176 replies to this topic

[Krivvan]

Like the title said, I remember earlier today in one of the premade vs pug threads someone saying that people were just using anecdotal evidence and that if someone posted screenshots then it would be obvious that almost every single game a pug plays will be against a premade.

I decided to take them up on that challenge:





These are my stats screens before and after to show that I didn't skip any games in between and that these games all occurred consecutively (I hope it also doesn't come off as me bragging...the initial k/d ratio was due to playing in a small premade for a few games during the start of the OB):





All the wins and losses should add up along with the kills, deaths and cbills.


So in the end out of 2 games I've only been able to confirm that 2 of the opposing teams were premade teams (and one was a small partial premade).

Neither of the premade teams announced that they were premade until I asked.

Even assuming all the unconfirmed opponents were premades, that means only approximately 5 premade teams fought in 19 games. Somewhat close to what the earlier thread about the "Real cause for lopsided losses" found.

This is all just another attempt of mine to actually record some data about how common premade teams are except this time with screenshots instead of tables that I could just make up.

All in all this is about 3 hours of playing.

EDIT: Woops, game 5 should say PUG VS. Unknown. I had a hunch they were premade, but I forgot to ask them. A little bit of forum profile searching means I'm confident that they aren't a corp though.

EDIT 2: Modified the image because of updates.

Game 15 video here:


Game 16 video here:


Game 16 is an example of simple advice to the pug team turning it into a stomp (although the fact that the enemy team seemed to have a non-responsive mech is also a huge factor).

Game 18 video here: www.youtube.com/watch?v=wB9xT50fkpQ

If anyone asks what my build is, here it is: https://dl.dropbox.c...JennerBuild.png
And yes I know I have 4 useless jump jets. I still leave them on.

Edited by Krivvan, 01 November 2012 - 07:59 AM.

[LogicSol]

That about matches up with my general experience. 1 in 4 to 1 in 8.
Good work.

[Johnny Reb]

I'm a member of a clan but have only been puging since OP. Ran into the same premade(not impressed and neither were the pugs when they flashed the recruitment site). However, In the roughly 120 matches since OB only ran into premades about 9 times.(one being my own) That said I have been playing some during off hours. Only see premades around 7pm-1am est.

Edited by Johnny Reb, 31 October 2012 - 08:00 PM.

[JollyRoger4u]

well, statstics are only viable with a minimum of 1000 samples (if my memory from the school days hold up at least ) so you got a lot of work to do! go git em!

seriously though, good work and nice to see that someone actually try to find actual facts instead of making *beep* up as they go along

gg

[Cochise]

My experience has been about 1-10. I thought Garth was going to release some stats about this. Anyway, nice work.

[Lavrenti]

I've asked groups I've played with/against if there are any premades in the match too. The problem is that not all premades will respond to such questions, and some will actively deny it if you ask before the match (not sure why, don't want to scare off opponents perhaps?). I haven't been keeping track, but your results roughly fit with the sort of feeling I've been getting from my games - about 1/4 of the time it's clear you're up against a premade, and another 1/4 or so it's either a partial premade or a well-coordinated PUG which usually results in a similar stomping. The remainder of the games can go either way.

[HAV0C]

<S> Thank you for this effort Krivvan. Well done sir.

I can't speak for any other teams, but Death's Hand Brigade will announce if we're running a full 8 man pre-made group during the countdown or initial few moments of the game.

This generally consists of "Good luck from Death's Hand Brigade" or "GL HF from DHB" At the very least.

We do this in an effort to identify ourselves first and also to try and find out if we're up against another pre-made group so we can try and get a good match in against them (professional courtesy as it were).

I know I played for approximately 2 hours last night and in all the matches I played (15 or so I believe) We met only two pre-mades who identified themselves as such. One or two others were suspect, but at the very least the vast majority of matches were not against pre-made groups. So I agree and concur with your assessment here.

<S> Great work and thanks for posting it. See ya dirtside sir.

-Colonel Andrew "Havoc" Davis
Death's Hand Brigade Mercenaries

[Krivvan]

JollyRoger4u, on 31 October 2012 - 08:00 PM, said:

well, statstics are only viable with a minimum of 1000 samples

Well then, time to get started!

[Stonefalcon]

Yeah I have to agree this data seems concurrent with what I've been able to ascertain.

[FerretGR]

Cheers, Krivvan. Number 1 guy on your team in every game: well done.

Given that most games end up being ROFLstomps one way or the other (I'd estimate, conservatively, 80%), and given that even the most liberal estimates have a PUGger meeting up with a premade 25% of the time, there's obviously an issue beyond the premade one. Sure, a premade will stomp just about any PUG, but it's obvious that the lack of a matchmaker pairing teams based on skill level is the bigger problem. Wish the devs had tackled this instead of the phases.

