Matchmaker Ideas?
#1
Posted 25 February 2014 - 01:55 PM
After a bit of considering I've decided to post an ideas that maybe would help the matchmaking.
1. Implement a battle value system that takes into account the following: tonnage, player elo (personal w/l rate and avg xp), mech elo (how much a particular mech variant wins/loses, and xp it gains on avg), and team modifier (bump elo by some number for being teamed up).
2. Use that bv system to build matches in a fairly simple manner by having matchmaker pull the first 8 ppl in the queue and split them into 2 teams. Total the 2 teams bv, add player to the weaker team, retotal and repeat until one team reaches 12 ppl. Then look through the queue for a person who can fill one more spot and bring the 2 teams to within 10% parity in BV. If no person can be found within a short period of time launch with game being teams being uneven.
Please note new players will need a baseline elo value and the weight given to each aspect of the bv would need to be carefully considered. Personally I think something like 40% player elo, 30% mech elo, and 30% tonnage would be a good starting point with a 15% boost for being teamed up.
Now how would this play out? Couple of things would happen. Great players with great mechs would likely get a lot of poor players on their teams to balance things back out. Conversely very bad players with very bad mechs would add very little to the bv so would end up on the numerically larger teams more often to balance things back out. Where things would get interesting is very good players playing mechs which are middling to poor elo would basically handicap their bv to get themselves more interesting matches.
Key thing is the matchmaker would account for the things everyone thinks matters to match outcomes and the devs could modify the weight each of these things is given AND matchmaker would not be constrained to make 12v12 matches but simply focus on balancing bvs.
#2
Posted 25 February 2014 - 02:42 PM
#3
Posted 25 February 2014 - 02:55 PM
Mystere, on 25 February 2014 - 02:42 PM, said:
I completely agree about that. I would hope that community warfare is somewhat more like what happens with World of Tanks where you choose what goes in, know what the map is and can have a plan in place.
#4
Posted 26 February 2014 - 06:30 AM
Mystere, on 25 February 2014 - 02:42 PM, said:
I agree. CW should be a "lobby" system that hopefully lets you know what you're getting yourself into before you drop with your group.
"Oh, I see. This one is going to be River City Night. I like that one, let's enter the lobby. Oh dear. It looks like half the people here are in Trial Stalkers and don't appear to know what they're doing. No thank you, I'll just leave this lobby and find a better one."
In that instance, if you decided to drop into what appears to be a complete roflstomp, you'd have noone to blame but yourself.
BUT...I'd still like to see the ability to have random matches that don't effect CW at all. Something similar to what we've got now.
In fact, since there's no good tutorial system anywhere on the horizon, I'd like to simply say that requiring new players to play in the "PUG Queue" until a certain point before allowing them to even consider getting involved in CW would be a good idea.
If nothing else, you could call the "PUG Queue" a Simulator or something. You've gotta put in Simulator time and show you know how to pilot a mech before they'll actually let you walk one around a parking lot...kinda like Driver's Ed.
#6
Posted 28 February 2014 - 03:06 PM
Joseph Mallan, on 26 February 2014 - 06:34 AM, said:
It doesn't look like it's going to be....at least...if CW is nothing but Private and Premium type matches. Granted, they may still involve the PUG queue into the overall CW stuff...we'll have to see.
It'd suck if you had a group like The Law that did their thing in Premium matches to defend Coventry...and have managed to successfully beat all comers for 20 different fights....only to lose the planet because the Matchmaker threw down 50 matches to PUGgers that lost. But, you win some, you lose some, I guess.
#7
Posted 28 February 2014 - 07:50 PM
It would need a lot of tweaking, but the end result would be better than class-vs-class.
#8
Posted 01 March 2014 - 07:49 AM
Willard Phule, on 28 February 2014 - 03:06 PM, said:
It doesn't look like it's going to be....at least...if CW is nothing but Private and Premium type matches. Granted, they may still involve the PUG queue into the overall CW stuff...we'll have to see.
It'd suck if you had a group like The Law that did their thing in Premium matches to defend Coventry...and have managed to successfully beat all comers for 20 different fights....only to lose the planet because the Matchmaker threw down 50 matches to PUGgers that lost. But, you win some, you lose some, I guess.
I thought I read private matches were not going to decide things like planetary conquest?
#9
Posted 01 March 2014 - 08:26 AM
Bodha, on 25 February 2014 - 01:55 PM, said:
After a bit of considering I've decided to post an ideas that maybe would help the matchmaking.
