

Theory on Match Making
Started by Evex, May 15 2012 12:06 AM
30 replies to this topic
#21
Posted 15 May 2012 - 10:40 AM
Not to mention pilot level, win/loss, etc
Lots of metrics they can bring to bear as part of the matchmaking process.
Lots of metrics they can bring to bear as part of the matchmaking process.
#22
Posted 15 May 2012 - 01:10 PM
I like that idea but I think it would make sense if top priority is first give to groups that go in together and then the system figures out the rest for the randoms.
#23
Posted 15 May 2012 - 01:29 PM
Come on guys you cant rank by tonnage ... it has to be a BV type metric. If this also includes some modification for the skills a character has purchased then I can live with that.
After all skills will presumably be bought with XP ... XP that will be earned by fighting and killing/damaging/scouting people ... so you can measure 'skill' to some extent. (YES I know we will prob be able to buy XP boosters etc.)
After all skills will presumably be bought with XP ... XP that will be earned by fighting and killing/damaging/scouting people ... so you can measure 'skill' to some extent. (YES I know we will prob be able to buy XP boosters etc.)
#24
Posted 15 May 2012 - 01:53 PM
You have to assume for match making purposes, grouped players are organised and more effective than an equivalent number of none grouped players.
if you're grouping and lack any form of teamwork within your group vs enemies then you are handicapping yourself.
So a group of 3 players all using 40ton mechs would have to count as more tonnage than their mechs alone where as 3 individuals shoved together on a team ungrouped would only count as based tonnage, because they likely will not stick to each other, support each other and focus fire on an enemy.
That's how I would consider it anyway.
Is it possible 3 pug players could individualy and purely by chance work better together then 3 grouped players? yes, is it more likely though? no and that's how you have to form your matchmaking calculation...the law of averages.
if you're grouping and lack any form of teamwork within your group vs enemies then you are handicapping yourself.
So a group of 3 players all using 40ton mechs would have to count as more tonnage than their mechs alone where as 3 individuals shoved together on a team ungrouped would only count as based tonnage, because they likely will not stick to each other, support each other and focus fire on an enemy.
That's how I would consider it anyway.
Is it possible 3 pug players could individualy and purely by chance work better together then 3 grouped players? yes, is it more likely though? no and that's how you have to form your matchmaking calculation...the law of averages.
#25
Posted 15 May 2012 - 01:57 PM
FrostPaw, on 15 May 2012 - 01:53 PM, said:
You have to assume for match making purposes, grouped players are organised and more effective than an equivalent number of none grouped players.
if you're grouping and lack any form of teamwork within your group vs enemies then you are handicapping yourself.
So a group of 3 players all using 40ton mechs would have to count as more tonnage than their mechs alone where as 3 individuals shoved together on a team ungrouped would only count as based tonnage, because they likely will not stick to each other, support each other and focus fire on an enemy.
That's how I would consider it anyway.
Is it possible 3 pug players could individualy and purely by chance work better together then 3 grouped players? yes, is it more likely though? no and that's how you have to form your matchmaking calculation...the law of averages.
if you're grouping and lack any form of teamwork within your group vs enemies then you are handicapping yourself.
So a group of 3 players all using 40ton mechs would have to count as more tonnage than their mechs alone where as 3 individuals shoved together on a team ungrouped would only count as based tonnage, because they likely will not stick to each other, support each other and focus fire on an enemy.
That's how I would consider it anyway.
Is it possible 3 pug players could individualy and purely by chance work better together then 3 grouped players? yes, is it more likely though? no and that's how you have to form your matchmaking calculation...the law of averages.
So what happens when three friends who have never played a Mech Simulation game in their lives pick up this awesome new MMOSim called MWO. If these guys try and play together they will constantly find themselves out numbered and outgunned. Now they arent having fun. Add in a little trash talking from elitist experienced gamers and we just lost three potential players.
Fighting in an organized team should give you an advantage, not just put you on Par.
