14 replies to this topic

[Greyfyl]

My suggestion for a more interesting/balanced approach to individual matches:

First of all it would require a slight change in how you pick 'ready' mechs in the mechlab. You would still start with 4, but you must have one of each weight class. I would suggest the ability to buy additional 'ready' slots with c-bills/mc, maybe up to a maximum of 6. Any purchased ready slots can be filled with any weight mech. See images below.








Once you have the minimum ready slots filled (they can be filled with trial mechs if necessary) with at least one from each weight class you can hit the launch button. Once you hit the launch button the matchmaker will try to find an ELO match for you personally (this would be for puggers) and put the two of you on opposing teams. Once each team is full you will go to the new drop preparation screen. This will show you only the players on your team.



At this point you will know the map and the drop configuration (which will be random). You will also at this point be placed in a numerical order from 1-8 with your other teammates. This should be a weighted system that tends to put the more experieced players towards the top, but can allow all players a chance at being #1. With all players in the drop prep screen you can use team chat to discuss what mechs to bring, and after 20 seconds or so the first player will be able to choose his mech from the mechs available at the top of the screen.



Each player will pick his mech as it becomes his turn. I would also think that a timer would be needed at this point to keep anyone from holding up the entire team. Once everyone has selected their mechs and readied up on both teams- the game launches.

Edited by Greyfyl, 06 March 2013 - 11:04 AM.

[Redshift2k5]

I don't know about making players wait to go one at a time. Let us all pick, and during a 30 second countdown we can change our minds. Ie, make it more like LoL's drop lobby where you are picking your character within as certain time and have the opportunity to chat about your picks.

[Greyfyl]

TY for the feedback, you may be right about not having to wait but If there is no order, it will turn into a mad rush to pick the assault. People will end up being pissed-off because someone beat them to the mech they wanted.

You would be given time to chat about your picks prior to the first player choosing his mech, and throughout the whole process.

Edited by Greyfyl, 06 March 2013 - 11:21 AM.

[Deathlike]

Anything that actually requires you to pick out a mech for the match like that, in non-league (private) matches, would never be approved.

It's a logistics problem, and MM is supposed to be the "solution" (a constantly tweaked one at that, which isn't the case quite yet). You're supposed to pick the mech you want to drop with, and that's about it. Unless you're dropping into a specific area (specific map theme or actual map) in clan warfare, it's not happening for PUGs.

[Greyfyl]

Deathlike, on 08 March 2013 - 08:10 PM, said:

Anything that actually requires you to pick out a mech for the match like that, in non-league (private) matches, would never be approved.

It's a logistics problem, and MM is supposed to be the "solution" (a constantly tweaked one at that, which isn't the case quite yet). You're supposed to pick the mech you want to drop with, and that's about it. Unless you're dropping into a specific area (specific map theme or actual map) in clan warfare, it's not happening for PUGs.

Just curious, but why would it never be approved? You still have to have 4 ready to go mechs - in a way it would promote grinding our more mechs and ready bays. I would think that would be a positive from pgi's pov. The current MM has many issues that will never be resolved with just elo. It needs help.

And what in the world does it's a logistics problem mean? I don't think we really have to head out a re-arm all the mechs in the field after each match, so are you talking about server load, bandwidth, or what?

Edited by Greyfyl, 08 March 2013 - 09:56 PM.

[Jonathan Paine]

I really like this suggestion. Would need some tweaking to work with the elo system per weight class.

[Deathlike]

Waiting for people to select their mechs brings me back to MW3+MW4 where these things were far worse than it is now. In your scenario, you'd have 20 seconds to pick mechs/discuss, whatever. That comes to about 140 seconds (7 people picking ahead), which is TWO WHOLE MINUTES (even a failed MM search is shorter than this). You're adding this ON TOP of MM, which doesn't make this easier or better.

Even if you adjusted the time for picking, I doubt this would actually go over well... despite the demand for better communication, this won't always go over well with people. It's worse if you force the new player into a trial mech (as if trial mechs were barely competent with custom mechs).

This also causes a greater handicap with strictly F2P players that won't have the desirable mechbay options. You would have to at least reconsider reorder who goes first (least experience first) and some would consider this a bad idea... especially if the mech they bring is woefully inadequate, wasting an particular mech slot that someone else would be more able to fill. (Note that since you have more control over this, it can and will drive people crazy. With the current randomization, you are forced into those situations, but it's not in anyone's particular control other than the MM... and blaming MM is far easier to do these days.)

