117 replies to this topic

[Hauser]

DeaconW, on 10 August 2013 - 07:46 AM, said:

I was here with weight class matching and respectfully disagree. I think the MM should do both...try to match weigh and ELO (if they are going to keep ELO) in some type of formula that is weighted (no pun intended) toward weight, not ELO. It should also do ELO matching by range, not average. At least, team average matching is what it appears to be doing.

They're already using weight class. They are using the actual weight of the mech. This is better for the lighter mechs in their weight class. An Atlas and an Awesome are not a fair match.

http://mwomercs.com/...-making-update/

Quote

Is the match maker taking weight class into account?

Yes the version currently live on production does take weight class into account. The way the match maker works it tries to group players within certain thresholds and grows those thresholds over time to ensure a match is made within currently a 2 minute time period.

These curves will likely see some tuning for both skill and weight as we monitor the data coming in and work to ensure more and more matches are created within good thresholds. This tuning will happen over the course of upcoming patches.

As for the average Elo, they are not doing that. It starts with a value and then expands the range around that value over time. It might appear to be average because the match maker is expanding the range and could be dumping less apropriate people into a match.

Quote

How does the match maker compose a teams Elo rating, is it average rating or closest to a target?

It's closest to a target value, so the match maker starts trying to make a match for an Elo of say 1300 and will pull in players to those teams closest to those values; however, as mentioned earlier within growing thresholds and those curves will be tuned. Currently it may be a bit 'sloppy' about how it's filling those buckets but over time it will be tuned to be much more precise.

We need to do this carefully over time as generally the cost of precision is time to find a match we want to slowly find a very nice balance between time to find a match and the number of matches that are correctly composed.

[Nicholas Carlyle]

Why does everyone feel like they have to explain everything to the Nth degree.

The matchmaker sucks. We all see the tonnage differences and major steamrolls consistently.

Just fix it

Edited by Nicholas Carlyle, 10 August 2013 - 08:22 AM.

[Hauser]

DeaconW, on 10 August 2013 - 08:15 AM, said:

That is not the point...what most players care about is the majority of the matches being evenly matched, challenging, and most of all, fun.

This. This is important.

I'm commenting allot on the match maker because people don't seem to understand how it works and are complaining about the wrong thing because of it.

For me allot games aren't fun when people are selfishly sitting behind cover not doing much. The games that are fun happen to be the games that are either all out brawls or games with good manoeuvring on both teams. Attempting either without the whole team being in on it is liable to get you killed.

I really don't know how to fix that.

Nicholas Carlyle, on 10 August 2013 - 08:21 AM, said:

Why does everyone feel like they have to explain everything to the Nth degree.

The matchmaker sucks. We all see the tonnage differences and major steamrolls consistently.

Just fix it

Because neither statement is true. Apparently yes. Yet when you look at more closely it turns out the tonnage difference is small and matters little. Comparable teams will steamroll each other either way in clan matches. It is the nature of the game, not directly an indication of in balance.

Edited by Hauser, 10 August 2013 - 08:37 AM.

[DeaconW]

Hauser, on 10 August 2013 - 08:20 AM, said:

They're already using weight class. They are using the actual weight of the mech. This is better for the lighter mechs in their weight class. An Atlas and an Awesome are not a fair match.

http://mwomercs.com/...-making-update/


As for the average Elo, they are not doing that. It starts with a value and then expands the range around that value over time. It might appear to be average because the match maker is expanding the range and could be dumping less apropriate people into a match.

Thx for the link to weight class being a part of it now. Wonder how much because I still see a lot of imbalance in matches.

And you have hit on another problem with ELO-based matching...since the MM will expand to put everyone of any skill together to simply get a match, the idea of "skill-matched" gaming is simply bogus in MWO.

[Ubikjam]

I'm not sure how this is supposed to be PGI's fault, sometimes a series ofpeople have a bad game and you get stomped.

12v12 has the benefit of making tactics more important than before, so individual skill is less of a crutch for a pugging team. Problem is noone uses any tactics at the moment - clearly encouraging better team play is on the agenda with CW.

[Leafia Barrett]

Karenai, on 10 August 2013 - 02:49 AM, said:

I have the feeling they reset ELO from time to time. Maybe not for everyone, but sometimes it feels that way. Getting thrown together with people in trial mechs, who run into walls the whole time (I mean not 1 or 2 but sometime 4-6 on your own team), feels like ELO reset.