#26
Posted 15 May 2012 - 02:02 PM
A group certainly counts for more than an equal number of uncoordinated random people on average. But, it doesn't seem fair (especially to less-skilled players forming into small friend groups) to give them a concrete handicap. Simply match them with similarly sized teams. Obviously pilot level, win/loss, and other metrics can also be put into play to put roughly equivalent groups together, just as I would expect pilot level or some other metric to sort out which random single-queue players should face each other.
you have to account for group size/coordination or tight-knit groups will steamroll every matchup they're in. you also have to account for pilot level/xp/win:loss/kill:death or other metrics as well to match up similarly skilled players.. But you don't make the higher level players bring less tonnage/BV
you have to account for group size/coordination or tight-knit groups will steamroll every matchup they're in. you also have to account for pilot level/xp/win:loss/kill:death or other metrics as well to match up similarly skilled players.. But you don't make the higher level players bring less tonnage/BV
#27
Posted 15 May 2012 - 02:19 PM
I was hoping there would be a more complicated system, then Mechlab --> Match --> Map --> Play. I know that in realtime, travelling the Inner Sphere would take weeks to cross. But I shouldn't be able to play on New Avalon and then play my next match on Hesperus II, 700 light years away.
Maybe the ability to play on any planet within 120LY from your last position would be fair?
Maybe the ability to play on any planet within 120LY from your last position would be fair?
Edited by Zakatak, 15 May 2012 - 02:19 PM.
#28
Posted 15 May 2012 - 02:56 PM
Zakatak, on 15 May 2012 - 02:19 PM, said:
I was hoping there would be a more complicated system, then Mechlab --> Match --> Map --> Play. I know that in realtime, travelling the Inner Sphere would take weeks to cross. But I shouldn't be able to play on New Avalon and then play my next match on Hesperus II, 700 light years away.
Maybe the ability to play on any planet within 120LY from your last position would be fair?
Maybe the ability to play on any planet within 120LY from your last position would be fair?
We have no idea how they will handle the meta game. Especially how it connects to joining matches ... I for one would love info on this!
#29
Posted 15 May 2012 - 03:31 PM
From Ask the Devs 2 - Answers:
See? We won't need a complicated matchmaking system that restricts player choice. Paul says right here that the 'Mechs are mostly balanced against each other. Such a system would only serve to frustrate players who want to play their favorite class but can't because the BV balancing will not allow it. From what it seems Paul says, the different 'Mech classes are fairly evenly matched and such a frustrating matchmaking system will not be needed.
Dammit, Paul!
I believe that, ideally, the game should balance itself. In a truly balanced game, matchmaking should come down to player skill, not the equipment they have. If the matchmaking system is based on the BV or tonnage of the 'Mech, then it indicates that this game is clearly unbalanced. After all, if the game is balanced, a light chassis should be just as helpful to the team as an assault chassis.
A matchmaking system based on BV and tonnage would only frustrate players and does not help at all towards balancing the game.
No matchmaking based on my preferred equipment, please.
Hopefully, a matchmaking system based on equipment is not in place when the game goes into beta and the devs realize they won't need one.
Paul said:
Third, what we’ve found is that there is no ideal loadout, it comes down to how you play with the loadout that you’re using that matters. You may think that everyone would jump in an assault ’Mech loaded to the nines with weapons but this isn’t the case. All ’Mech categories are effective against assaults. In fact, all ’Mechs are very capable of taking down any other category of ’Mech. We’ve seen the entire gamut of gameplay playout on the field and it comes down to working together and actually sticking together to be successful. I’ve seen assaults blow holes through lights… I’ve seen lights peck and chip away at assaults and core them… I’ve seen heavies engaged in closed quarter combat only to be outflanked by a fast mover punching holes in the back armor and I’ve seen slower heavies/mediums stand their ground and devastate enemies in outnumbered combat.
See? We won't need a complicated matchmaking system that restricts player choice. Paul says right here that the 'Mechs are mostly balanced against each other. Such a system would only serve to frustrate players who want to play their favorite class but can't because the BV balancing will not allow it. From what it seems Paul says, the different 'Mech classes are fairly evenly matched and such a frustrating matchmaking system will not be needed.
Paul said:
Lastly, there will be a team balancing mechanic in place when it comes to finding matches. We are still working out the final details on how we want this to work whether it goes with BV or based on tonnage or a combination therein.
Dammit, Paul!
I believe that, ideally, the game should balance itself. In a truly balanced game, matchmaking should come down to player skill, not the equipment they have. If the matchmaking system is based on the BV or tonnage of the 'Mech, then it indicates that this game is clearly unbalanced. After all, if the game is balanced, a light chassis should be just as helpful to the team as an assault chassis.
