123 replies to this topic

[NRP]

I've been told that the lowest ELO players are grouped in Charlie lance. I always end up in Charlie lance when I drop solo.

Therefore, I'm a scrub.

[Deathlike]

Ghogiel, on 20 June 2014 - 01:40 PM, said:

Ace level players no one has even heard of have Elo so high that carnage and ladys assault mech Elo can't make up for my deficit Elo in meds.

yeah...

My meds might have near as good stats as those CTFs even after tanking my SHDs in 12s anyway. who knows.

Well, I can subscribe to the First Come First Served Rule.

It could be that the premade in Alpha Lance was the MM's "original pick" for its Elo averaging purposes. You just happened to be the same-size, so by default they get Alpha Lance.

That's the only other logical explanation.

NRP, on 20 June 2014 - 01:43 PM, said:

I've been told that the lowest ELO players are grouped in Charlie lance. I always end up in Charlie lance when I drop solo.

Therefore, I'm a scrub.

BTW, I had to convert the Wolverine-6R from a brawler to a lurm boat. There's only enough times where you can get screwed on Alpine and feel like garbage.

[Ghogiel]

Deathlike, on 20 June 2014 - 01:46 PM, said:

Well, I can subscribe to the First Come First Served Rule.

It could be that the premade in Alpha Lance was the MM's "original pick" for its Elo averaging purposes. You just happened to be the same-size, so by default they get Alpha Lance.

That's the only other logical explanation.

Maybe Lady with his ozzy ping was group lead or something and it just drops us second >_>

Edited by Ghogiel, 20 June 2014 - 01:48 PM.

[Mc JR]

Wintersdark, on 20 June 2014 - 01:17 PM, said:

Anecdotal, but my observations have been, over absurdly many drops, that this is absolutely provable.

My assault Elo is reasonably high.

Where do you find your ELO?

Fatal25, on 20 June 2014 - 01:21 PM, said:

If you read this developer post from December 2012....

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

it seems to just say your ELO (whatever that really is) has some bearing on matchmaking, but I don't see anything that says it has to do with lance assignment.

This is true I don't think anyone says ELO does nothing, it has to do something, but i agree it says nowhere on any mwo website that ELO is linked to lance assignment. (BTW: thank you for finding this)

[Monky]

I mostly solo drop and am almost always Alpha lance. Sometimes Bravo, I don't recall ever being Charlie lance though.

[Roadkill]

WarHippy, on 20 June 2014 - 11:58 AM, said:

If there is a large gap in Elo from Alpha to Charlie then yes, but if there is a small difference in Elo then its safe to assume they are just as reliable or derpy as you.

Oh god, we're doomed.

[NeonKnight]

Sadly i SWEAR TO GOD the match maker simply does the following:

• CHOOSE MAP
  
• GRAB FIRST AVAILABLE PREMADE - Assign to SIDE A - Lance Alpha
  
• GRAB SECOND PREMADE - Assign to SIDE B - Lance Alpha
  
• GRAB PLAYER - Assign to SIDE A (starts to fill slots)
  
• GRAB PLAYER - Assign to SIDE B (starts to fill slots)
  
• REPEAT UNTIL 12 per SIDE.

and as someone who plays in a premade, this means for me:

I am ALWAYS ALPHA on the same damn side each and every time.

Please PGI, when you do sit down to rewrite the match maker code, can you just please gather players like something like the following:

• CHOOSE MAP
  
• GRAB FIRST AVAILABLE PREMADE - Assign to LANCE 1
  
• GRAB SECOND PREMADE - Assign to LANCE 2
  
• GRAB PLAYER - Assign to lances as needed to fill until 6 full lances.
  
• RANDOMLY ASSIGN ODD LANCES TO ONE SIDE
  
• RANDOMLY ASSIGN LANCE 1 (and 2 for their side) to ALPHA, BRAVO or CHARLIE
  
• RANDOMLY ASSIGN LANCE 2 to one of the remaining two lances
  
• ASSIGN LANCE 3 to last lance
  
• START MAP

This allows EVERYONE the chance for some variety.

[blackicmenace]

Hellen Wheels, on 20 June 2014 - 09:33 AM, said:

You were told wrong. I drop solo exclusively, and 99% of the time I wind up in Alpha lance.

I solo pug 99.9% of the time and 99% of the time I am in Charlie lance.

[Roadkill]

Deathlike, on 20 June 2014 - 12:35 PM, said:

So, I looked it up for you, just to make this clear (and proof, because you know, links or it didn't happen):

http://mwomercs.com/...ost__p__3298159

Awesome, thanks. I apparently really suck at forum search... I'd have probably never found it on my own.

