44 replies to this topic

[Deathlike]

MrMadguy, on 18 July 2015 - 10:50 PM, said:

And I don't really care about my W/L ratio and how fair it to have it almost equal to 1 in a long term. I care about following situation: imagine, I have only 1-1.5 hours every day to play MWO, so I don't have a time to wait till MM will "warm up" and ELO will settle down - I just don't want to waste this time on streak of roflstomps! I want MM to work the following way - it should care about my personal performance both in a long and in a short term. There should not be a situation, when for example while event, population of players suddenly dramatically changes, which causes ELO levels to shift, but it takes ages for my ELO level to be finally corrected, cuz average ELO level of both teams don't change - my personal level changes relatively to average.

MM should care about relative ELO level - not just average. Or more simply: it should also care about K/D or match score - not only about W/L.

It's not actually possible for many of the things cited here.

While it should care about relative Elo level, the problem is "the current population at the time of when you play" changes the makeup of its ability to make "fairer matches" as you move from the middle to an extreme (sounds like you're going into the higher level), the disparity increases as there are far less people in the upper tier regions than there are in the middle tier.

With regards to factoring K-D ratio and score, those aren't the most useful things. Getting a UAV to collect a better score doesn't mean you were actually responsible for the win (or any sort of influence to it). K-D ratios can be inflated as having a good game (say Aces of Spades - 8 kills in a match) doesn't actually mean you're a good player. Some people like having high K-D ratios, but don't focus on the win, thus affecting where they stand Elo-wise.

Ultimately, if you get the W, however which way, you are going to gain more Elo. The question is by how much.

The harder your "opponent's Elo average" happens to be, the more the gain. Beating the Lords (if they are still around) would generally mean an increase of Elo, but beating people "of your Elo average" changes very little on the Elo side. Remember, it's AVERAGES, so a very good 6-man team could be offset by a very poor 6-man team (both of which are on the same side) when the MM does its team construction.

In any case, you can't really have it the way you described it, given the current conditions AND to what Elo is mostly responsible for.. the Win. Your Elo simply changes more depending on whether "they were considered better or worse" than you.

[EekaBlitzer]

Sadly MM is the major festering albatross shaped **** round the neck of MWO.
It does not work , period. I am getting more than fed up with being throw into groups that are only ever going to lose, because it has somehow been determined that I need to kill 4+ enemy mechs in order to make the game a close fight.

[Sky Legacy]

MM works and I can prove it to you. Go into MWO, click on 'statistics' on the left hand side and let it open the website. Then when it opens your statistics, look at your total win/loss record of your entire MWO career. I guarantee you it will be somewhere around 50/50 or you will have possibly even more wins than losses. That's proof that whoever is claiming they constantly lose or get stomped are just using selective memory and confirmation bias because you in actuality win just as much as you lose which is the point of MM.

[km1710]

EgoSlayer, on 18 July 2015 - 08:49 AM, said:

3) Stomps happen because of combat loss grouping. And happen more frequently in *evenly matched teams*. because once they are down units they don't have the skill reserve to overcome the remaining enemy numbers.

is that the only reason or are there others for this claim? just asking out of curiosity - because I would think that if two teams are "evenly matched", in other terms assuming similar coordination and fire direction, the total damage traded and most importantly the focus of this damage should tend to be the same for both teams, leading to an evenly distributed mech loss.

Elo does work in groups due to the large numbers' law and also beacuse in teams it may be factored... but this game doesn't have a big enough playerbase, for this reason in my opinion they should have chosen a MM approach like world of warships. From a thread about that game opened some months ago, one could infer that MWO playerbase size is comparable to that

[MrMadguy]

Sky Legacy, on 25 July 2015 - 01:03 PM, said:

correct me if I'm wrong but isn't that a poor way to do matchmaking, is with a simple win/loss ELO only? Because I could be doing great in every single game, get 6 kills, 1000 damage etc and yet my team just happens to keep losing and so my ELO is super low. Or vice versa I could be complete garbage get 0 kills and 50 damage each game but I just happen to keep winning cus my teammates carry me so my ELO is high. Wouldn't it be smarter to do matchmaking based on personal score.

i.e. as someone might of said, if you lose but you get 6 kills and 1000 damage maybe your ELO should still skyrocket. Or if you win and get 0 kills and 50 damage, maybe your ELO should plummet, etc.

Finally! Somebody understands, what I'm talking about. My own experience of about a year of playing this game shows, that this ELO hell exists and you can easily stuck in it forever.

P.S. I play in public queue only - never played in premade.