713 replies to this topic

[Knight Captain Morgan]

TLDR: the population has dropped enough that match quality is only gonna get worse from here on out. + a bunch of numbers meant to distract the casual solo pugs left that competitive premade teams will still be allowed to roflstomp them at will.

[Remover of Obstacles]

OZHomerOZ, on 08 June 2020 - 04:36 PM, said:

I'm pleasantly surprised

No risk no reward and the status quo clearly is not working

Totally agree.

OZHomerOZ, on 08 June 2020 - 04:36 PM, said:

Edit: Off the cuff your formula looks ok, maybe reduce the ends to 22 and add two no change PSR in the middle.

I like the no change option for the middle ground.

OZHomerOZ, on 08 June 2020 - 04:36 PM, said:

Also I'm good wih the current match score system, just speaking for my self.

I agree. I feel the match score system isn't the real problem here.
Tweeking ithe matchscore system might be nice. You could reward pilots that tank damage well. But it gets complicated since some mechs have far superior hit boxes, quirks and tons of armor which require scaling. Maybe AMS gets toned down a tad. All in all, I am afraid there is little juice to be gotten from a lot of squeezing of the match score fruit.

On the other hand, accepting that we will be doing a reset which will throw pilots of different skill levels together. I think we want a system based on match score since it will more quickly separate Tier 1 pilots form Tier 5 pilots then a Win/Loss system. I think we can assume that in a typical match there could be Tier 1, Tier 2 and Tier 3 pilots on both teams. Why slow the process of tier separation because you happen to win about half of your games?

Likewise, if the reset puts all existing pilots in Tier 3. The Tier 3 pilots should pretty much stay there if we allow the middle match score performers to receive zero points or no change to PSR.

Edited by Remover of Obstacles, 09 June 2020 - 08:59 AM.

[Fiona Marshe]

Reducing the Match Score for damage done to about 20% of its current value would boost the teamwork components.

[LordNothing]

typical bait and switch. i am very disappointed.

Edited by LordNothing, 08 June 2020 - 07:04 PM.

[Kosomok]

Hawker Siddeley, on 08 June 2020 - 05:41 PM, said:

Sorry I should have exampled my point better. How do you differentiate someone who ignores the call, or YOLOs in, wasting their mech but still taking as much damage as someone who actually tanks the push? How do you differentiate between someone using their armour effectively and someone who isn't?


You brought up an example of how increasing the weight of capping on MS could change the dynamic and I said why it wouldn't, why you would still ultimately see what currently happens.

And ultimately my point that in most cases caps do not win conquest matches was never addressed. A murderball that gets the initial 2 cap then rolls into the enemy team is still more likely to win than the team that has people not contributing to the team by running around capping. Encouraging and rewarding people for hampering their team like that is not something that should be rewarded.

One way to differentiate who charges and who effectively tanks is to look at damage done vs. damage taken. A second way would be to look at TTK (time to kill--when in the match that the payer died, if they do so).

Just recently I have seen caps win conquest matches several times... even when it was only 1 v 4 or 5 left on the other side. Is it necessarily regular? No--but then there is no incentive for most players to play the game that way.

Part of the reason the way the game is played (in whatever mode) that it is played is that the existing system preferentially rewards playing the game in certain ways. It rewards particular patterns and activities. If you change the incentives, you will change the patterns of play--because people respond to incentives.

The question then becomes, what do you want to reward--what play styles do you give preference to?

A major problem in the system, IMO, is that the MS criteria is the same across all the classes--which puts lighter classes at a disadvantage in many cases or even actually removes the justification to really even play them--or, at least, to play them in an even somewhat realistic fashion. MS criteria and rewards need to be a function of mech class/role.

[Thebackson]

Paul Inouye, on 08 June 2020 - 02:52 PM, said:

As Russ mentioned, I'm going to provide a list of all events that currently trigger and apply value to Match Score.