The proposed system simply favors those that have paid or played more of this game, and would really be a poor choice for any new player coming into this game.

Edited by Deathlike, 09 March 2013 - 01:16 AM.

[Greyfyl]

Deathlike, on 09 March 2013 - 01:14 AM, said:

Waiting for people to select their mechs brings me back to MW3+MW4 where these things were far worse than it is now. In your scenario, you'd have 20 seconds to pick mechs/discuss, whatever. That comes to about 140 seconds (7 people picking ahead), which is TWO WHOLE MINUTES (even a failed MM search is shorter than this). You're adding this ON TOP of MM, which doesn't make this easier or better.

You are not taking 2 things into account...first of all the time for mm should be substantially less as you are now only finding a personal ELO match instead of trying to match entire teams.

Second.....most players will probably take 2-3 seconds to pick their mech. They know what they want to play, and it's not like you are allowing them to customize the mech at this point - just picking one.

Deathlike, on 09 March 2013 - 01:14 AM, said:

Even if you adjusted the time for picking, I doubt this would actually go over well... despite the demand for better communication, this won't always go over well with people. It's worse if you force the new player into a trial mech (as if trial mechs were barely competent with custom mechs).

I believe new players are often in trial mechs anyways are they not. Yes, until they get 1 mech of their own for each weight class they would have a greater chance of being 'forced' into a trial mech. There almost certainly should be a player on the opposing team in the same situation using personal elo instead of team elo, or at the very minimum a better chance for that to be the case.

Deathlike, on 09 March 2013 - 01:14 AM, said:

This also causes a greater handicap with strictly F2P players that won't have the desirable mechbay options. You would have to at least reconsider reorder who goes first (least experience first) and some would consider this a bad idea... especially if the mech they bring is woefully inadequate, wasting an particular mech slot that someone else would be more able to fill. (Note that since you have more control over this, it can and will drive people crazy. With the current randomization, you are forced into those situations, but it's not in anyone's particular control other than the MM... and blaming MM is far easier to do these days.)

The current system allows new players that don't even know how to group weapons to take an assault slot as often as they like. IMO, while that may be nice for allowing the player to run the mech he wants, it often severely reduces the effectiveness of the team he gets placed on.

I get what you are saying but IMO

Having more mech diversity and more even matches >>>>> being able to blame the matchmaking.

Deathlike, on 09 March 2013 - 01:14 AM, said:

The proposed system simply favors those that have paid or played more of this game, and would really be a poor choice for any new player coming into this game.

I think you are probably right to an extent on this issue, but tweaks could certainly be made to make the system more new player friendly. I think matching player ELO as opposed to team ELO would be a good starting point. I also think ELO needs to be tweaked to take into account player performance into account, not just win/loss. If someone does 1400 damage and gets 7 kills but still loses, chances are his elo should be going up, not down because he lost.

Edited by Greyfyl, 09 March 2013 - 05:44 AM.

[Greyfyl]

Jonathan Paine, on 09 March 2013 - 12:56 AM, said:

I really like this suggestion. Would need some tweaking to work with the elo system per weight class.

Thank you - I certainly know what I proposed here is very rough and would need tons of work.

But to be honest - the current system feels like it had no thought at all put into it. The matches are just as uneven now as ever, and often more unbalanced then before. I know that they will tweak ELO and MM, but how much tweaking can you do to a system that is just flawed from the start?

Edited by Greyfyl, 09 March 2013 - 06:25 AM.

[Deathlike]

Greyfyl, on 09 March 2013 - 05:44 AM, said:

You are not taking 2 things into account...first of all the time for mm should be substantially less as you are now only finding a personal ELO match instead of trying to match entire teams.

Second.....most players will probably take 2-3 seconds to pick their mech. They know what they want to play, and it's not like you are allowing them to customize the mech at this point - just picking one.

You are changing the dynamics of how these matches are to be played out.. it's a lot more significant than just time spent. ELO isn't perfect when it isn't given enough data samples/points to be made, thus making it a wildcard value until it is normalized/stabilized.

Quote

I believe new players are often in trial mechs anyways are they not. Yes, until they get 1 mech of their own for each weight class they would have a greater chance of being 'forced' into a trial mech. There almost certainly should be a player on the opposing team in the same situation using personal elo instead of team elo, or at the very minimum a better chance for that to be the case.