1. Implement a battle value system that takes into account the following: tonnage, player elo (personal w/l rate and avg xp), mech elo (how much a particular mech variant wins/loses, and xp it gains on avg), and team modifier (bump elo by some number for being teamed up).
2. Use that bv system to build matches in a fairly simple manner by having matchmaker pull the first 8 ppl in the queue and split them into 2 teams. Total the 2 teams bv, add player to the weaker team, retotal and repeat until one team reaches 12 ppl. Then look through the queue for a person who can fill one more spot and bring the 2 teams to within 10% parity in BV. If no person can be found within a short period of time launch with game being teams being uneven.
Please note new players will need a baseline elo value and the weight given to each aspect of the bv would need to be carefully considered. Personally I think something like 40% player elo, 30% mech elo, and 30% tonnage would be a good starting point with a 15% boost for being teamed up.
Now how would this play out? Couple of things would happen. Great players with great mechs would likely get a lot of poor players on their teams to balance things back out. Conversely very bad players with very bad mechs would add very little to the bv so would end up on the numerically larger teams more often to balance things back out. Where things would get interesting is very good players playing mechs which are middling to poor elo would basically handicap their bv to get themselves more interesting matches.
Key thing is the matchmaker would account for the things everyone thinks matters to match outcomes and the devs could modify the weight each of these things is given AND matchmaker would not be constrained to make 12v12 matches but simply focus on balancing bvs.
At this point we know that we are getting another terribad system to replace the current one (basically the current one, except wait times increased as 3 people per team will be forced to play medium) at the end of the April. What's the point of talking it out? We all know that PGI aren't interested.
#10
Posted 01 March 2014 - 08:41 AM
#11
Posted 01 March 2014 - 08:49 AM
NextGame, on 01 March 2014 - 08:26 AM, said:
The hope PGI may correct issues relating to mediums and/or introduce aspects like role warfare that then give more purpose to the idea of using the medium roles.
But yes without something to better reflect the situation of IS economics to represent the best reason of strategically using Medium Mechs more in CW. Then it comes down to pilot choice and preferences.
So if liberal freedom is applied to ignore Mediums as per the precedents associated to their builds it is hardly surprising that pilots, tactical coordinators will choose or recommend other classes.
And yet the build choices or limitations applied to Medium Mechs at the same time also applied or interpreted from the same precedents from BT to be a disadvantage in MWO:
1) Nerfed Heat dissipation which encourages ALPHA builds more.
2) SRMs HD still problematic,
3) ML, MPL not effective by comparison due to beam mechanics and higher heat values (see 1).
4) Oversized proportions making them easier to hit relative to armour than larger mechs.
5) Some XL need for insurgent performance when linked with (4) making these builds fragile.
6) Lack of tonnage restrictions means lots of opponents having much more defense to try hit and run tactics.
7) The ranged meta confidence diminishes the idea of mobility tactics to flank and ambush opponents.
8) Engine limitations (which do associate to build potentials naturally anyhow).
9) Restricted Variant and Chassis ECM use.
As a result Mediums have been the recognised pinata in MWO since its inception. And other choices simply then being more preferable to use.
The faster phoenix Mechs and especially the Shadowhawk being a more recent introduction that allows for the medium to perform in the Meta. But this mostly associated with the streak light killer or light direct fire support in the AC/PPC sniping meta.
Nothing really has been done to encourage the short game use with the majority of the Medium Mechs and this I believe due to poor design choices in either game mechanics or as specifically applied to the Medium class who currently are trying to perform in a demographic with the current Meta they are not really designed for. And this then removes one of the more natural counters to the current meta with direct fire support with Mechs who can be best applied to hit entrenched sniping positions with surprise positioning/flanking mobilty.
Having said that they are not completely useless and possible with a few corrections here and there more specific to the Medium class or applied role warfare could encourage more use with them. This could then incentivise their use and would make them sit better naturally in more flexible tonnage restrictions as they then would be a more viable choice.
As such we are now faced with a more simplistic view of rigid class numbers in lances due to these problems existing in the game.
#12
Posted 01 March 2014 - 09:04 AM
Willard Phule, on 01 March 2014 - 08:41 AM, said:
PLease know I feel PGI should be picking players right now to be core members of Canon units, so we have something in place before the map launches. CW would be a rats nest if they don't get something in place to start from.
#13
Posted 01 March 2014 - 09:09 AM
5 user(s) are reading this topic
0 members, 5 guests, 0 anonymous users