59 replies to this topic

[zmeul]

12 matches played, for the 1st two I didn't took screens; 1st was a win, 2nd a loss

Spoiler

out of 12 matches I played only 3 were wins
8 with AS7-RS and last 4 with STK-5M; it went from bad to just "hilarious" .. so I just quit

I cannot understand PGI's ELO; I don't consider myself a good player, somewhere average perhaps
but there is no way to tell, no way to evaluate yourself - there is no way to tell if you doing better or just plain worse
at the rate I'm losing it can't go up, can it?!

ELO should put people of the same skill against each other, but I can't see it
why? I don't know .. maybe because there are very few people of my skill playing, or just there aren't enough active players to make a proper ranking

with an working ELO I should've have a 50% chance to win a match, again .. I don't see it
yes, I know the data pool is insignificant, but frankly I got mad and stopped playing

Edited by zmeul, 01 March 2013 - 07:51 PM.

[Pan]

Let me see if I can do this......

You should not have pug farmed.

amirite guys?

[zmeul]

Pan, on 01 March 2013 - 07:54 PM, said:

You should not have pug farmed.

huh?!

[Homeless Bill]

Just play more matches, and it will slowly balance; repetition is what makes the system work. Here's what stands out to me: only one 8-0 roll and only a single 8-1 roll. Pre-Phase 3, almost 50% of matches I played were 8-0 for one side.

[Pan]

Forgive me

zmeul, on 01 March 2013 - 07:59 PM, said:

huh?!

Oh never mind me I was making a funny. Folks will be along any moment to tell you its all your fault. To be honest the MM is a lil strange for me too. I was experiencing some crazy weight and mech mismatching for awhile. I think it had to do with the times I play. Times are pretty sporadic for me. During prime time its stomp fest. I am 50/50 on the win lose side of things. Some people people claiming great games but I am not seeing it. Now those folks will be by to rip me apart too.

[Mawai]

He is asking if you have spent any time playing in a pre-made team during the ELO seeding period from February 5th onward.

Many folks who play pre-made achieve better than average win/loss success ... not as a result of their individual skill but as a result of the coordination resulting from even a minimum level of team work.

People who played a lot with pre-made groups in the time frame since Feb 5th find themselves with very high ELOs that are not generally reflective of their actual skill. These folks are finding themselves dropped in matches against other pre-made groups and with people whose individual skill sustains the high ELO they currently have. Folks in this situation are finding themselves losing matches ,,, sometimes a lot of matches ... until their ELO reaches the appropriate level for their skill.

In addition, ELO is determined on a per class basis ... light, medium, heavy and assault. The initial ELO score is around 1300 which will put you in the middile range of player ability. However, this may or may not be appropriate for your actual play in that class of mech. As you either win or lose your ELO will adjust to the appropriate level.

Ultimately, you should expect about a 50/50 chance to win any match. A W/L ratio close to 1:1 and good, fairly even matches, is the expected outcome of the process but it will take some length of time playing in each mech class for your ELO to settle to something close to what it should be.

In addition, if they have not implemented some sort of group compensation effect, then I would expect a weekend spent in pre-made groups will boost your ELO ... which may give you a number of losses if you start dropping in PUGs again for a while.

[zmeul]

Mawai, on 01 March 2013 - 08:08 PM, said:

He is asking if you have spent any time playing in a pre-made team during the ELO seeding period from February 5th onward.

since I play MWO, back from CBT to today I go solo
last week it was 1st time when I played maybe 2-3 matches with 2 people who recognized me from my WoT days
I definitely didn't pug stomp to boost my rating

back when suicide farming was a thing I did it a lot, since then, I saw my W/L going up (0.77) and my K/D ratio is now 0.92
looking at those stats, something is rotten in Canada

Edited by zmeul, 01 March 2013 - 08:24 PM.

[Thirdstar]

zmeul, on 01 March 2013 - 08:20 PM, said:

back when suicide farming was a thing I did it a lot

Why would you do that?

But if you say it was at the start of Open Beta i'll probably be a bit more understanding.

[Pat Kell]

I may be wrong (completely) but as i understand it, ELO makes matches based on total ELO score rather then individual. It takes the total ELO score of all 8 on 1 team and trys to match it up with 8 other players who have a similiar TOTAL ELO score. So if you have a low ELO score, you may be put in with some high ELO score people while facing a bunch of medium ELO score people and it will seem sometimes like a mismatch. So you may have 1 elite and 7 rather new or lower skilled people against 4-5 average and 3-4 lower level people.

And what Mawai said sounds accurate. If you don't play alot or drop it groups a lot, that will give ELO fits when you pug.

[Thirdstar]

Pat Kell, on 01 March 2013 - 08:31 PM, said:

I may be wrong (completely) but as i understand it, ELO makes matches based on total ELO score rather then individual. It takes the total ELO score of all 8 on 1 team and trys to match it up with 8 other players who have a similiar TOTAL ELO score. So if you have a low ELO score, you may be put in with some high ELO score people while facing a bunch of medium ELO score people and it will seem sometimes like a mismatch. So you may have 1 elite and 7 rather new or lower skilled people against 4-5 average and 3-4 lower level people.

And what Mawai said sounds accurate. If you don't play alot or drop it groups a lot, that will give ELO fits when you pug.

Yesterday was the first time since Elo was implemented that I saw some serious wonkyness, also mixed with people playing Capwarrior online but that's not entirely related.

However, one evening of play does not a trend make so i'll reserve judgement till I get a few more days of play in.

[MischiefSC]

You may be suffering from the converse problem - if your stats are low enough you're going to be on teams full of sub-par players. This is going to cause big swings in your win/loss rates. My best recommendation is try to organize your fellow pugs for success and play a lot of games. Elo should average you out at 50/50 and then it's a matter of personal skill to grind that up. What was your win/loss rate from before?