Each of these triggers has a value assigned to it. The values are essentially the weighting of the variable in the overall sum of all actions listed. At the end, the number is multiplied by a decimal multiplier that reduces the overall sum to a number that is manageable.

win - Base value for winning.
loss - Base value for losing.
teamkills - Loss for team killing.
spottingassist - Gain for spotting an enemy and that enemy takes damage. (Press R)
componentdestroyed - Gain for destroying a component on an enemy.
death - Loss for dying.
capturewin - Bonus gain for winning by capture.
captureassist - Gain for helping others capture a capture point by being in the capture radius.
suicide - Loss for suiciding.
saviorkill - Gain for helping a teammate under fire and you get the kill shot on his highest damaging opponent.
defensivekill - Gain for helping a teammate under fire and you get the kill shot on an opponent damaging them.
uavkill - Gain for any kills happening under a UAV you deployed.
uavlockeddmg - Gain for any damage done to a target that is under your UAV.
uavdetection - Gain for any new enemy detected by your UAV.
counterECM - Gain for countering enemy ECM.
counterECMLockedDmg - Gain for damage done to enemies under your counter ECM.
turretkill - Gain for killing a turret.
killblow - Gain for getting the killing blow on your enemy.
killassist - Gain for damage done to enemies upon kill but you didn't explicitly get the kill shot.
teamdmg - Loss for team damage done.
damagedone - Gain for damage done to enemies.
killmostdmg - Gain for getting the kill shot and you did the most damage to your enemy.
solokill - Gain for killing an enemy without the assistance of your teammates.
scouting - Gain for targeting enemies without damage being done.
flanking - Gain for being out of LoS to your enemy and behind enemy line.
capture - Gain for capturing a capture zone.
capturepulse - Gain for time you are capturing in a capture zone.
firstcapture - Gain for capturing the first capture zone in a match.
brawling - Gain for being in combat agaist multiple opponents.
tagdmg - Gain for any damage done to an enemy you have tagged.
tagkill - Gain for any kill done to an enemy you have tagged.
narckill - Gain for any kill done to an enemy you have narced.
hitandrun - Gain for attacking an enemy and escaping their LoS for an amount of time.
tagstealth - Gain for tagging an enemy behind enemies and not being targeted by them.
lanceformation - Gain for time spent near lancemates.
protectmedium - Gain for killing an enemy who is attacking a medium class teammate.
protectlight - Gain for killing an enemy who is attacking a light class teammate.
protectproximity - Gain for killing an enemy who is near any teammate.
powercell_pickup - Gain for picking up a power cell.
powercell_dropoff - Gain for dropping off a power cell.
incrusion_destruction - Gain for destroying objective objects in Incursion.
kill_powercell_carrier - Gain for killing a power cell carrier.
ams_missile_destroyed - Gain for missiles destroyed by your AMS system.
 
matchscore_scale - A multiplier used to reduce the sum of all match score activity to keep numbers reasonable.

Please use this specific thread for proposals. It will help keep all suggestions in one place for easier feedback management. Make sure you click "Like" on the ones that are the best in your view.

If the community can agree on weighting of these events, we will review them on our side to make sure they're fair to as many play styles as possible. Once that is done, we'll implement the numbers and try them out.

Please keep in mind, we're going to need buy-in from the majority of players.

Is there a way to find out the current values these are currently at to understand the scale to apply changes.

For example AMS should probably be toned down very slightly due to quad ams builds or better yet keep them where they are but cap the maximum amount they can contribute.

Also tone down damage done but slight increase for killing blow and kmdd. but without knowing what the current values are it is hard to know how they relate to the other MS values.


But YAY this is awesome, exactly what I was thinking should happen.

Edited by Thebackson, 08 June 2020 - 07:13 PM.

[Slothasaurus]

Sorry if these have been mentioned before but giving more weight to damage taken would be nice. There's something to be said for a mech leading a charge taking lots of shots so the rest of the team can get into position to fire. I have done it many times being in a light or medium find a bigger mech to buddy up with. He eats the damage. He should get rewarded for that.

Also more penalty than there currently is for team damage. There are times before the fight really gets into it I have taken more team damage than enemy fire. I have been spectating people and seen people unload on an enemy but a teammate might of been a bit off to the side and take half the volley. If people have even a bit of fire discipline or if they make the one-off mistake it won't affect them. It will affect the people that continue to not be careful.

Edited by Slothasaurus, 08 June 2020 - 07:30 PM.

[MOBAjobg]

Wow, I love this PSR formula computation so it's a go for me now.

[ShogunKid]

So far, the clearest voice in this discussion to me has been from JustcallmeASH on his reddit post ---> https://www.reddit.c...utm_name=iossmf

He explains what was proposed by PGI and consolidates three of the best options given by the community (credit to Kami, Arjoe, and Dogmeat) -- one zero sum option and surprisingly, two not zero sum options. My vote is we get behind one of these options and go for it!

I would also strongly suggest we consider tweaking Match Score values AFTER we figure this PSR thing out.

