435 replies to this topic

[Kmieciu]

Rhalgaln, on 30 September 2015 - 12:56 AM, said:

I like the suggestions on dynamic Battle Values.

So if all mechs have an abstract Value based upon their successful usage in historic games, you can easyily balance pug games by a restricting (battlevalue+pilotELO) per side.
This allows lower skill pilots to take higher battle value mechs and forces big groups of high tier players to use less meta mechs.

If the BV is recalculated weekly you might break the domination of so called meta-mechs
The default BV can simply be based on tonnage, equippment(ECM, MASC), quirks or maybe even on loadout different from mech-config to mech-config. The more mechs of a given loadout exist, the higher will be the BV.
Maths should limit to a maximum which is abaout the maximum score of reaching Tier 1.

Then limit Team BV to 8 times that value (BV+ Skill) per Side for default.
For each group size limit BV to 80% of that score per player in group.

Match Teams with equal to 10% different BV's

Sorry for OT, but we had tonnage restrictions already and they failed

Battle Value should be calculated using global win ratio of a mech variant. Because winning is the ultimate goal. Damage and kills are less important.
Battle Value should be kept secret, because exactly which mech gives the best chance of winning would make even more people use that mech.

Edited by Kmieciu, 30 September 2015 - 01:58 AM.

[TyphonCh]

I'm going to elaborate a bit on what I quoted Kiiyor on page 5 after mulling it over...
Here's a personal experience from a casual player in a casual unit for you to chew on Russ. 9 times out of 10 when I'm dropping in the group Q with a friend or two (which is still rare, I'm more of a pugger but a couple friends and I recently started a unit so that may change) we drop in similarly tonnage mechs, either to complement each other or to maintain the same speed to stick together.
Example, when there are 2 or 3 of us. We all own the ecm variant Commando. We painted them white and red, put in the biggest engine, 3 streak 2's and a medium laser. We call them the Canadian Commandos. It's a ton of fun for us to play like that.... honestly if not being able to drop in any mech I choose just in a 2 man group?? I wouldn't play group Q anymore. 2 man's have a hard enough time as it is

The new tonnage restrictions you impose would eliminate smaller groups dropping in all assaults or all lights. Which you just said yourself, groups of 2 and 3 comprise %80 of your group Queue. Why should they have restrictions? They can drop in this odd combination of assault/light, heavy/heavy, heavy/medium, etc. But why not light/light or assault/assault?

I can't imagine a group of 2 having a massive impact on a game even if they are running two mechs at the high end of the spectrum. I would suggest giving them full range for tonnage, 40 - 200

A group of 3 maybe more so. I'd give them stricter limitations of possibly between 100 - 280. 3 arctic cheetahs or 3 direwolves dropping together is more of a game decider than 2

As for the rest of the numbers, they're in a good spot I feel. The whole concept seems to be well thought out and I hope it works! Group Queue needs help!

EDIT: It's the middle of the night and I'm at work so I haven't thought out every possible kink of what could happen if say 6 groups of 2 drop with all lights, all assaults, etc and you get a totally lop sided game. I'm just talking from my own personal experience.

If you could somehow loosely weave 3/3/3/3 into the equation to stick two or three man groups into a more appropriate game, I could see it working. Reading back on some of the posts now, a concern is massive boating of powerful mechs that fit into the tonnage window. I still feel like 3/3/3/3 needs to be enforced in some way

Edited by Team Chevy86, 30 September 2015 - 02:52 AM.

[Karl Streiger]

Kmieciu, on 30 September 2015 - 01:56 AM, said:

Battle Value should be calculated using global win ratio of a mech variant. Because winning is the ultimate goal. Damage and kills are less important.
Battle Value should be kept secret, because exactly which mech gives the best chance of winning would make even more people use that mech.

This could normalize the global win ratio.... but shouldn't this global win ratio be linked with the number of Mechs in usage?
For example if everybody uses the UrbanMech- its obvious that its global win ratio would be 50/50 - although not a single other Mech is in usage because the UrbanMechis so "great".

Anyhow - without the ability to balance the Mechs on a tonage based system (what should be working when you really want to use a tonnage based MM)MWO really really need a more abstract system - and this dynamic global BV system is exactly what could do the job - without complex calculations that could be exploited (like the BV in TT)

Edited by Karl Streiger, 30 September 2015 - 02:31 AM.