The longer the wait, the looser the ELO restriction becomes. If you're waiting for a solid minute or two, you can probably expect a pretty unbalanced match.

[DeaconW]

Ubikjam, on 10 August 2013 - 08:37 AM, said:

I'm not sure how this is supposed to be PGI's fault, sometimes a series of people have a bad game and you get stomped.

Because PGI built and is advertising their MM as making balanced matches based on ELO, and IMO it does not better than rolling a die or straight up weight-matching did. If they scrapped ELO tomorrow I'd stop complaining at all if I got ROFLstomped because I'd agree with you. It is the *claim* that the majority of the games are balanced when, for many of us, they are not, that I take issue with. This is in addition to my general issue with them using ELO at all.

If you really want to think about what is wrong with using ELO for MWO, think about this: ELO was designed for rating individual players in a one-on-one game with a fixed board and fixed piece capabilities. It is being used by PGI on a multiplayer game with a constantly changing "board"(different maps) and constantly changing "pieces"(different mechs, different weapons on mechs). Just doesn't sound right when you look at it that way, does it?

[Hauser]

DeaconW, on 10 August 2013 - 08:34 AM, said:

Thx for the link to weight class being a part of it now. Wonder how much because I still see a lot of imbalance in matches.

And you have hit on another problem with ELO-based matching...since the MM will expand to put everyone of any skill together to simply get a match, the idea of "skill-matched" gaming is simply bogus in MWO.

You're welcome!

I think you should be careful by estimating weight at a glance. It's easy to see the four Atlai on one team but miss the four Centurions that add the extra weight on the other. Hammerborn did a nice survey of that a while ago and it turns out the difference is small.

And yeah. It feels as if the match maker can't find enough players of one sort. It is a process so while it starts expanding it will already have assembled 8 guys that are a really good match, another 8 that are a bit off, and finally the last 8 that are far out either way. That would be fairly consistent with my experience, though I would also like to see the telemetry PGI has on this.

Edited by Hauser, 10 August 2013 - 08:56 AM.

[DeaconW]

Leafia Barrett, on 10 August 2013 - 08:49 AM, said:

The longer the wait, the looser the ELO restriction becomes. If you're waiting for a solid minute or two, you can probably expect a pretty unbalanced match.

Agreed...but it is important to understand that the inverse is NOT true...you may get a match right away because you were the last person needed to fill that unbalanced match that started matching 3 minutes ago...

[Hauser]

You know, I'm wondering. Are matches against your equals fun? They're certainly fair, but is a fair game fun?

Wouldn't matches against people that have 90% of your skill be more fun? You'd have a moderate challenge and then you'd win.

Can a game even be fun when you're only winning 50% of the time?

What makes a match fun anyway?

Edited by Hauser, 10 August 2013 - 08:56 AM.

[Odins Fist]

xenoglyph, on 10 August 2013 - 01:28 AM, said:

They got top men on it. top. men.

Raiders of the Lost Ark quote ^

Bravo..

[CeeKay Boques]

Weight balance has very little to do with it...

Because many "Steamrolls" involve crazy *** lights in a group cleaning clocks.

It is absolutely the fact of more firepower on less targets is a slope that is hard to recover from. If you are down, you should be panicking, and finding great places of terrain to fight from, because its not looking good. I've had just as many 12-8s as I have 11 to 2s.

[DeaconW]

Hauser, on 10 August 2013 - 08:55 AM, said:

You know, I'm wondering. Are matches against your equals fun? They're certainly fair, but is a fair game fun?

All depends on how narcissistic you are, I suppose.

I, for one, like hard fought, to the last man, battles against skilled players much better than being on the winning side of a ROFLstomp. It is clear, however, that the online world is populated by a lot of people who have no problem getting their "fun" at the expense of others. Those are the guys who post "gg close" after a ROLFstomp, for example. There are others of similar, but darker ilk who will cheat on games (something I have never understood). But the truth is that wins that are accomplished by actively cheating or simply being given to you by the MM aren't "real" wins. Just another reason we need a real lobby where competitive players can set up their own matches with their own parameters.