[Andrzej Lechrenski]

The Problem*

Before we can work together as a community to fix our problem, we are going to first need to clearly define it. With regard to match maker, I propose that this is the specific issue we need to address: players who do not play competitively (hereafter referred to as casual players) are being placed in matches with players who do play competitively. This is a problem both for the casual players, who have a miserable time due to being either "stomped" or "carried" (neither of which is especially fun), as much as it is a problem for the competitive players who feel they are either being unfairly handicapped or else breezing through the game with no challenge. I'm going to assume that we have consensus on at least that much.

With the player pool as small as it is, there is clearly never going to be a 'perfect' solution, in which everyone will be matched with players very close to their level of skill and competitiveness, but that is why I have defined it broadly instead of in terms of shades of grey. Still, I think that we all agree that there is room for improvement in terms of making the game a more consistently satisfying experience for a broader number of people. I assume we have consensus here as well.

Further, I believe there a few more assumptions that we can make. For the competitive players, it seems almost universally agreed that ranking should be done through a zero-sum system because, over time, that prevents the grouping of players with vastly disparate levels of skill. Traditionally, this has been done by evaluating only wins and losses, because, all else being equal, past wins are highly predictive of future wins. With regard to casual players, the priority seems instead to see individual performance "rewarded" and poor performance "punished", regardless of the outcome of the match. Some competitive players agree with that last notion as well, although I would venture the guess that they mostly due so as a vestige of rage against the (more or less) random match making we have at the moment, as opposed to having thought about it analytically. Regardless, I think it is clear that the collision of these ideas is the greatest factor in our so-far lack of consensus. We need to reconcile this.


Solutions

Note here that it is best to avoid the words "reward" or "punish" in terms of the PSR numbers given here. Consider only what will happen over time (and not as a conclusion to an individual match) if these systems are implemented. Also, I am simply using Russ’ suggested metric. Some people may imagine “equals” signs here and there or wish to change a few magnitudes. Those are a trivial thing in the big picture.


Solution 1

Just rank everybody strictly on relative match score. Ignore wins and losses. To see what that looks like, I'll just copy and paste Russ' numbers from the OP.

player 1 +24
player 2 +22
player 3 +20
player 4 +18
player 5 +16
player 6 +14
player 7 +12
player 8 +10
player 9 +8
player 10 +6
player 11 +4
player 12 +2
player 13 -2
player 14 -4
player 15 -6
player 16 -8
player 17 -10
player 18 -12
player 19 -14
player 20 -16
player 21 -18
player 22 -20
player 23 -22
player 24 -24

This is the 'sol ipsis' approach of "every mech is an island". Team-play is strongly discouraged, if not even selected against. I passionately hate this idea, and I have a strong feeling that most people who care at all about making this game a pleasant experience for anyone but the most selfish (who, oddly, still enjoy losing) would agree. It is antithetical to a 12v12 team game as it essential makes it a 24-person free-for-all. However, for people who never use comms, have chat turned off, ignore the company commander marks, and don't care about getting help from (or helping) their team, I can understand this being a viable option.


Solution 2

Use match score only in a relative (within the same match) sense and give priority to W/L ratio. Winners gain PSR, losers lose PSR. Here is an idea of what that might look like using Russ' PSR numbers:

Winning team:
player 1 +24
player 2 +22
player 3 +20
player 4 +18
player 5 +16
player 6 +14
player 7 +12
player 8 +10
player 9 +8
player 10 +6
player 11 +4
player 12 +2

Losing Team
player 13 -2
player 14 -4
player 15 -6
player 16 -8
player 17 -10
player 18 -12
player 19 -14
player 20 -16
player 21 -18
player 22 -20
player 23 -22
player 24 -24

Team play is the strongest factor here. The longer that time passes and the more matches that are played, the more people who focus on securing a win for their team will find themselves against other people doing the game, and vice versa for people who don't. After a long enough period of time, the winning team no longer has to feel bitter about "carrying" people with low match score, because they can simply assume that at least those people tried. Likewise, the losing team can assume that everybody put in some effort. Communication is a must for this to be an enjoyable experience. I like this because I like winning, I like to assume that my team is going to try to work as one, and I like using comms. This solution also has the major advantage of removing most ambiguity to what ones needs to do in order to obtain good match score since the primary factor is just “did I win or lose?”


Solution 3

Use match score relatively, but "reward" high-performance loser and "punish" low performance winners. There have been various iterations of this already proposed. Here is one example, again using Russ' numbers. Your 'ideal' numbers may differ, or even include some changes of 'zero', but I think you'll get the idea.