[Pat Kell]

Thirdstar, on 01 March 2013 - 08:35 PM, said:

Yesterday was the first time since Elo was implemented that I saw some serious wonkyness, also mixed with people playing Capwarrior online but that's not entirely related.

However, one evening of play does not a trend make so i'll reserve judgement till I get a few more days of play in.

I did have a couple of matches today where the whole enemy team charged our base..nothing new there for me tho as I remember the days when it was a good idea to do that as far as rewards and such so its really no problem for me to be able to rally the pugs around me. usually. 1 of those matches about half my team headed for their base when ours began getting capped and the other half including me went back for ours. We had no chance and were gunned down without making much of a dent and their7-8 remaining guys easily out capped our 4. but oh well, is just a game and I will continue to adapt to whatever new sillyness PGI and the community throws at me.

Edited by Pat Kell, 01 March 2013 - 08:47 PM.

[zmeul]

MischiefSC, on 01 March 2013 - 08:38 PM, said:

What was your win/loss rate from before?

before I started farming? was a long time ago can't recall exactly, but was above 1

Edited by zmeul, 01 March 2013 - 08:46 PM.

[Mechabrozilla]

I'm in the same boat as the TS I'll consider myself fairly average to a little above average. Meaning with manual aim weapons and situational awareness I feel I do pretty well and on average will generally be in the upper half of the scoreboard for my team. Over the hero challenge weekend I felt the new ELO thing was good and working I was put in with matches that were fairly even with hardly any lopsided matches.

Then after the weekend event ended it was like oh noes the horrible pugging experience has returned. Being thrown on teams with guys that had worse aim then stormtroopers and running around like complete noob puggers. And I'm fairly sure I didn't go against to many pre mades. After a couple matches I watched my team more instead of the enemy and noticed it was like I again got cursed to be on teams where the guys wouldn't even try to aim and did the lemming death march against a team of guys that could aim.

After the weekend I've switched to trying to lvl up my Trebs, prior to this haven't played medium mechs much. But after a few matches I go back and use my other work horse mechs that I usually do pretty well in and I still end up in groups that it feels like maybe 1-3 guys are completely carrying the team and the rest couldn't hit a target unless it was less then 150m in front of them.

[Mackensen]

I think Elo is working alright. My W/L and K/D ratings are nothing to brag about but they are rising steadily and I have fun as they do.That is good enough for me. I play to relax and have fun, not to hate, whine and complain about "cheese ". Since the introduction of Elo I get less team mates and opponents with names related to below belt body parts and fecal matter.

[Xie Belvoule]

zmeul, on 01 March 2013 - 07:50 PM, said:

12 matches played, for the 1st two I didn't took screens; 1st was a win, 2nd a loss

Spoiler

out of 12 matches I played only 3 were wins
8 with AS7-RS and last 4 with STK-5M; it went from bad to just "hilarious" .. so I just quit

I cannot understand PGI's ELO; I don't consider myself a good player, somewhere average perhaps
but there is no way to tell, no way to evaluate yourself - there is no way to tell if you doing better or just plain worse
at the rate I'm losing it can't go up, can it?!

ELO should put people of the same skill against each other, but I can't see it
why? I don't know .. maybe because there are very few people of my skill playing, or just there aren't enough active players to make a proper ranking

with an working ELO I should've have a 50% chance to win a match, again .. I don't see it
yes, I know the data pool is insignificant, but frankly I got mad and stopped playing

[Squigles]

Of your losses, 1 was due to a base cap with only 1 enemy mech remaining, and another loss was a close 6/8 game. 3 Wins, 2 extremely narrow losses. With only 1 true steamroll loss, and the fact that your 2 narrow losses could have gone the other way with a well placed shot or 2, that's not really that bad of a distribution for such a small sample size of 12 games.

[JeepStuff]

zmeul, on 01 March 2013 - 07:50 PM, said:

12 matches played, for the 1st two I didn't took screens; 1st was a win, 2nd a loss

Spoiler

out of 12 matches I played only 3 were wins

12 matches isn't nearly enough to have Elo working for you. Think 25 games bare minimum, and probably double that to be honest. But I believe it will settle in for you and eventually most of your games will be good ones. For me, Elo is GLORIOUS. But it took between 25-50 games before I felt like the matchmaker was consistently putting me with players of similar skill.

[Xagul]

ELO isnt working at all.... Played about 100 games last week, 70% at the tournament last weekend. im allways in the Top4 in my team, my teams arnt getting better the get worse more and more that i played.... how is it possible that i get in such teams? is ELO matching 1 good player into a crap team? think so.....
2/10 games are balanced, good team fights, nice dogfights.... the rest? worse then mention it....

ah i played all games SOLO and i wasnt active since Dec. last year... so i dont got any games for the "pre-elo"

this sucks....

I dont give a **** if i HAVE to wait 5mins for balanced games!!!

[Chrithu]

Some thing from observations I've made:

Your Elo seems to be seperatly tracked for each chassis/variant(?) or at least each weight class. Meaning After playing A LOT of matches in my CTF's I have a 50/50 win chance with them and most of the time ppretty balanced matches (weight macthing problems which are known and to be fixed soon aside). When I then started to roll in my Cicadas I was back in "hell" getting mostly 8-0 or 0-8 macthes having a slightly higher win than loss rate in them. Which pointed at my Elo rising.

My guess is that it takes your Elo at least 50 matches to arrive at your bracket on the scale.

Then you have to know that the accuracy of the skill matching is "thrown overboard" (loosened) pretty fast in order to keep the time to find a match low.