148 replies to this topic

[Commander Kobold]

Phoenix Gray, on 17 March 2013 - 07:54 AM, said:

I see three FOUR ECM mechs on the winners. No ECM on the losers. Any TAGs or PPCs? If not, then someone(s) wasn't prepared for the environment. That, or if the game has an algorhythm for balancing ECM, then it sure wasn't working.

I also note wildly disparate ping rates.

PPCs are not a really effective counter vs ecm

[MischiefSC]

SgtKinCaiD, on 17 March 2013 - 07:47 AM, said:

What a great and balanced match ! Did I mention the tonnage between teams ?

Clearly this is photoshopped. Both sides had A1s (splatcats) who didn't do 1,000 damage each win or lose. Everyone knows that A1s are OPed and do 1,000 damage per game no matter what or at least always place top of the list for damage. Even if they disco at the start of the match.

Joking aside the weight mismatch is painful in that one. It happens though, I've been on both sides of it. It's getting fixed.

[Killerwithin]

SgtKinCaiD, on 17 March 2013 - 07:47 AM, said:

What a great and balanced match ! Did I mention the tonnage between teams ?

Seems about right. I once dropped and we had one heavy the rest was medium and lights and they had 4 assaults 2 heavies 2 lights this was also late at night.

[Noth]

T Hawk, on 17 March 2013 - 05:51 AM, said:

Example?

League of Legends. Uses elo, team based, an get unlucky or lucky streaks with team mates, yet you still the typical all star names at the top of the boards.

[LordofDecay]

ELO doesn't bother me. What bothers me the most is the mis-matched weight ratios of the drops. I know it's scheduled to be fixed but it's very frustrating.

[Glythe]

Murzao, on 17 March 2013 - 06:44 AM, said:

Games are only extremely lopsided when one team doesn't focus fire (or mainly uses LRMs/SSRMs/other non DD weapons or just can't aim)........or your team gets run over 5-1 repeatedly super fast due to poor positioning.

Bhan Zor, on 17 March 2013 - 07:33 AM, said:

Balanced match outcomes have become rare and you cant explain that away with tactical mistakes all the time. There is a reason people make these mistakes; lack of experience being the most important one. And when they are repeatedly matched up against people who apparently know more than just the basics, it ends in a "roflstomp" - isnt the purpose of Elo to avoid this crap? If it doesnt do that, its worthless.

Pretty spot on reply.... was going to say something myself but it was already covered.

Let us consider another game and look at what happens there. In WoT the matchup is random everytime. There's a mod you can install to see how skilled everyone is in the game. Red is bad (under 48%), Green is decent (around 53%), and purple is fantastic (over 55% or something like that).

There are enough purple players and green players so that Purple could only play with purple and green no matter what tier you play. There is no reason a red player should ever see a purple player. And yet earlier during prime time in a platoon of 2 we had 2 purple players and the rest were all red. With little exception this kept happening over and over. It's quite literally the equivalent if you as an adult were put into the boxing ring with a 12 year old child. Sure he might move fast in a straight line but where's he going in the ring? His punches won't hurt you and you can barely tap him and really hurt him.

Overall this mod hurt tanks because it let you see instantly who you needed to kill to win. Being purple I was always targeted first and before this mod that rarely ever happened (because I positioned myself well). The take home message is that Elo should never be shown to players before the match starts. After the game is a completely different story however.

And now back to MWO..... there's no reason the terribad should ever play with really good players. Good players resent the noobs who have nothing to contribute to the team. Bad players hate getting 1 shotted from half way across the map (or from a high dps build up close). Putting them in a game together makes for a bad experience for both parties almost without exception.


Overall I saw many close games with random MM that functioned on weight. But ECM is the next big thing that must be balanced for both teams. When you combine these two you see 4 ECM ravens kill half a team in moments (often with the team unaware the enemy even has 4 scouts). The enemy team might not even have a light and there's nothing fair about such a match at all (especially when you can have 6 ecm mechs vs none).


Elo is the last variable to balance as it does influence the game but wow it's so broken when it's the only factor.

[FunkyFritter]

When matchmaking is working most players will win close to 50% of their games. If you're somehow way below that something is up, but usually when people complain about bad teammates they're just not used to even fights.

[Glythe]

LordofDecay, on 22 March 2013 - 08:15 PM, said:

ELO doesn't bother me. What bothers me the most is the mis-matched weight ratios of the drops. I know it's scheduled to be fixed but it's very frustrating.

How are they going to fix Elo? There is no way you can account for a player in a different weight mech to be matched against a better or worse opponent in another different weight mech. They are going to try and use statistics but it won't work. Chess works for Elo because we are completely identical before the game starts. Meanhhile mechs are built on the concept of paper rock scissors (mobiliy armor firepower) there's so much more going in between the confrontation of two enemies. The only way you could even remotely get close with what they are trying to do is to keep a ton of data on each player and that is extremely counter productive. Chess statistics also never had to deal with 8 players on a team.

FunkyFritter, on 23 March 2013 - 01:33 AM, said:

When matchmaking is working most players will win close to 50% of their games. If you're somehow way below that something is up, but usually when people complain about bad teammates they're just not used to even fights.

First I am used to even fights because we practice 8 man groups against allies. It's easy to see when allies are smart and when they are stupid. Smart allies (even without comms) will see friendlies that are about to lose a fight before it happens. They move in to prevent a mismatch and save their allies. Stupid allies will sit back and get a few hits in but not really contribute while the enemy team fights an ally 3 to 1 (or worse odds). Stupid allies also do things like send 5 mechs to cap a point that will be captured by a light long before they ever get there. I don't know how they do not know better because we do not have very many maps and they should have figured out the patterns by now.

Right now the mismatch maker will sabotage your team with noobitry if you drop with a team of 4 with a high Elo score if it can't find similarly ranked enemies. Brain dead allies may as well be DC because you can't use them for anything. How does that make for better games?

Elo in a fps game is stupid because it is an artificial means to try and sabotage win rates 50% or higher. Over the long term random MM got it right far more often. This isn't chess where everyone has the same pieces and the same opportunity on exactly the same level playing field.

[Zolaz]

I think the programming for PGI's ELO system is right on par with its other programming ability. Someone over there needs to take a look at the code and the results. PGI you dont even have to tell us how bad you screwed up ... JUST FIX IT!