31 replies to this topic

[TOPGUN Stinger]

badaa, on 25 June 2014 - 02:40 PM, said:

meta

Stinger: God help us.

[Davers]

You have to win the ones you were supposed to lose, and not lose the ones you were supposed to win.

But there are other factors than just straight Elo. That's the stuff PGI is keeping under wraps.

So just try to bring your A game to every match.

[80sGlamRockSensation David Bowie]

You will know when you are in the highest levels of "elo" matchmaking.

You will see the same familiar names every match.

Those stats you posted, you will no longer give any ounce of care as to what they are.

You and only you will carry then entire other 11 players, day in, day out. You must actively carry the team for a victory to happen.

You will enter solo, and still be put in Alpha lance. Every. Time.

Breaking 1000 damage and 6 kills suddenly becomes mundane, and "meh".

Routinely headshotting people and being accused of hacking by simply playing will become the norm.




This is when you have entered the higher brackets.

[Flyby215]

Question for the OP: Why do you want to get into a higher bracket?

Longer wait times to find a match, you're hopelessly outgunned unless you're on TS, poptarts everywhere playing the same build over and over pulling massive damage, same players over and over...

For sake of curiosity though, back when Elo first started I played a meta-mech with a 3.06 W/L ratio and 4.9 K/D (formerly 5 but a couple drunk matches tanked it...) after 350 matches. I suspect that artificially inflated my elo, and it became more normalized once I started playing my Misery. It has over 1,400 matches with a 1.68 W/L and 3.42 K/D ratio. Between those two my assault class ends up vs the toughest competition MWO has to offer. (other assaults rounding out the class were played to mastery and stopped).

My heavy class ends up in a tough bracket as well, it is primarily carried by my Jester (2000+ matches) which far outweighs any other heavies I play (usually to master then it gets stripped and never played again). My Jester currently sits at 1.52 W/L and 3.22 K/D (proudly in a 2xERLL 4xML, 2xAMS configuration).

Anyway, you asked for numbers so those are samples of what it takes to land yourself against SwK HoL SJR on a fairly regular basis when solo dropping.

[Fatal25]

This is old, but may still apply.

http://mwomercs.com/...79-matchmaking/

[Lynx7725]

Since there's some relevancy, I may as well ask here.. Does ELO brackets even affect matchmaking that much?

I'm fairly sure there are some form of ELO bracketing, but my feel is that matchmaker works with whoever is online at the moment. Even high ELO players have to sleep at times, so the quality of the teams in your match is more predicated on the general pool of players that are online rather than ELO.

Getting to a higher ELO bracket is a bit odd if you play in a timezone where most people in your bracket aren't playing. That means the opposite team is less likely to get someone in your bracket. Otherwise, certain ELO brackets would never get games.

I play at various timings and I can honestly say, there are certain timings where players aren't even aware of basic fundamentals like avoiding moving in the open, vs. other timings where players don't even give you a shot. ELO doesn't seem to have any bearing at all. To me, ELO just help -- a bit -- to balance games so that you get challenging games. It doesn't mean your games would always be with people of your skill level.

My suggestion is just to concentrate on figuring out how to help your team win games, even in PUGs. Fundamentally, that;s more rewarding, both personally and financially in game. Once you settled into that role (king-maker?), your ELO, if it still remains important to you, would naturally improve over time.

[XxXAbsolutZeroXxX]

Flyby215, on 25 June 2014 - 06:57 PM, said:

Question for the OP: Why do you want to get into a higher bracket?

For the most part, the only games I lose are ones where I make silly mistakse. Or players on my team make mistakes. I never feel like I lost a match because players on the other team were more skilled.

In a lot of games, it seems like I have extremely low quality opponents. Sometimes, I can take them on 2 or 3+ versus myself and come out on top. Not because I'm good. Moreso because they stand out in the open and don't move & maybe don't understand the basics. A few days ago there was a K2 standing out in the open on canyon network. I alphaed at long range into his arm. Shot his arm PPC off. Waited for him to run for cover. He didn't. I shot off his other arm. He still stood in the same place. I shot off his leg. Still he made no attempt to move or retreat. He was firing PPC's and AC's at me the whole time. But, I was moving and most of his shots were missing.

