Jump to content

- - - - -

Matchmaking Phase 4 Follow-Up - Feedback


277 replies to this topic

#81 Hauser

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 976 posts

Posted 22 April 2013 - 11:15 AM

View PostPhoenixFire55, on 22 April 2013 - 10:55 AM, said:

Lets put it this way ... McPopperson (and others) kept their entire team busy long enough for you mr 3L-son to go and cap.


They cooperated and won the game! Everybody did what they had to do and we even have a result to measure it by! Wonderful.

Edited by Hauser, 22 April 2013 - 11:15 AM.


#82 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 22 April 2013 - 11:44 AM

View PostHauser, on 22 April 2013 - 10:52 AM, said:

Let me put it another way then. You're treating elo as a digital e-peen rather then a part of the match maker. The only thing you should measure the effectiveness of elo by is the quality of the matches it helps the matchmaker produce. If a rating is deserved or not is utterly irrelevant.


Well what I'm seeing is that my current games aren't any different from what I had in closed beta. 8:0 stomps still happen very often and either way. Close and exciting matches do happen too, but far less often, just as it was in closed beta. I still see complete newbs in my games, just like I've seen them in closed beta. I still randomly get good teams and bad teams.

I understand how a matchmaker with Elo is supposed to work, I'm just saying its NOT working as it supposed to. Nearly everyone has same Elo score right now which only means there is no point in it.

View PostHauser, on 22 April 2013 - 10:52 AM, said:

The end result is indeed that, noobs will play with noobs if they manage to hang on long enough. And this is what we need because you dont want to play with them (on your team) and they don't want to play with you (on the other team).


If only it was true my win/loss ratio would be twice better ... lol

View PostHauser, on 22 April 2013 - 10:52 AM, said:

The end result the system is aiming for is that indeed you will win 50% of your games. More if you're improving, less if you're getting worse.


If everyone wins 50% of his/her games everyones Elo is gonna be same. No point in haveing it.

View PostHauser, on 22 April 2013 - 10:52 AM, said:

Also do me a favour, if you are going to type this much please make a coherent argument. I'm not interested in replying to every mismatched paragraph.


Likewise ...

View PostHauser, on 22 April 2013 - 10:52 AM, said:

Now I regret to see that you think this is a bad environment for new players. I regret too see that you think that a noob friendly environment is a bad thing. I regret to see that you suggest trashing your elo and stomping on noobs so you can have an easy experience. An attitude like this prevents the game from growing.


Here we go into personalities again...
This game is supposed to be 'thinking person game'. That eliminates noob-friendly environment by default. Only thing we can and must have is some sort of tutorial and different treatment in terms of matchmaking for newbie players with less then say 100 games.

Also, if you wanna make it personal we can discuss it somewhere else. Right now its you who's not making any 'coherent argument'.

#83 Symbiodinium

    Member

  • PipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 162 posts
  • LocationNew Zealand

Posted 22 April 2013 - 12:35 PM

View PostPhoenixFire55, on 22 April 2013 - 11:44 AM, said:


I understand how a matchmaker with Elo is supposed to work, I'm just saying its NOT working as it supposed to. Nearly everyone has same Elo score right now which only means there is no point in it.


It's unclear to me how Elo can work at all with team games with different teams in every round if Elo is only based on the team win/loss. It works in individual competitions, but when I'm just one person out of a team of eight, with different teammates each round, I'm not sure what the Elo is measuring. That is, I think my ability to determine the outcome of the game as an individual is less than the random variation introduced by random teammates and frequent disconnects (on either team). Starcraft 2 uses an Elo system as well, including with team games, but you have a different Elo score for each team/combination of players. Combined with disconnects and players leveling inferior mech variants, it's unsurprising that Elo scores are very evenly distributed.

Edited by Symbiodinium, 22 April 2013 - 12:36 PM.


#84 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 22 April 2013 - 12:43 PM

View PostSymbiodinium, on 22 April 2013 - 12:35 PM, said:


It's unclear to me how Elo can work at all with team games with different teams in every round if Elo is only based on the team win/loss. It works in individual competitions, but when I'm just one person out of a team of eight, with different teammates each round, I'm not sure what the Elo is measuring. That is, I think my ability to determine the outcome of the game as an individual is less than the random variation introduced by random teammates and frequent disconnects (on either team). Starcraft 2 uses an Elo system as well, including with team games, but you have a different Elo score for each team/combination of players. Combined with disconnects and players leveling inferior mech variants, it's unsurprising that Elo scores are very evenly distributed.