TBH, premade ELO needs to be calculated differently as the dynamics of a premade is far more significant than regular PUGs. Very few people like the trial (stock mechs). They simply are bad at what they are used for. Just look through every stock mech configuration in this game... they are for the most part, barely usable. There's a reason why PGI is having a future build-a-better heavy variant contest.

Quote

The current system allows new players that don't even know how to group weapons to take an assault slot as often as they like. IMO, while that may be nice for allowing the player to run the mech he wants, it often severely reduces the effectiveness of the team he gets placed on.

In a game where you are intentionally limiting what mech you can take, the gripes of "he took this slot and wasted it" would certainly increase. Have you ever noticed the threads mentioning the pitiful amounts of damage people manage to deal? You would actively annoy veteran players from their newbie counterparts to a greater level. It's easier to facepalm when you know you have someone who probably won't be able to contribute, than to rage against the newbie that took your favorite chassis slot. The rage is the same... where that rage is directed to is very different. Heck, this can happen between two newbies or two vets. It's even easier to point fingers at what is the perceived problem. Adding issues that wasn't a problem before is not a good way to go.

Quote

I think you are probably right to an extent on this issue, but tweaks could certainly be made to make the system more new player friendly. I think matching player ELO as opposed to team ELO would be a good starting point. I also think ELO needs to be tweaked to take into account player performance into account, not just win/loss. If someone does 1400 damage and gets 7 kills but still loses, chances are his elo should be going up, not down because he lost.

It depends on the context of his performance. I doubt any system would accurately reflect that. On the other hand, new player ELO will never correctly reflect his actual skill.. I'm not a fan of the MM at times, but it's supposed to be the fall guy for weirdness... not other players. That's not the community anyone wants to be a part of (blaming others).

Greyfyl, on 09 March 2013 - 05:48 AM, said:

Thank you - I certainly know what I proposed here is very rough and would need tons of work.

But to be honest - the current system feels like it had no thought at all put into it. The matches are just as uneven now as ever, and often more unbalanced then before. I know that they will tweak ELO and MM, but how much tweaking can you do to a system that is just flawed from the start?

It's flawed, and I've asked for revising (no comments as of yet); It will never be perfect, but it can be tweaked and it can be better. The issue is with some of the details and not entirely understand the dynamics of certain things... as a player... not as a developer.

Edit:
Here's an example of a poor trial mech...

Before the recent patch, there was a increase an trial K2 usage. The mech has 2 ppcs, 2 meds, 2 mguns, with 20 SHS. Just by converting to DHS (w/o adding any DHS), you would be cooling down at the VERY same rate as the SHS (w/o any increases to heat containment, which SHS would do, I think). Anyone who has used PPCs can imagine how how this mech has a very weak heat managing system.

We could also goto the other end, where the CDA-2A on the other end of the scale.. has only 2 meds, 1 small laser as its stock mech default. Not only is this not very effective (vs any decent customized mech), it isn't even competitive. It certainly will allow you to scout and cap, but it can't hold a candle to having your own customized mech.

There's a lot more stuff that has to be considered... primarily the implications of the change (always think, WORST CASE SCENARIO), because there's always one that you didn't think of.

Edited by Deathlike, 09 March 2013 - 08:07 AM.

[Greyfyl]

Wow, it's amazing how you see the negatives of my proposal so easily but not the overwhelming negatives of the existing system.

"In a game where you are intentionally limiting what mech you can take, the gripes of "he took this slot and wasted it" would certainly increase. Have you ever noticed the threads mentioning the pitiful amounts of damage people manage to deal? You would actively annoy veteran players from their newbie counterparts to a greater level."

So which is worse? In my system you would at least have some kind of weight class balance. Right now its 2-3 ddc per match, I see alot of complaining about that too. The other point that you missed completely is that ELO should be getting newbies away from the veteran players. It should be a gradual progression for new players. Right now rookies can still be thrown in way over their heads against veteran players with the ELO supposedly balancing out the overall 'team elo'. It's not working.

In my system it would be much easier to put 500 elo players in groups against other 500 elo players because you don't have to ALSO find a weight match. And to be honest, from what I understand the queue doesn't currently work like that at all anyways.

Anyways, you have your opinion and I have mine, I think the current system is god-awful and boring. I know I'm not alone, after all - the lack of weight balance is one of the main reasons why the 8-man games are deserted.