I'm glad that they did finally get it implemented as it's far more accurate than a single rating for all Mechs.

[blackicmenace]

If ELO is based off Wins and losses only that does not seem like a very accurate assessment of skill. Sure it does to a degree but its not like you are solo you have 11 other people to factor into the win or loss. Teams that work together best are likely going to get the win. This game needs to have voice coms built in so everyone can get on the same page not that most clans with teamspeak/vent would mingle with us solo scrubs.

[Wintersdark]

Deathlike, on 20 June 2014 - 01:46 PM, said:

Well, I can subscribe to the First Come First Served Rule.

It could be that the premade in Alpha Lance was the MM's "original pick" for its Elo averaging purposes. You just happened to be the same-size, so by default they get Alpha Lance.

That's the only other logical explanation.

Actually, I could get right behind this explanation.

As I said earlier, it doesn't make sense to me to rank players in the match via Elo, given what I see at high, medium and low elo's. If you solo drop, you'd find yourself at all points of the scale depending on if you where the high one or the low one in the group (you'd not always going to be the high player, after all - even if you have an above average elo, the system should be slotting you in with better players as well as worse ones).

However, if the MM is (and we know it is) grabbing a group from the queue as the first part of building a match, then trying to build a match with that as the target point, then it stands to reason that the first few solo puggers it grabs will be roughly the same Elo rating as that team, with later additions diverging increasingly. If its a reasonably high-elo team, statistics indicate that more people added will be lower than higher, given the bell-curve distribution of player Elo rankings.

However, this would get more interesting if the MM was grabbing multiple groups at once(with the goal of "using up" groups in the queue before filling with solo players), as additional groups would tend to push average weightings around a lot. Time to talk to Mr. Berg (if he can/will talk about this)!

Quote

BTW, I had to convert the Wolverine-6R from a brawler to a lurm boat. There's only enough times where you can get screwed on Alpine and feel like garbage.

I love my mediums. A lot. I'm nowhere near as strong with them as I am with my Assaults, or even my Heavies, but I love them all the same. But that situation right there is why most of my mediums are built for range - If I set a medium up as a brawler, I *WILL* drop almost exclusively on Alpine with it, and perform atrociously badly as a result.

[NRP]

Deathlike, on 20 June 2014 - 01:46 PM, said:

BTW, I had to convert the Wolverine-6R from a brawler to a lurm boat. There's only enough times where you can get screwed on Alpine and feel like garbage.

Yeah, that match on Alpine last night was boring as hell.

[Lyoto Machida]

blackicmenace, on 20 June 2014 - 02:38 PM, said:

If ELO is based off Wins and losses only that does not seem like a very accurate assessment of skill. Sure it does to a degree but its not like you are solo you have 11 other people to factor into the win or loss. Teams that work together best are likely going to get the win. This game needs to have voice coms built in so everyone can get on the same page not that most clans with teamspeak/vent would mingle with us solo scrubs.

Eww...plebians. I shall not have my headset speakers infected by them.

[blackicmenace]

Lyoto Machida, on 20 June 2014 - 02:56 PM, said:

Eww...plebians. I shall not have my headset speakers infected by them.

Did I mention it should also have the option to mute players?

[eaglemaster42]

NRP, on 20 June 2014 - 01:43 PM, said:

I've been told that the lowest ELO players are grouped in Charlie lance. I always end up in Charlie lance when I drop solo.

Therefore, I'm a scrub.

Is that really how it works? I guess I'm a scrub too.

[Wintersdark]

Mc JR, on 20 June 2014 - 01:58 PM, said:

Where do you find your ELO?

You don't.

However, I've been playing this game a very long time, with many, many drops under my belt. I know who's better than me, who's worse than me. I know who's good to very good at the game(from regular pugging play, grouping with people, watching tournies, etc), and I see them frequently when dropping in my Assaults. As such, I know my Assault Elo is far enough up the chain to frequently see those at the top of it. Not that I'm right up at the top by any means - populations there are thin given the whole bell curve distribution - but close enough.

All those people vanish once I'm in mediums/heavies, unless they're in non-standard weight classes themselves. I see a MUCH wider range of players in those mechs too, particularly my mediums. As such, it stands to reason those two weight classes are more "average" overall. I see more new players in my Mediums(several) than my Heavies(few), so I can safely assume my Heavy Elo is somewhat higher than my Medium Elo.

I know for a fact my Light Elo is at the bottom of the steering wheel underhive, because I deliberately put it there initially, and hold it there by dint of being freaking terrible in lights. If I were to reset my light Elo (you can't do that) and start at the New Player point again, I wouldn't go so far down (I'm at least moderately good at the game, and can build a good mech) but when you're right at the bottom there matches are extremely random, because there are so very many inconsistencies there.