Whole theory is based on the assumption that 1 player if he is good improves chances for his team to win and visa versa. So in theory he'll win more then 50% of his game. In reality it doesn't matter how good you are, you can't swing a match in your favor on your own in 99% of times when you are in a worse team, so such things as 'win against a better opposition' that are supposed to increase your Elo just never happen.

Edited by PhoenixFire55, 22 April 2013 - 12:52 PM.


#85 RenegadeMaster

    Member

  • PipPipPipPipPipPipPip
  • 587 posts
  • LocationUSA's Caustic Valley: Arizona

Posted 22 April 2013 - 01:17 PM

The fact that these changes were made a day before a competition and that I did not play much in the ~24 hour window makes it more difficult from in game experience to tell how much of the impact on match quality is the result of Matchmaking changes or the nature of the challenge.

What I do know is that I have participated in at least 10% more matches that are very close in score (within 2 kills) or last two thirds of the match. Also, I am seeing more of the same players in drops recently. However, I recognize that the challenge could be contributing to either factor. In any case, I do appreciate the methdology behind Matchmaking and the recent changes to it.

#86 Wispsy

    Member

  • PipPipPipPipPipPipPipPipPip
  • Talon
  • Talon
  • 2,007 posts

Posted 22 April 2013 - 01:34 PM

View PostPhoenixFire55, on 22 April 2013 - 12:43 PM, said:


Whole theory is based on the assumption that 1 player if he is good improves chances for his team to win and visa versa. So in theory he'll win more then 50% of his game. In reality it doesn't matter how good you are, you can't swing a match in your favor on your own in 99% of times when you are in a worse team, so such things as 'win against a better opposition' that are supposed to increase your Elo just never happen.


Well you can. Then you increase in Elo score until you cannot, leaving you around the right kind of area. Also you make it out as if one team is always much much better then the other team. This is not really the case in my experience, even if it is 8-0 then it can still be a close game with a number of heavily damaged people getting that first couple of kills then snowballing the numbers advantage. That or one team got the positional advantage and that means an awful lot and can make it seem like your team is worse then they are.

#87 Lord Psycho

    Member

  • PipPipPipPipPip
  • Overlord
  • 177 posts

Posted 22 April 2013 - 01:55 PM

Keep this in place for another month or two and then tell us what happened. Hope fully this is the same results you'll get back.

Besides that..i've been seeing alot of different people around...

I've been dying a bit more often then I like for this tournament.... ELO dropped to nothing I bet....

Either way, it's nice but it's also been really weird. Hopefully it's just a tournament thing..I can't pilot an atlas for the life of me anyways.

Edited by Lord Psycho, 22 April 2013 - 01:58 PM.


#88 hammerreborn

    Member

  • PipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 3,063 posts
  • LocationAlexandria, VA

Posted 22 April 2013 - 02:22 PM

Quote


If everyone wins 50% of his/her games everyones Elo is gonna be same. No point in haveing it.


Quote


I understand how a matchmaker with Elo is supposed to work, I'm just saying its NOT working as it supposed to. Nearly everyone has same Elo score right now which only means there is no point in it.


The first statement directly contradicts the second.

Edited by hammerreborn, 22 April 2013 - 02:23 PM.


#89 Hauser

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 976 posts

Posted 22 April 2013 - 03:42 PM

View PostPhoenixFire55, on 22 April 2013 - 11:44 AM, said:

Well what I'm seeing is that my current games aren't any different from what I had in closed beta. 8:0 stomps still happen very often and either way. Close and exciting matches do happen too, but far less often, just as it was in closed beta. I still see complete newbs in my games, just like I've seen them in closed beta. I still randomly get good teams and bad teams.


My experience is rather different. When I drop in a medium I tend to get reasonable teams. When I drop in a heavy I get the games you describe. I suspect my elo for my mediums is rather high, I've got good ratios on them. My heavy ranking is probably average. So I would definitely say it the system is working.

You're seeing those new guys probably because they're new and still have an average elo rating. Something they shouldn't have, or at least should be rid of at an accelerated rate.

View PostPhoenixFire55, on 22 April 2013 - 11:44 AM, said:

Only thing we can and must have is some sort of tutorial and different treatment in terms of matchmaking for newbie players with less then say 100 games.


And I couldn't let this one go unnoticed.