A matchmaking system based on BV and tonnage would only frustrate players and does not help at all towards balancing the game.
No matchmaking based on my preferred equipment, please.
Hopefully, a matchmaking system based on equipment is not in place when the game goes into beta and the devs realize they won't need one.
#30
Posted 15 May 2012 - 03:36 PM
The queue should DEFINITELY take into consideration pre-made groups and solo/duo queues. It should attempt to find another group thats queuing, but if it can't it doesn't mean you dont get to play, just means takes a bit longer because it gives up on finding another premade team.
A group of 8 guys on teamspeak vs 8 random guys playing together in the game is NOT a fair match, however at times, may be necessary to reduce queue times.
A group of 8 guys on teamspeak vs 8 random guys playing together in the game is NOT a fair match, however at times, may be necessary to reduce queue times.
#31
Posted 15 May 2012 - 09:52 PM
Ghost73, on 15 May 2012 - 03:31 PM, said:
From Ask the Devs 2 - Answers:
See? We won't need a complicated matchmaking system that restricts player choice. Paul says right here that the 'Mechs are mostly balanced against each other. Such a system would only serve to frustrate players who want to play their favorite class but can't because the BV balancing will not allow it. From what it seems Paul says, the different 'Mech classes are fairly evenly matched and such a frustrating matchmaking system will not be needed.
Dammit, Paul!
I believe that, ideally, the game should balance itself. In a truly balanced game, matchmaking should come down to player skill, not the equipment they have. If the matchmaking system is based on the BV or tonnage of the 'Mech, then it indicates that this game is clearly unbalanced. After all, if the game is balanced, a light chassis should be just as helpful to the team as an assault chassis.
A matchmaking system based on BV and tonnage would only frustrate players and does not help at all towards balancing the game.
No matchmaking based on my preferred equipment, please.
Hopefully, a matchmaking system based on equipment is not in place when the game goes into beta and the devs realize they won't need one.
See? We won't need a complicated matchmaking system that restricts player choice. Paul says right here that the 'Mechs are mostly balanced against each other. Such a system would only serve to frustrate players who want to play their favorite class but can't because the BV balancing will not allow it. From what it seems Paul says, the different 'Mech classes are fairly evenly matched and such a frustrating matchmaking system will not be needed.
Dammit, Paul!
I believe that, ideally, the game should balance itself. In a truly balanced game, matchmaking should come down to player skill, not the equipment they have. If the matchmaking system is based on the BV or tonnage of the 'Mech, then it indicates that this game is clearly unbalanced. After all, if the game is balanced, a light chassis should be just as helpful to the team as an assault chassis.
A matchmaking system based on BV and tonnage would only frustrate players and does not help at all towards balancing the game.
No matchmaking based on my preferred equipment, please.
Hopefully, a matchmaking system based on equipment is not in place when the game goes into beta and the devs realize they won't need one.
I didn't think its possible ..but I disagree with practically everything you just said from your opinions all the way down to your interpretations of what the dev's said. A bv ranking system would include a value for scout modules ... it would include the fact that yes a light mech with 2 mg's and a med laser could in fact hurt an assault, and given enough time kill it, Repeated 'lucky' head hits will kill anything. Unlike in WoT where weapons have to penetrate, MWO has an ablative armour value system which keeps all mechs effective against all classes.
You also say that in a 'truly balanced game' Everything will come down to skill and that a light mech will be as useful as a any other weight class.
However to say that a BV system would not be needed because all mech classes will already be balanced is crazy. E.g. to suggest that a 'locust' (should we get one) has the same chance of taking down an Atlas, as the atlas has of squishing the locust is frankly ludicrous. If it was true then they would have made a far bigger design mistake!
What Paul was trying to say, I think, (Who truly can understand the words of gods?) is that a team that does not know where the enemy is will lose, so you need scouts and you need artillery mechs and you need grunt mechs ... balanced teams work in this game.
You also suggest that a matchmaking system based on the BV of a mech can not be balanced. Why not? So long as it is a well thought out system that includes everything from Engine size and relative speed, ease to hit vital modules and equipment (the critical location system) weapons, tech level, EWar equipment, over all size, mobility and a host of other things, then the relative value of the mech will stay true.
Player skill can never be measure by the system, Character experience can, so that should be factored in also.
Frankly I don't see how a BV system could be frustrating for anyone.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users