[Deathlike]

You're missing the point that the MM itself has to be tweaked. That is part of the problem. You are changing one problem and adding new problems that needs their own solution. Even if on the surface it sounds good, but there are logistics problems that would have to be addressed. I do have to look at the downsides, even if I don't have a major problem with it (your idea sounds well intentioned, just flawed IMO).

Newbies will have to play veterans at any given point.. that is the nature of PUGs. This is not going to change. The rate of this occurring should change through ELO and MM tweaking, but it will always be an imperfect science. Blaming newbies for losing is one thing.. it happens all the time... but now you have a function that can LITERALLY determine who you can blame for the lose. Do you really want the whining to be "X refuses to only take a different mech outside of Y, and is screwing the entire team/game" or "ELO/MM is broken". It's easier to deal with the latter, unfortunately.

I have no actual issues in weight differences, if the skills in the current mechs in the current tonnage are "relatively accurate" to the opposing team. However, there is a certain amount of leeway that this can be allowed (probably would have to factor in total tonnage).

Weight balancing is nearly impossible on 8 man premades... just with the # of 8-mans vs PUGs in general (in at least an order or 2 difference in magnitude). It's actually harder to put together the 8-man IMO than it is to deal with 4-man premades (it's a #s issue of course).

Edited by Deathlike, 09 March 2013 - 11:42 AM.

[Greyfyl]

Deathlike, on 09 March 2013 - 11:41 AM, said:

You're missing the point that the MM itself has to be tweaked. That is part of the problem. You are changing one problem and adding new problems that needs their own solution. Even if on the surface it sounds good, but there are logistics problems that would have to be addressed. I do have to look at the downsides, even if I don't have a major problem with it (your idea sounds well intentioned, just flawed IMO).

I'm not missing the point at all - the current system is boring and flawed at the core. Tweaking it really doesn't bring anything to it at all. You keep saying there are 'logistic problems' as if that makes your whole argument somehow valid. It means nothing.

Of course my system is flawed - I came up with it in 1/2 hour and I'm not a paid developer. Again that doesn't change the fact that the current system is god awful and needs something a hell of alot bigger than just a tweak to how elo works.

Edited by Greyfyl, 10 March 2013 - 06:20 AM.

[Treckin]

Greyfyl, on 06 March 2013 - 10:58 AM, said:

My suggestion for a more interesting/balanced approach to individual matches:

First of all it would require a slight change in how you pick 'ready' mechs in the mechlab. You would still start with 4, but you must have one of each weight class. I would suggest the ability to buy additional 'ready' slots with c-bills/mc, maybe up to a maximum of 6. Any purchased ready slots can be filled with any weight mech. See images below.








Once you have the minimum ready slots filled (they can be filled with trial mechs if necessary) with at least one from each weight class you can hit the launch button. Once you hit the launch button the matchmaker will try to find an ELO match for you personally (this would be for puggers) and put the two of you on opposing teams. Once each team is full you will go to the new drop preparation screen. This will show you only the players on your team.



At this point you will know the map and the drop configuration (which will be random). You will also at this point be placed in a numerical order from 1-8 with your other teammates. This should be a weighted system that tends to put the more experieced players towards the top, but can allow all players a chance at being #1. With all players in the drop prep screen you can use team chat to discuss what mechs to bring, and after 20 seconds or so the first player will be able to choose his mech from the mechs available at the top of the screen.



Each player will pick his mech as it becomes his turn. I would also think that a timer would be needed at this point to keep anyone from holding up the entire team. Once everyone has selected their mechs and readied up on both teams- the game launches.

Fantastic!

[MaddMaxx]

Greyfyl, on 08 March 2013 - 08:37 PM, said:

Just curious, but why would it never be approved? You still have to have 4 ready to go mechs - in a way it would promote grinding our more mechs and ready bays. I would think that would be a positive from pgi's pov. The current MM has many issues that will never be resolved with just elo. It needs help.

And what in the world does it's a logistics problem mean? I don't think we really have to head out a re-arm all the mechs in the field after each match, so are you talking about server load, bandwidth, or what?

Think about it. Unless I am guaranteed to get first pick a minimum of 1 in 4 games, the whining would be tremendous. Despite having prepped 4 choices, everyone, upon seeing the Map, will want that one for that Map. When they do not get it multiple times in a row, that system crumbles under its own fail.