Quote

In other words noobs play with noobs and learn nothing. Noob friendly game, you are gonna win 50% of your games even if you are a complete *****? Is that what we need? Don't know about you but I'd rather see a game that motivates people to improve their skill to win more games. Right now there is more reason to 'noobify' yourself to play in easy mode vs non-so-good players.


edit:

To clarify; how can you on one hand suggest too keep new players separated from the general population, but be against the idea of matching people based on their performance in game? It achieves the same objectives but better.

Now I'm not saying it's doing what it is suppose to. I've so far been trying explain why it can do what it is supposed to do.

Edited by Hauser, 22 April 2013 - 03:57 PM.


#90 Bunko

    Member

  • PipPipPipPipPip
  • 140 posts
  • LocationJapan

Posted 22 April 2013 - 03:51 PM

Getting slightly better matches now.

Doesn't this all get chucked once Lobbies come into the game?

#91 Hauser

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 976 posts

Posted 22 April 2013 - 03:59 PM

View PostBunko, on 22 April 2013 - 03:51 PM, said:

Getting slightly better matches now.

Doesn't this all get chucked once Lobbies come into the game?


The lobby system grew out of the dropship mode. So I suspect it will be a pre-drop lobby where you can coordinate which 4 mechs you bring. People can still be put into a lobby by the match maker.

League of Legends has a similar system.

Edited by Hauser, 22 April 2013 - 04:00 PM.


#92 jay35

    Member

  • PipPipPipPipPipPipPipPip
  • The Hammer
  • The Hammer
  • 1,597 posts

Posted 22 April 2013 - 08:16 PM

View PostHayashi, on 19 April 2013 - 08:15 PM, said:

My 8 man has never gotten to drop even once. Just set the ELO tolerance to be good for 4man to PuG only, then completely eliminate ELO for 8 man games. When we're talking coordinated groups, ELO's irrelevant anyway - the results will be far more determined by the way the group works together, over individual skill - so forcing it on 8mans is pointless.

Good point!

#93 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 23 April 2013 - 12:07 AM

View Posthammerreborn, on 22 April 2013 - 02:22 PM, said:

The first statement directly contradicts the second.


How does nearly everyone winning 50% of their games contradicts with nearly everyone haveing the same Elo score?

#94 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 23 April 2013 - 12:15 AM

View PostHauser, on 22 April 2013 - 03:42 PM, said:

To clarify; how can you on one hand suggest too keep new players separated from the general population, but be against the idea of matching people based on their performance in game? It achieves the same objectives but better.


Just where in this topic did I say that I'm 'against the idea of matching people based on their performance'??
I am saying again and again that current Elo based on WINS/LOSSES has NOTHING to do with performance. Your game score, even in the state it is now, but better still when its gonna be improved, has MUCH more to do with your in game performance.

View PostHauser, on 22 April 2013 - 03:42 PM, said:

Now I'm not saying it's doing what it is suppose to. I've so far been trying explain why it can do what it is supposed to do.


And I'm in turn also saying it's not doing what it is supposed to, and that it can't in current form do what it's supposed to.

You are defending the idea of 'matching people of nearly same skill', I keep telling you that making Elo based off game score will do just that and much much better then what we have now. What do you have against it?

#95 Wispsy

    Member

  • PipPipPipPipPipPipPipPipPip
  • Talon
  • Talon
  • 2,007 posts

Posted 23 April 2013 - 12:54 AM

View PostPhoenixFire55, on 23 April 2013 - 12:15 AM, said:


Just where in this topic did I say that I'm 'against the idea of matching people based on their performance'??
I am saying again and again that current Elo based on WINS/LOSSES has NOTHING to do with performance. Your game score, even in the state it is now, but better still when its gonna be improved, has MUCH more to do with your in game performance.



And I'm in turn also saying it's not doing what it is supposed to, and that it can't in current form do what it's supposed to.

You are defending the idea of 'matching people of nearly same skill', I keep telling you that making Elo based off game score will do just that and much much better then what we have now. What do you have against it?


Actually if you perform better your team has a higher chance to win. You can literally kill all 8 enemy mechs on your own if you are good enough or they are bad enough. If you were to base it off the scoring system we see at the end of the screen then very high level of skill players would be in the same place as very low skill players due to the big flaws in the scoring system (main one being it is pretty much just your damage score, which tells very little). Unless they completely revamp the scoring system and actually make it something useful (has not been since the start) then win/loss is a far better way to go.