Winning team:
player 1 +24
player 2 +22
player 3 +20
player 4 +18
player 5 +16
player 6 +14
player 7 +12
player 8 +10
player 9 +8
player 10 -2
player 11 -4
player 12 -6

Losing Team
player 13 +6
player 14 +4
player 15 +2
player 16 -8
player 17 -10
player 18 -12
player 19 -14
player 20 -16
player 21 -18
player 22 -20
player 23 -22
player 24 -24

This is almost as good as Solution 2, but makes people who like seeing themselves rewarded and the people they view as inferior punished. I'm not sure that a team should be competing within itself, but whatever. I live in America so I'm used to it. In the end, I cannot convince myself that this will lead to a better or worse result that Solution 2, but rather simply differing in how many green or red arrows are witnessed along the way. If people want Barabbas, I'll halfheartedly shout it with them.


Discussion

The common theme of the three solutions above is that match score is only used to score player against player within the same match. That is because the three systems above are truly zero-sum, and, due to it's (more-or-less) random distribution, match score, as an absolute value, is not.** This must be done in order to make match maker fair. However, there is a big problem: the concerns of casual players wishing to see themselves 'improve' over time playing the game, or with a 'good match' are not addressed. I think there is a simple solution for that. Decouple the tiers used for match making from the 'experience bar' displayed in the in-game UI. People want to see that bar go up. It makes them feel good about themselves, feel good about the game, and feel like they are getting something for an extraordinary effort. On the other hand, we all care about the quality of the game when our boots hit the ground.

Here is one way I can imagine that going:

Match maker: uses only PSR. Period. You guys have the numbers stored on-server anyway. Simply do not bound it. Release your "valves" as you see fit to get people matches in a 'reasonable' amount of time, but keep it just as opaque as you always have. It doesn't matter for the end user as long as they are getting fun games.

Tier Bar: Change it entirely. Make it 1 to X (with X being higher) like this is some sort of role playing game (maybe X could be as high as 100, who care?), and keep evaluating match scores for "tiers" the same way you always have. People can keep feeling like they are 'leveling up', and keep feeling good about themselves, but at the same time not have that mess up the actual match making. I'd wager that a lot of people like to see that bar go up even if they won't admit it.




* I'm not going to address groups here. If you want to complain about groups, there is already a thread where you can do that. This thread is for being constructive.

** There are nits to pick there since match score, over a large enough sampling of both matches and players, makes a more-or-less normal distribution curve, but the fact stands that the sample size within a single match is too small to be useful for that.

Edited by Andrzej Lechrenski, 08 June 2020 - 07:55 PM.

[TheCaptainJZ]

Guys...
Less talking and more number posting, please. Already page 4 and no firm match score proposals. Hopefully Paul will provide current values.

Source for some Rewards information: https://mwomercs.com...-get-xp-and-cb/
Contains XP and CBill values but not Match Score, but might give some idea what is valued. I'm sure someone has reverse engineered Match Score somewhere...

Assign a number to each category. GO!

Edited by TheCaptainJZ, 08 June 2020 - 07:41 PM.

[Kamikaze Viking]

nocomfort, on 08 June 2020 - 06:37 PM, said:

How about this: Reset PSR immediately using the last 25/50/100 games as a player's seed, and let us play in the new tiers that result with the current system while we're waiting for the twelfth of never to freeze over and everyone to agree where the new deck chairs are going to go. Give us something right now so we can at least be satisified that maybe something we propose will be taken seriously.

I remember this was definitely part of my idea at one point. I'm annoyed i forgot to put it into my proposal last week.

Since they are going with the full reset, i just think they need to make sure the PSR accelerator is applied for longer than the 20 matches previously stated. More like 100-150.

[Andrzej Lechrenski]

TheCaptainJZ, on 08 June 2020 - 07:40 PM, said:

Guys...
Less talking and more number posting, please. Already page 4 and no firm match score proposals. Hopefully Paul will provide current values.

Source for some Rewards information: https://mwomercs.com...-get-xp-and-cb/
Contains XP and CBill values but not Match Score, but might give some idea what is valued. I'm sure someone has reverse engineered Match Score somewhere...

Assign a number to each category. GO!

Thinking takes time.

[Billyum]

I personally think that in order to get the most accurate PSR possible, the primary factor needs to be Win/Loss Ratio. I say this because the objective is to win the match (Aside from having fun!). Match score can only account for so much, and as it's been stated by multiple people, there are many ways to inflate your match score to make it not fully accurate.