Those types of games make me feel like I'm wasting my time. I would much rather lose to someone skilled than beat people who don't know any better than to stand out in the open and let people tee off on them.

I guess I want to raise my ELO so I can have a higher quality of opponents and maybe a higher qualify of allies as well.

I would also like to see how broken the jump sniper meta is. I don't see a lot of coordinated or skillful jump sniping.

At the moment, my prime timberwolf has the best stats. Its sitting at 2.0 W/L and 3.5 KDR after 13 games. I have a feeling those numbers will drop significantly, though. A lot of my mechs start out with those types of numbers and decline.

Edited by I Zeratul I, 25 June 2014 - 07:53 PM.

[LordKnightFandragon]

Deathlike, on 25 June 2014 - 04:54 PM, said:

3 requirements:

1) Win early and often.

2) Contribute to your team significantly, as much as possible. It doesn't mean have a super k-d ratio (it helps, but it's not a guarantee of winning), but also capping or RTB whenever necessary so you don't lose outright.

3) Do it with meta, if only to accelerate the process.

Soo, even though I have an a -4WL and a -12 KD, since I deal an avg of about 350dmg, get assists left and right and pop components off alot, the game thinks I need to be put on the crap team more often? That why im always with the window licking team that loses 12-2? But why was it like this clear back in beta? Where I would walk out, and die in 10 seconds or less? Was my acct given a random ELO of 45000?

This ELO rating thing is really damn ********....no doubt.

[Targetloc]

LordKnightFandragon, on 25 June 2014 - 07:57 PM, said:

Soo, even though I have an a -4WL and a -12 KD, since I deal an avg of about 350dmg, get assists left and right and pop components off alot, the game thinks I need to be put on the crap team more often? That why im always with the window licking team that loses 12-2? But why was it like this clear back in beta? Where I would walk out, and die in 10 seconds or less? Was my acct given a random ELO of 45000?

This ELO rating thing is really damn ********....no doubt.

When the frog sees the well he realizes that both teams were licking windows, which is why he's on 'the window licking team' more than 50% of the time.

[Deathlike]

LordKnightFandragon, on 25 June 2014 - 07:57 PM, said:

Soo, even though I have an a -4WL and a -12 KD, since I deal an avg of about 350dmg, get assists left and right and pop components off alot, the game thinks I need to be put on the crap team more often? That why im always with the window licking team that loses 12-2? But why was it like this clear back in beta? Where I would walk out, and die in 10 seconds or less? Was my acct given a random ELO of 45000?

This ELO rating thing is really damn ********....no doubt.

The "unintended" side effect is that you get more "underhive" teammates... which means you are forced to carry more.

There's nothing to be done to to "averages". when variance becomes a significant factor at times.

[Triskelion]

Mizeur, on 25 June 2014 - 02:25 PM, said:

Elo isn't perfectly correlated to W/L. Before every game, the matchmaker predicts a winner. If the predicted winner wins, there's no change. If the predicted winner loses, then there's a change to everyone's rating. I don't know the specifics for PGI, but generally the change is proportional to the difference of each team's elo. So a 10 point gap would be a small change. A 100 point gap would be a bigger change.

It generally takes lots of games--think hundreds--to really adjust your elo if you've been playing awhile.

Jeez, that seems a bit rough.

I'm surprised it's not linked to some kind of performance system, which most game devs seem to forgo these days in fear of elo exploits, or some such other cosmic horror. This game sort of seems to lend itself to practicing good behavior though. Or perhaps just getting a group of 4 and running a dedicated exploitative strategy. *Cough*

Edited by Triskelion, 25 June 2014 - 09:48 PM.

[blackicmenace]

FupDup, on 25 June 2014 - 05:41 PM, said:

So you want to climb to the peaks of Mount Tryhard?

I must be in the Scrub tier Ghetto.