You keep acting like you cannot carry the team to victory, you can. You can easily carry the team to victory until you get close to your deserved Elo score. After that you will end up mostly with people of a similar skill level where individual performance does have a large impact on the team as it is only 8 people. As you start to improve and become better then those in your bracket then you will start to win more and your Elo will increase a little until you become a detriment to your team and lose more. Yes some individual games are unwinnable and some are unloseable, random is random but when you play enough (not all that much) they mostly even out and become irrelevant, a large number of "stomps" from what I have seen could have gone either way but one or two players managed to get that early advantage through skill/luck/positioning and then the opposite team crumbles to the number advantage.

Do not try to make it based off match score, damage is not a measure of skill, winning is. I have said this multiple times before but you can get 4+ solo kills and cap and still end up at/near the bottom of your team on the scoreboard. Please do not try and tell me that those other 7 guys on my team who somehow appear to have blown off every limb of the remaining 4 stationary assaults were actually all more useful then I just because they had higher scores.

#96 Hauser

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 976 posts

Posted 23 April 2013 - 01:31 AM

View PostPhoenixFire55, on 23 April 2013 - 12:07 AM, said:

How does nearly everyone winning 50% of their games contradicts with nearly everyone haveing the same Elo score?


1. New player joins the game. He is given an average rating to start with.

2. The match maker sees his (average) rating so he's put in a game with people who have a similar (average) rating.

3. The new player can't perform as well as the other (average) players so he will lose a few games.

4. As a result his elo is adjusted downward.

5. Now he is matched against people with a lower elo.

6. Repeat steps 3 to 5 until he performs just as well as every other player at his rating (e.g. winning 50% of his games when matched together).

We now have a player who does not have the same elo as every one else who's winning 50% of his games. Now suppose this new guy hangs out for a while and learns to play.

7. The match maker sees his (low) rating so he's put in a game with people who have a similar (low) rating.

8. The new player preforms better then other (low) players so he will win a few games.

9. As a result his elo is adjusted upwards.

10. Repeat steps 7 to 9 until he performs just as well as every other player at his rating (e.g. winning 50% of his games when matched together).

Again we now have a player who does not have the same elo as every one else who's winning 50% of his games.

Edited by Hauser, 23 April 2013 - 01:35 AM.


#97 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 23 April 2013 - 01:47 AM

View PostWispsy, on 23 April 2013 - 12:54 AM, said:

Do not try to make it based off match score, damage is not a measure of skill, winning is. I have said this multiple times before but you can get 4+ solo kills and cap and still end up at/near the bottom of your team on the scoreboard. Please do not try and tell me that those other 7 guys on my team who somehow appear to have blown off every limb of the remaining 4 stationary assaults were actually all more useful then I just because they had higher scores.


View PostPhoenixFire55, on 21 April 2013 - 12:20 AM, said:

Never said that the current match score is perfect. PGI will have to keep improving it. IMO you need to include factors like 'damage taken', 'distance travelled', 'locks aquired/kept', 'time on hostile cap zones', 'time on your cap zones when enemy mechs are on them' etc. But 99% of matches I've played people who really did well had higher game scores.


Also, I don't know why you of all people worry bout being placed bottom of the team. Each time I've been in game with you you've been top or near top of your team in a jenner, dragon, any mech really. As I said, even with the current very imperfect game score good players end up with good game scores almost all the time, regardless of what mechs they use.

I still think that 1 man can't carry his team, I've never seen that happen. If smb claims to have 8 kills in a game then it only means his teammates contributed greatly by stripping off enemy armor, taking fire etc. One person can't swing a balance in a non-close matchup, much less so he'll be able to do it in 12 vs 12. And if its a close matchup you can't say that anyone is carrying the team, because in close matchups everyone contributes. It might happen so that in the end somebody will decide the game, but it doesn't mean he carried his team.

#98 Hauser

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 976 posts

Posted 23 April 2013 - 01:48 AM

View PostPhoenixFire55, on 23 April 2013 - 12:15 AM, said:

Just where in this topic did I say that I'm 'against the idea of matching people based on their performance'??
I am saying again and again that current Elo based on WINS/LOSSES has NOTHING to do with performance. Your game score, even in the state it is now, but better still when its gonna be improved, has MUCH more to do with your in game performance.


I'm getting the impression you are against matching people based on this:

Quote


In other words noobs play with noobs and learn nothing. Noob friendly game, you are gonna win 50% of your games even if you are a complete *****? Is that what we need? Don't know about you but I'd rather see a game that motivates people to improve their skill to win more games. Right now there is more reason to 'noobify' yourself to play in easy mode vs non-so-good players.