[Screech]

Matches have seemed a bit more balanced with 12v12 then previously with 8v8. The matchmaker does what it can but honestly it is limited due to the small pool it draws from and imbalances between the mechs, weapons and most importantly the people themselves.

It is a poor craftsman who insults his tools.

Oh, 100% pugging here.

Edited by Screech, 10 August 2013 - 09:33 AM.

[Mystere]

Appogee, on 10 August 2013 - 01:19 AM, said:

I 95% PUG, and I see this kind of match very frequently. Do any other (solo) players experience this...?

I'm a 100% solo player and I do not see the problem you are seeing. You must currently be in Elo hell!

[Ubikjam]

DeaconW, on 10 August 2013 - 08:49 AM, said:

Because PGI built and is advertising their MM as making balanced matches based on ELO, and IMO it does not better than rolling a die or straight up weight-matching did. If they scrapped ELO tomorrow I'd stop complaining at all if I got ROFLstomped because I'd agree with you. It is the *claim* that the majority of the games are balanced when, for many of us, they are not, that I take issue with. This is in addition to my general issue with them using ELO at all.



If you really want to think about what is wrong with using ELO for MWO, think about this: ELO was designed for rating individual players in a one-on-one game with a fixed board and fixed piece capabilities. It is being used by PGI on a multiplayer game with a constantly changing "board"(different maps) and constantly changing "pieces"(different mechs, different weapons on mechs). Just doesn't sound right when you look at it that way, does it?

So the fact that they are trying to prevent stomps is what upsets you?

Edited by Ubikjam, 10 August 2013 - 09:30 AM.

[MWHawke]

There is no best way. There is only the PGI way.

Deal with it.

[DeaconW]

Ubikjam, on 10 August 2013 - 09:29 AM, said:

So the fact that they are trying to prevent stomps is what upsets you?

Nice try. No, that is not what I am saying. I am saying that:

1. Their method of trying to balance matches to prevent stomps doesn't work to actually balance matches (or ROFL stomps would be very rare.)

and

2. Their advertising it as if it does balance matches is therefore disingenuous.

Honestly, I'd probably be at peace with it if they just stopped doing #2 (ooh...a pun is so waiting there...)

Another factor I forgot to mention: ELO can't take into account is whether I am playing in a 4 or 8 man or playing solo. Those styles of play and results are very different.

ELO could probably come closest to "working" in a solo only, weight class matched que. Otherwise, I question how much better it is than a random matchmaker...

MWHawke, on 10 August 2013 - 09:39 AM, said:

There is no best way. There is only the PGI way.

Deal with it.

Same answer for heat scaling as well...at least they are consistent?

[Urdnot Mau]

@ OP

You had more assaults. You had more fast mechs. So, what is wrong about that match?

1- Bad players are being carried by good players. Assault mechs doing less than 200 damage? Light mechs outdamaging assault mechs?? Do i really need to say that there's something wrong about it?
Bad players will have a hard time getting better, since they are not the ones leading matches and getting crushed by other good players. Bad players should play with bad players and good players should play with good players.

2- They had more players. Bad players get in a bad position and they die quickly before dealing much damage, leaving you severely out-numbered.

so, 1: you had more bad players; 2: you had less team mates, wich could be either an asset or a distraction. 1 Less bad players does not make a difference, but 1 less good player makes all the difference.

I solo a lot and i get in matches with team mates l ike these and it makes me very frustrated...

[Jman5]

Hauser, on 10 August 2013 - 08:51 AM, said:

You're welcome!

I think you should be careful by estimating weight at a glance. It's easy to see the four Atlai on one team but miss the four Centurions that add the extra weight on the other. Hammerborn did a nice survey of that a while ago and it turns out the difference is small.

And yeah. It feels as if the match maker can't find enough players of one sort. It is a process so while it starts expanding it will already have assembled 8 guys that are a really good match, another 8 that are a bit off, and finally the last 8 that are far out either way. That would be fairly consistent with my experience, though I would also like to see the telemetry PGI has on this.

I've noticed that as well. Whenever I organize the assaults into their own lances I see an uptick in people complaining about weight imbalance/matchmaker.

I do a lot of end of round screenshots to compare the weight and it's generally pretty close. That said it has gotten slightly worse since 12v12 was introduced, but the average percentage difference is probably identical.