To those that disagree, I would recommend looking at the in game leaderboard's top players under both the 'Average Match Score' and 'W/L Ratio' tabs. I know that the season just started, but this should be enough of an indicator that having a high match score doesn't equal helping your team win the match.

Edited by Billyum, 08 June 2020 - 07:46 PM.

[Zanotam]

A good matchmaking system with 12 people per team is going to need ~150-200 matches to rank people appropriately. That's just the truth. So maybe accelerate it for the first 50-100 games, but we just gotta accept it's going to take a somewhat active player still like a month to end up close to their rank assuming the rest of the system is ranked correctly and with a full reset it will potentially take 2-3 months with uncertainties of lower games per week players thrown in along with people who maybe haven't seriously tried for years getting back in the swing of things as well.

After that though there needs to be a monthly 'decay' towards the 'base' which pushes everyone's score at the end of the month towards whatever the 'base' point which IIRC is something like 1200 or 1500 in an ELO ranking but is obviously arbitrary somewhere around probably teh bottom of tier 3 to make tiers slightly more meaningful for higher tiers where skill level disparities will matter more.

[Kamikaze Viking]

Andrzej Lechrenski, on 08 June 2020 - 07:38 PM, said:

The Problem*

players who do not play competitively (hereafter referred to as casual players) are being placed in matches with players who do play competitively.

/snip

Great post. I Like your Solution 3. it works very similar to the way ive been thinking.

I think the Values for the amount of PSR shift might need to be scaled back, BUT we need futher details on the Max PSR range and the Tier Thresholds for that (something in the back of my head says PSR range is 0-2300, but i cannot remember where that info comes from)

[Andrzej Lechrenski]

Kamikaze Viking, on 08 June 2020 - 07:48 PM, said:

Great post. I Like your Solution 3. it works very similar to the way ive been thinking.

I think the Values for the amount of PSR shift might need to be scaled back, BUT we need futher details on the Max PSR range and the Tier Thresholds for that (something in the back of my head says PSR range is 0-2300, but i cannot remember where that info comes from)

That is why I think PSR should be an unbounded number. It shouldn't matter to us, personally. It should just be used for match making, and having bounds only hurts that.

[MOBAjobg]

Andrzej Lechrenski, on 08 June 2020 - 07:38 PM, said:

*snipped Long Wall of Text.

Solution 3

Use match score relatively, but "reward" high-performance loser and "punish" low performance winners. There have been various iterations of this already proposed. Here is one example, again using Russ' numbers. Your 'ideal' numbers may differ, or even include some changes of 'zero', but I think you'll get the idea.

Winning team:
player 1 +24
player 2 +22
player 3 +20
player 4 +18
player 5 +16
player 6 +14
player 7 +12
player 8 +10
player 9 +8
player 10 -2
player 11 -4
player 12 -8

Losing Team
player 13 +6
player 14 +4
player 15 +2
player 16 -8
player 17 -10
player 18 -12
player 19 -14
player 20 -16
player 21 -18
player 22 -20
player 23 -22
player 24 -24

This is almost as good as Solution 2, but makes people who like seeing themselves rewarded and the people they view as inferior punished. I'm not sure that a team should be competing within itself, but whatever. I live in America so I'm used to it. In the end, I cannot convince myself that this will lead to a better or worse result that Solution 2, but rather simply differing in how many green or red arrows are witnessed along the way. If people want Barabbas, I'll halfheartedly shout it with them.

*snipped further

Should player 12 be -6 instead of -8, just checking.

Edited by MOBAjobg, 08 June 2020 - 07:55 PM.

[Andrzej Lechrenski]

MOBAjobg, on 08 June 2020 - 07:54 PM, said:

Is player 12 should be -6 instead of -8, just checking.

whoops

Edit: fixed.

Edited by Andrzej Lechrenski, 08 June 2020 - 07:56 PM.

[Rex Gordan]

Andrzej Lechrenski, on 08 June 2020 - 07:38 PM, said:

The Problem*

Before we can work together as a community to fix our problem, we are going to first need to clearly define it. With regard to match maker, I propose that this is the specific issue we need to address: players who do not play competitively (hereafter referred to as casual players) are being placed in matches with players who do play competitively. This is a problem both for the casual players, who have a miserable time due to being either "stomped" or "carried" (neither of which is especially fun), as much as it is a problem for the competitive players who feel they are either being unfairly handicapped or else breezing ......

This post is GOLD! So well written! I like how he breaks it down into options. I like option 2 personally. Let's rally behind Andrzej Lechrenski's options here.