Now please go back to my post to Christoff. It explains why good players will win more games.

View PostHauser, on 20 April 2013 - 11:41 AM, said:

Now I get the feeling you're not familiar with statistics. You keep on mentioning rather specific examples that work towards your argument but you don't take into account that with random match making those examples can go either way. The better players can on either team. So I'll reduce this to a very small and simple example.

Imagine a game of tug of war. Each players skill in game can be represented by a number, the higher the number the better they are. This number is the players inherent aptitude at playing tug of war, you can not measure it, it is a hidden value. I do expose them however so I can simulate the game.

We have 3 players, A, B, C and D. They have a power of 1,2,2 and 3 respectively. You can add up these powers to see how strong a team is. The stronger team will win. If the scores are equal it's a 50/50 chance either way.

Given a random matchmaker the possible teams are:

AB | CD = 3 vs 5
BC | AD = 4 vs 4
AC | BD = 3 vs 5

Now if you look at all possible outcomes we get this:

A has a 16.6% chance of winning.
B has a 50% chance of winning.
C has a 50% chance of winning.
D has a 82.6% chance of winning.

B and C are both average players and as such have an average chance of winning and losing. Their elo won't change much.

A and D are a different story though. D is usually carrying the team, unless he is paired with A. A on the other hand is poor player and while he might occasionally be carried by D, he'll still lose most of his games.

As such when using elo, D increase in rank while A goes down. Atleast until the point where their scores are so extreme that winning from A or losing against D yields no change in elo.

Now keep in mind that while the BC | AD match gave both teams a 50% chance of winning, the match maker is not averaging out elo. Rather it searches for people that match a specific rating, which shows you need a good population for elo to work. As such this population is too small to provide good matches, but I'm trying to keep this example simple.

So in summary, if you consistently do the majority of the work for your team (like D does), then you will go up. Perhaps not in every match, but on average you will go up.


#99 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 23 April 2013 - 01:53 AM

View PostHauser, on 23 April 2013 - 01:31 AM, said:

1. New player joins the game. He is given an average rating to start with.

2. The match maker sees his (average) rating so he's put in a game with people who have a similar (average) rating.

3. The new player can't perform as well as the other (average) players so he will lose a few games.

4. As a result his elo is adjusted downward...


3 and 4 are false. It doesn't matter just how good/bad he performs. There is a high chance that the other team has the same new player. They both have no affect on the game outcome whatsoever. Chances of a completely new player being placed on a good team or a bad team are same. Thus chances of a completely new player to win/lose are the same. He'll be winning 50% of his game right from the start. His Elo won't change at all unless he does one of the following:

- joins a quality group that will win way more games then they lose (because yes, 4 players on comms CAN swing a game into their favor) - then and only then his Elo will go up
- starts to afk, TK etc. - then and only then his Elo will go down

Main problem that this works not just for new players, but pretty much for all players.

#100 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 23 April 2013 - 02:04 AM

View PostHauser, on 23 April 2013 - 01:48 AM, said:

I'm getting the impression you are against matching people based on this:


I won't deny that I am in general against it. But this is irrelevant in this topic, I'm trying to work with what we have. We have Elo and PGI will most surely stick with it just like they stick with most of their questionable decisions. There are far worse things then Elo that we have. Thing in question here is matchmaking, my suggestions are towards improving the matchmaking in terms of really matching players by skill. I've just had two 8:0 roflmao stomps in a row, just like I used to have in closed beta. I can't call it quality matchmaking.

View PostHauser, on 23 April 2013 - 01:48 AM, said:

Now please go back to my post to Christoff. It explains why good players will win more games.


Your example is good and true, but irrelevant to what we discuss. I will fully agree with you that 1 person can be a deciding factor in 1 vs 1 games, 2 vs 2 games, probably even in 3 vs 3 games, but it is not so in 8 vs 8 and surely not so in 12 vs 12. This 'margin', 'difference' he makes is becoming less and less the more people you have in your teams. You can calculate this yourself if you extend your example into 8 vs 8 and 12 vs 12. Chances of a good player winning will be like 52-54%, chances of a bad player winning 46-48%. This is exactly what we see on graphs, 90% of people are there in between 46-54% wins, only people outside these boundaries are people doing the two things I've mentioned in my reply to Wispsy just above...





5 user(s) are reading this topic

0 members, 5 guests, 0 anonymous users