Anyways, I digress.

You can work out your Elo ratings roughly by deduction, when you know enough about the game and roughly where other people are. You do need to realise though that your personal stats are irrelevant to your Elo, even your win/loss ratio.

[Deathlike]

Wintersdark, on 20 June 2014 - 02:49 PM, said:

I love my mediums. A lot. I'm nowhere near as strong with them as I am with my Assaults, or even my Heavies, but I love them all the same. But that situation right there is why most of my mediums are built for range - If I set a medium up as a brawler, I *WILL* drop almost exclusively on Alpine with it, and perform atrociously badly as a result.

Before I sold the Hunchback-4P (wanted to grind something else), I ended up using a Large Laser on it with 6 Meds... only to avoid Ghost Heat.

I used to have patience trying to brawl with the 9 med 4P, but with the meta for what it is and being limited in range IS a downside, you have to find a way to cram just 1 LL/PPC, AC5, AC10, a pair of LRMs (equaling an LRM15 or better) or something along those lines to stay useful at any range.

Brawling is still frowned upon... so being "that medium holding out in the back" is not how I'd like to be at all times... but it is what it is.

[Wintersdark]

blackicmenace, on 20 June 2014 - 02:38 PM, said:

If ELO is based off Wins and losses only that does not seem like a very accurate assessment of skill. Sure it does to a degree but its not like you are solo you have 11 other people to factor into the win or loss. Teams that work together best are likely going to get the win. This game needs to have voice coms built in so everyone can get on the same page not that most clans with teamspeak/vent would mingle with us solo scrubs.

Wins and losses is the ONLY thing that can be used. Nothing else works. But it's not just wins and losses, it's wins and losses relative to the opposing teams' rating. Thus, losing a match to a higher ranked team has little to no impact on your ranking, for example.

You can't use stats and such, because so much of what determines who wins a match is not quantifiable. Damage? No. You could be some meathead firing all yer guns from the back, grossly inaccurately and not actually accomplishing anything. Damage in and of itself is worthless. Kills? Kills are important, but it doesn't matter if you or your teammate gets the kill. What if you crippled a mech, and moved on to a threatening target, then some opportunistic moron chases after the crippled mech? He gets a kill, you don't, but he actually hurt your team's chances to win while you helped it.

It goes on and on with all the stats. There's only one thing that works. Wins. Maybe you're terrible at the game, can't hit a Dire Wolf's center torso from 50m. But you can lead. You're great at getting pugs to act (comparatively) cohesively. Maybe you're a great scout, or just have a good grasp of when to lead a push. There are countless actions that contribute to victory but cannot be measured.

Hell, this is why premades end up with higher Elo ratings. Not necessarily because the players are better at the game, but because they have comms (or at least just know each other's play styles well) and can act more cohesively. Because of that, they win more.


TLDR: Elo isn't necessarily a measure of skill, but rather a measure of how much a given player contributes to his teams victory on average. These are not the same thing.

It's a rough valuation, by necessity - it's working on very abstract values, after all - and it requires a LOT of matches to really get tuned well for you... And the more different mechs you use, the more "averaged" it'll be across those mechs and thus the more.

[Almeras]

Alpha is your pre-made, bravo is normally duo's or threes sometimes 4mans, Charlie is the short bus.

If Alpha lance is sub 200tons (lights and mediums) you'll have a nasty handicap.
If Alpha lance is sub 150tons you'll probably loose.
If Alpha lance is 325tons (2 assaults 2 heavies) you'll be evens.
If Alpha lance has silly builds you'll loose.
If Alpha and Bravo both have 4mans then expect the same on the other side.
If Charlie has all your assaults you'll most likely loose.
If Charlie lance is all in 3rd person or champion mechs you'll probably loose
If Charlie has lots of LRMs you probably win.

This pretty much rings true as the MM is blind to elo vs meta. The 'elite' 4mans know the best way to engineer a win is to make sure the 4man take all the weight and go full meta so to eat the other teams Pugs (Charlie lance) as quickly as they can before their own pugs get them selves killed. Once the enemy is 4 down chances are they aren't coming back up. If your 4man is light then the chances are they will loose the pug eating race and find them selves 4 vs 8+

Its gotten to the point where i can predict a win or a loss 75% of the time just by looking at the launch board.

[Fatal25]

36 games and finally just got in Bravo lance for some unknown reason. We won 12-10. No clue who on my team was actually grouped or not. I don't know why PGI can't just explain the voodoo they use to make the lances, not that I guess it really matters, just mix it up for the sake of solo droppers.