[DeRazer]

A lot of the complaints about this seem to be around the "abuse" of tonnage by large groups boating meta mechs.

My solution - Variety Tax which penalises groups using too many of the same mech (variants count as same [necessary due to clan Omnimechs but could potentially be waived for IS]) but has no impact on the matchmaker ability at all as it comes into play AFTER the battle.

This is NOTHING to do with Lore but purely a system to reward variety in play.

Groups 2-3 No penalty.
Groups 4-6 Minimum of 3 different mechs to avoid tax
Groups 7-10 Minimum of 4 different mechs to avoid tax
Group 12. Minimum 5 different mechs to avoid tax.

Each group CAN still field whatever they want within tonnage limits BUT for every Mech BELOW the difference limit the whole group lose 15% Cbills and XP.

Examples:
Group of 3 - do whatever they want. Good for new starters, good for casual group drops.
Group of 6 - If they field 3 different mechs all is good, but if they do just 2 then they lose 15%.
Group of 12 - if they really did do just 6 ACH and 6 DWF they would be 3 under the variety limit and be losing 45% of CBills and XP.

You could even ENCOURAGE variety by rewarding large groups who used variety of mechs - 2.5% BONUS for every mech ABOVE the min limit. A 12 man group using 12 different mechs (remember - variants count as same mech) would be getting 15% Cbills/XP Bonus.

[TexAce]

I'd say maximum for 12mans should be 720 tons.

Remove the Minimum tonnage completely. Seriously, why do you hate lights so much? As if they were any threat at all in current gameplay/meta.

Or at least remove it for any group smaller than 6-man.

[Leopardo]

dude good idea but ppl play mostly for win not for farm i think. so for moment it will works....but not for long.
but still good to have different mechs in drops.

Edited by Leopardo, 30 September 2015 - 03:21 AM.

[Tyrandia]

My main question is: Why don't we allow groups of 2 to queue in with the solo queue drops? It'd be similar to duo-queuing with League of Legends.

Just my two cents.

[Chemie]

you won't see many assaults in the group queue with this type of restrictions.

[TyphonCh]

Tyrandia, on 30 September 2015 - 03:40 AM, said:

My main question is: Why don't we allow groups of 2 to queue in with the solo queue drops? It'd be similar to duo-queuing with League of Legends.

Just my two cents.

Kicking a dead horse. That idea has been juggled around along side solo's in the group Q... It's one or the other unfortunately. I have ideas for solo's in the group Q but that's a different topic and I hope it gets more attention

[Naduk]

DeRazer, on 30 September 2015 - 02:50 AM, said:

A lot of the complaints about this seem to be around the "abuse" of tonnage by large groups boating meta mechs.

My solution - Variety Tax which penalises groups using too many of the same mech (variants count as same [necessary due to clan Omnimechs but could potentially be waived for IS]) but has no impact on the matchmaker ability at all as it comes into play AFTER the battle.

This is NOTHING to do with Lore but purely a system to reward variety in play.

Groups 2-3 No penalty.
Groups 4-6 Minimum of 3 different mechs to avoid tax
Groups 7-10 Minimum of 4 different mechs to avoid tax
Group 12. Minimum 5 different mechs to avoid tax.

Each group CAN still field whatever they want within tonnage limits BUT for every Mech BELOW the difference limit the whole group lose 15% Cbills and XP.

Examples:
Group of 3 - do whatever they want. Good for new starters, good for casual group drops.
Group of 6 - If they field 3 different mechs all is good, but if they do just 2 then they lose 15%.
Group of 12 - if they really did do just 6 ACH and 6 DWF they would be 3 under the variety limit and be losing 45% of CBills and XP.

You could even ENCOURAGE variety by rewarding large groups who used variety of mechs - 2.5% BONUS for every mech ABOVE the min limit. A 12 man group using 12 different mechs (remember - variants count as same mech) would be getting 15% Cbills/XP Bonus.

there ya go thats exactly the kind of thing i was getting at in my prev post
i like it and/or something similar

[Darwins Dog]

Just repeat what others have said:

Yes to the idea, if it will help get better matches.

Remove or greatly relax restrictions on small groups so that two friends can run two assaults or lights.