[EarthenMight]

With the negative feedback pre-mades have been getting, I am shocked that any actually told you they were a pre-made team.

Truth be told you can not rely on the other team to tell you they are a pre-made or what you may think is a pre-made team. The only people that know they are a premade group is PGI, and they are not telling us anything.

[Lavrenti]

FerretGR, on 31 October 2012 - 08:09 PM, said:

Cheers, Krivvan. Number 1 guy on your team in every game: well done.

Given that most games end up being ROFLstomps one way or the other (I'd estimate, conservatively, 80%), and given that even the most liberal estimates have a PUGger meeting up with a premade 25% of the time, there's obviously an issue beyond the premade one. Sure, a premade will stomp just about any PUG, but it's obvious that the lack of a matchmaker pairing teams based on skill level is the bigger problem. Wish the devs had tackled this instead of the phases.

The thing is that better communication is a force multiplier, independent of player skill. A PUG that can't use VOIP is likely to get stomped by a group that does, even if all the mechs and skill levels are the same. So skill isn't the only determining factor. Personally I think that the phases will be a day late and a dollar short, they won't completely fix the premade problem since it seems even a 4-man premade will wipe the floor with 8 randoms. Better VOIP and command tools might actually do more good than the phases.

[Krivvan]

First Knight, on 31 October 2012 - 08:14 PM, said:

With the negative feedback pre-mades have been getting, I am shocked that any actually told you they were a pre-made team.

Truth be told you can not rely on the other team to tell you they are a pre-made or what you may think is a pre-made team. The only people that know they are a premade group is PGI, and they are not telling us anything.

Part of the issue though is that some people are saying that close to every game they play has a premade group advertising a clan/corp or something. I just wanted to show that this isn't usually the case.

[Vermaxx]

Ok, let's say half of them were premades First Knight, and most of them lied.

Their skill level didn't indicate to Krivvan, who is clearly observant, that they were a coherent team.

Meaning premades are everywhere, but most of them are terrible. Either way the problem has been exaggerated.

[EDMW CSN]

First Knight, on 31 October 2012 - 08:14 PM, said:

With the negative feedback pre-mades have been getting, I am shocked that any actually told you they were a pre-made team.

Truth be told you can not rely on the other team to tell you they are a pre-made or what you may think is a pre-made team. The only people that know they are a premade group is PGI, and they are not telling us anything.

Even if a third of his games or half were pre-made, he is doing insanely well.

[Stoindrae]

37.4% of all statistics are made up on the spot.

I've had runs like that as well .. tho over the past hour .. I tried to run a few simple pug drops before bed... 5 in a row were organized, focused firing groups of 8... while our side had 1-3 drops/afks/ or just plain sat at our base and refused to fight (you know who you are /shakes_fist)

No idea how many of the 5 were in a voice chat.... but I recognized people I have teamed with in TS3 in at least 3 of the matches. /shrug .... one group was doing the "SQUAWK" thing.. tho I have no fuggen clue what that is supposed to mean.

[Lavrenti]

Stoindrae, on 31 October 2012 - 08:19 PM, said:

.... one group was doing the "SQUAWK" thing.. tho I have no fuggen clue what that is supposed to mean.

I'm thinking it might be a low-rent form of electronic warfare. Spam in the global chat box makes it useless for people trying to communicate through team chat, reducing their ability to coordinate. Just a guess, though.

[Voodoo Circus]

Nice job mate
But you failed to point out the most important information contained in your research
That to me is the prime reason people consider they oponents to be premades
and therefore atribute their loss to a premade group

Look at the scoreboards of the Pug vs Pug games

Most of them are 8-0 for the winning side

Seems like losing a mech early on really tips the balance as a lot of resonable people have been saying
Plus in that 8-0 you can have mechs that are 1 or 2 hits away from getting killed as people will generaly retreat when
they got no weapons left

So when you get "stomped" in a 8-0 dont assume you are playing a premade

[Magik0012]

1 ) Excellent post.
2 ) You dirty, Jenner scum!
3 ) ;-)

[FerretGR]

Lavrenti, on 31 October 2012 - 08:17 PM, said:

The thing is that better communication is a force multiplier, independent of player skill. A PUG that can't use VOIP is likely to get stomped by a group that does, even if all the mechs and skill levels are the same. So skill isn't the only determining factor. Personally I think that the phases will be a day late and a dollar short, they won't completely fix the premade problem since it seems even a 4-man premade will wipe the floor with 8 randoms. Better VOIP and command tools might actually do more good than the phases.

Agreed... that's something else that I think should have been implemented before letting the masses in here. In-game VOIP and a lobby system, a skill-based matchmaking system... the problem is multilayered, but I think with premades and the Phase system they're peeling off a relatively insignificant layer.