so elite players are FORCED to pair up with complete newbies for the current MM system to function "correctly".

What should happen to some degree is overall matching should be lumping players of similar ELO together (+50 and -50, or -25 and +35 of your current ELO level, or some arbitrary range, as needed). So as you get better, you will play against more difficult players at a more steady progression.

The problem though with that though is that I'm unsure if we have enough ACTIVE players to do this with. Also, I think people to some degree are using "too much tonnage" as a crutch for their losses. If your ELO is actually good, you SHOULD be able to compensate with skill, but to some degree, there are true issues with balancing with respect to the mechs themselves... especially @ the medium weight classes that prevents this from being optimal.

I guess, there's no perfect system...

Edited by Deathlike, 15 June 2013 - 06:43 PM.

[p00k]

actually the way to fix it is to ditch matchmaking altogether and give us lobbies. something they should've done last fall before embarking on matchmaking phase 1 even. all the effort put into matchmaking phase 1, and phase 2, and phase 3, and all the elo tweaking and troubleshooting, all the programming time and qc time put into adjusting people's elo scores when they realized it was bugged, the effort to give people a cadet elo for their first 25, all of it was wasted.

instead, they should've just given us lobbies to start with. i mean, the only matchmaking system they needed has been in place for 12 years now (see MPBT3025). ******* idiots

[Deathlike]

Lobbies... MSN Gaming Zone... good times.

[SVK Puskin]

WolvesX, on 15 June 2013 - 05:50 PM, said:

I saw you playing and you got the tactic right and hit things = HIGH ELO

Maybe.

Traigus, on 15 June 2013 - 05:52 PM, said:

I'm pretty sure they put weight balancing back in, in the last couple of months (phase 3) just doesn't seem to count for much.

I do not think so, i still can see a big difference in weight.

WolvesX, on 15 June 2013 - 05:56 PM, said:

Most players can't hit a barn with a big bullseye on every side and a neon shield with a arrow stating "HIT ME".

So much for "convergence" is too stronk...


QFT

QFT?

Deathlike, on 15 June 2013 - 05:58 PM, said:

Just so you know.. we're in phase four aka "Weight balancing w/ELO, favoring ELO over weights in the worst case scenario".

Phase 4? Never heard about it.

[Levi Porphyrogenitus]

Elo tiers would fix a lot of problems. There are several ways to implement them, and here are a few:

1 - Team v Team. Each team is composed solely of players from a single Elo tier. If there aren't sufficient players available then it expands up/down by one tier (but the same direction for both teams). This continues until both teams are full.

2 - Lance v Lance. Each team is composed of two lances (later three) of which every member is from the same Elo tier. Each lance is matched against another lance of the same tier. Player shortfalls are handled like in (1). This gives greater variety within each team while retaining consistency between teams.

3 - Pilot v Pilot. Both teams gain one pilot at a time (at the same time as the other team gains theirs), and the pilots for both teams are drawn from the same tier. Tier matching is done only with consideration to each individual pilot, ensuring that a high-tier pilot will face a high-tier pilot and a low-tier pilot will face a low-tier pilot. This allows maximum flexibility while still maintaining balanced Elo distribution across the teams, and would allow for players to encounter people outside their own Elo tier far more often than the other two systems would (even the Lance system).

I tend to favor (3), but (2) would work ok.

As for the tiers, I imagine each tier would consist of a 100-Elo range. If the median is 1200, then the base Elo tier is 1150-1249. Tier -1 is 1050-1149. Tier +3 is 1450-1549. So on and so forth.

PGI could even reveal which tier you belong in for each category (L, M, H, A, Lance / Partial Team, and Full Team), without revealing the actual Elo values.

[SVK Puskin]

Profiteer, on 15 June 2013 - 06:33 PM, said:

I dislike the ELO system.

I know PGI don't want to split the community up, but I feel the only way to balance the games out is to have 3 different divisions: cadets/soilders/vets, playing only against each other.

For me, playing with/against newb/casual players is a terrible experience. PUG games are a total lottery; occasionally I'll get a competent group, but more often than not I'll get players that suicide charge in single file. Last night I had a guy on my team ask how to respawn...

ELO is like the NFL trying to balance games out by putting semi-pro players on some sides.

Sport works by putting players of similar skill in divisions together, so you get somewhat even games. I don't see why online gaming should be any different.

I was thinking about some objectives or challanges for new players. They could upgrade the testing grounds to player vs player so new players will be able only to do this and after reaching maybe 100 rounds or certain amount of demage, accuracy...they will be able to join real match.

[SVK Puskin]

p00k, on 15 June 2013 - 06:47 PM, said:

actually the way to fix it is to ditch matchmaking altogether and give us lobbies. something they should've done last fall before embarking on matchmaking phase 1 even. all the effort put into matchmaking phase 1, and phase 2, and phase 3, and all the elo tweaking and troubleshooting, all the programming time and qc time put into adjusting people's elo scores when they realized it was bugged, the effort to give people a cadet elo for their first 25, all of it was wasted.

instead, they should've just given us lobbies to start with. i mean, the only matchmaking system they needed has been in place for 12 years now (see MPBT3025). ******* idiots

Lobbies? What do you mean? I have no idea what is that.

[SVK Puskin]

Levi Porphyrogenitus, on 15 June 2013 - 07:31 PM, said:

Elo tiers would fix a lot of problems. There are several ways to implement them, and here are a few:

1 - Team v Team. Each team is composed solely of players from a single Elo tier. If there aren't sufficient players available then it expands up/down by one tier (but the same direction for both teams). This continues until both teams are full.

2 - Lance v Lance. Each team is composed of two lances (later three) of which every member is from the same Elo tier. Each lance is matched against another lance of the same tier. Player shortfalls are handled like in (1). This gives greater variety within each team while retaining consistency between teams.

3 - Pilot v Pilot. Both teams gain one pilot at a time (at the same time as the other team gains theirs), and the pilots for both teams are drawn from the same tier. Tier matching is done only with consideration to each individual pilot, ensuring that a high-tier pilot will face a high-tier pilot and a low-tier pilot will face a low-tier pilot. This allows maximum flexibility while still maintaining balanced Elo distribution across the teams, and would allow for players to encounter people outside their own Elo tier far more often than the other two systems would (even the Lance system).

I tend to favor (3), but (2) would work ok.

As for the tiers, I imagine each tier would consist of a 100-Elo range. If the median is 1200, then the base Elo tier is 1150-1249. Tier -1 is 1050-1149. Tier +3 is 1450-1549. So on and so forth.

PGI could even reveal which tier you belong in for each category (L, M, H, A, Lance / Partial Team, and Full Team), without revealing the actual Elo values.

Interesting idea. I have read in Ask The Developers 40 that they plan to add weight limit for a drop but i hope they will bring it fast becuase i am suffering from this unfair matchmaking system.

[Deathlike]

ENS Puskin, on 15 June 2013 - 07:02 PM, said:

Phase 4? Never heard about it.

http://mwomercs.com/...ted-april-19th/

[SVK Puskin]

Deathlike, on 15 June 2013 - 07:47 PM, said:

http://mwomercs.com/...ted-april-19th/

Thanks i missed this information but still there is so much weight class difference so even the Phase 4 sucks.