Bring in class restrictions for larger groups. 12 TDRs will become the new standard for 12 man teams with this system.

[Widowmaker1981]

Russ Bullock, on 29 September 2015 - 01:32 PM, said:

Thanks for the feedback thus far

Russ, i have a question regarding matchmaking that i would very much like answered:

What do you envisage the MM doing? Do you simply want to segregate brand new players from experienced players, or do you want to create matches actually balanced by Pilot skill? The current PSR system does the former (to an extent) but does not do the latter, at all, due to the enormous weight given to winning or losing (impossible to lose PSR in a win, trivial to gain PSR in a win). This works out ok if there is no such thing as a group queue, but since there is, good units that win the majority of their games will see every single member rise to T1 in short order, basically regardless of the individual skill of the player (better players will rise faster, but all players will rise). It is foolish to think that every person has the same skill level ceiling, they do not.

I would suggest as a means to fix this that you change the PSR algorithm for the group queue (is it possible to separate it?) such that the chance to gain/lose PSR is totally independent of win/loss and is only based on individual match score (win / loss is already accounted for in match score as when you win you have more chances for assists/dmg). That will mean people being carried by good groups but performing consistently badly will lose PSR rating as they should, not gain.

The tonnage limits look reasonable, given what you are trying to achieve. I think however what you are trying to achieve is pointless, because PSR doesnt work for the group queue as currently constituted.


Incidentally, though i see what you are going for, the proposed changes will likely kill this game for me, because a) I only really like running assaults and heavies in the group queue and i dont really like running in large groups with my unit because it gets boring very rapidly, and impossible to get those cool 1k dmg+ games. Its going to be impossible to run 2+ assaults in a 4 man, and thats really all i like doing in the group queue. I recognize im a special case here though and dont expect to be catered to. Its just a shame ive dumped a lot of money on this game and will likely be forced out by these changes.

To be quite honest, id rather you just capped group size at 6

Edited by Widowmaker1981, 30 September 2015 - 04:10 AM.

[Markoxford]

There is a lot of over complication in this thread and playing with minor figures that one minority likes, but another doesn't. It seems to me that the issue is 1) reduce MM time and 2) allow for flexible group sizes that have a sense of being balanced, ( mostly by penalizing the larger groups). Here is at least an idea of how to proceed ;

1) Limit groups to even numbers. MM takes less time and we get away from everything for everyone syndrome.

2) Remove PSR balancing all together in groups - it's groups, PSR doesn't really matter that much I would guess and a small group of tier 5 could end up learning a thing or two from a large group of 1s or 2s.

3) Keep 3/3/3/3 to keep some variation to the mechs that the larger groups brings.

Sorry, I know weight limits were key to earlier MW multiplayer games, but we had lobbies and players could set up the game however they liked to try to get others on their computers to play, (and it wasn't that great to be honest). MWO is not the same - we don't have a single player or co-op to take the edge off the online action, we have client/server infrastructure with 1000s of players in limited modes. Keep and refine PSR for single players, promote groups as the next step in players development and CW as the community game mode it needs to be. Add official ladders for 4, 8 and 12 man groups so people can see when they are stomped by a higher level team and gain from it by trying to play better.

[Sardauker Legion]

Small groups usually go same speed. The miminum limit does not allow scout lances.

The ton system creates problems for who owns not enough modules for all mechs.
Every incoming - leaving player in group forces me to change mech, so i loose more time in Hangar.

[Nightshade24]

BanditB17, on 29 September 2015 - 11:32 AM, said:

Group size 11? Does this mean group and solo queues will be combined?

it was an oversight by using the formula.


Kinda like the same way how you can estimate how much an Elemental battle armour (1 ton) would cost if it was a hero would be 75 MC... will never happen since battle armour will never be player controlled if added but at least we know what a 1 ton 'mech' would cost as a hero!

[kka]

I have a different approach as suggestion:

Only allow groups of 4, 8 or 12

Then come up with tonnage / class restrictions.

I think this is doable by players and would speed up the mm process a lot.

[JHackworth]

I'll echo Widowmaker here and reiterate a similar point/suggestion I have made in the past. Tonnage is a poor reflection of actual competitive value in this game but we currently don't have any way to account for this. There is no in-game incentive to take the less capable mech for any tonnage and as such any MM effort to try to balance based on tonnage with run into problems of mismatches. 3/3/3/3 mitigated this somewhat but it just pushed the same problem out to each class.

We keep trying to make every mech 'balanced' against every other mech--and it's not really helping the game. Everything is quirked so heavily now that quirks are driving mech choices in the place of more nuanced strategic thinking or some sort of game economics. There is nothing inherently wrong with having mechs that out perform their others. The problem is that there is no economic (broadly speaking) offset to bringing meta or OP mechs--there is no concept of 'value' and 'cost'.

One solution is to scrap group queue altogether and create an alternate lobby system where groups can see existing challenges, issue them or accept existing ones. If a 'buyer' and 'seller' match within certain tolerance, MM kicks in an automatically completes that transaction. Make this based on battle value instead of tonnage or numbers. In terms of BV, there are lot's of ways to get there and many of us have discussed them so i won't go over it here. This will allow for all kinds of different match ups, small skirmishes, full lances vs full lances, full companies vs full companies give or take. You want to try 4xAtlas DD-Cs worth 10,000 BV Fine, let's see if there is another group that will step up and field a team worth 10,000 BV.

Hell, if you really wanted to, you could factor in PSR so that BV is modified by pilot ratings. 4 x TBRs driven by 4 x PSR 1 pilots might be worth 10,000 BV--a fair match up against 4 Tier4 D-DCs.

The bottom line is that if you take the onus of creating a good match and put it on the players and instead of an algorithm, you'd run a better chance creating a perception that things are fair. At least people feel like they control their own destiny to some degree.

Another Solution, a less radical one keeping in the current 12v12 straight jacket, is to express this BV idea with a proxy mechanic such as "Operating Cost". Bring an Artic Cheetah, but since it is a high BV mech, you take cbill hit or a PSR points hit at the end of the match. Make that visible like a refit cost. Bring a crappy Commando and you get a decent bonus for a win because of very low operating costs. Again, throw in a PSR modifier and you are good to go.

2cents dropped

[Cricket504]

Seeing many of the responses posted, there are 2 very clear and common concerns.

1) 2 and 3 man groups can't drop with mech selections that would be enjoyable for many small group players. At the 2 and 3 mech groups, drop tonnage restrictions. Even with the numbers listed now heavies are going to dominate. So why put on the farce of tonnage restrictions. Even now you see the heavy queue is at the highest % any time there is not a new non-heavy.

2) 8+man teams have a serious advantage, hard lock a 1 group differential (12 man can only face a max of 2 teams on the other side). IF ~80% of the group queue is 2 and 3 man, then that is what the MM should be best set up to support. Large teams have CW so give the group queue to lean to the smaller groups.

[Daitarn4]

• it is impossible to try to balance the ability of the pilots or their organization.
  
• is unfair (three players or less) give tonnage limitations.

The solutions are to be found in the game and not changing the rules for players every month:

• Mech balance trying to make it more or less on the same level. No meta MEch!
  
• Create a system of 1v1 games - 4vs4 - 12vs12 system ranked to attract the best players.
  
• Machmaking must be weak against weak or strong against strong pilots. No other weird solution PGI..
  
• The growth of a pilot must be progressive, mixing weak and strong is useless and does not teach anything.

All these changes instead try to solve problems quickly by imposing constraints and flattening the game further.

Edited by Daitarn4, 30 September 2015 - 05:34 AM.

[Locabiosol]

I haven't read all... So just a few thoughts.
While I think a tonnage restriction is better than capping group sizes to 4 or doing 1/1/1/1 matchmaking to release some pressure from MM, I think that the limits are not well set.
I mean you can do a 12 man made of 6 ACH and 6 DWF with this limits. Seriously...please no. Why not adding the current 3/3/3/3 rule to the tonnage limits to prevent this kind of cheese? The possibility of 6+ assaults is just crazy.
The other thing is the limitation for small groups. Some people just like going with two lights or two assaults. Why not? There is nothing scary about it. I mean if you fear two locust for what ever reason, than please go atleast to 70 tons min, so you can bring two lights and max 170 for two 85 tonners, if you fear two DWF. I won't even bother seeing two friends run two DWF, but I can understand why people don't like it.