96 replies to this topic

[ambosen]

A pretty good start would be just considering how your actions within a game might affect other people on your team. Case in point: I saw a match lost today because some player in a Rifleman-IIC threw a fit following someone shooting at several lights he'd gotten himself surrounded by, getting at least one of them off his back, then he turned around and just start randomly opening fire on his own team's flanks.

This left a good chunk of his own team facing the wrong direction when the opposing team mounted a counter attack.

[NeirSolon]

I see this is a difficult topic to get one's head around.

To try to get to the heart of the matter, I have developed the 'Great Player Diagnostic Test':

https://mfbc.us/m/cwqmgk

Use this tool and apply it to a player you know and love to determine if they are a truely great player.

[Horseman]

Kubernetes, on 20 July 2020 - 11:09 AM, said:

Good grief, ANOTHER ONE.

No, no you're not. The only things in which you are superior to him in this game are dying and losing.

Just to drive the point home: https://leaderboard....101%0D%0ABrauer

Sjorpha, on 20 July 2020 - 12:43 PM, said:

A good player is a player who reliably wins more than he loses, especially if he also wins when matched against other strong players. Another way to put it is that a good player will, on average, increase the winning chances of teams he is in.

A bad player is a player who, on average, decreases his team's chances of winning.

And finally a scrub is someone who tries to argue against the above, who makes excuses for their own performance and who tries to bring their own subjective values into the argument (such as calling things cheese or raving about irrelevant moral standards like "honor" or "no-skill builds" and so on.)

Amen.

Nesutizale, on 20 July 2020 - 10:35 AM, said:

Conquest:
As long as its easier to kill the enemy then takeing the points...whats the point?
I think the only way to solve this is when you can get to the 750 points faster by capping then it takes to destroy the enemy.
Basicly the only way to archive that is by haveing Dropdecks or respawns.

If you ignore the objective and the opfor doesn't, you will lose by objective. I've had a match just yesterday where because one of my team's cappers tarried for a second too long on a previous cap, we lost a match that would otherwise be a win.

Quote

Escort (yah I know its not in atm)
While I liked the idea there where things that allways bothered me.
1) Let it be a mobile HQ with an escort of tanks and let the tanks add their firepower to the defance
2) If you can't program a good pathfinding system for the AI make a map that supports the mode and have predetermed pathes you make sure the AI will follow correctly.
3) An option to draw the escorting players away. Maybe in the way that the players know the convoys path beforehand. Then the enemy team can have false radar blips at their disposal they can drop to create false radar readings and lure out the escorting players. Maybe let every UAV appear as an enemy mech.

Escort was bad in that it didn't give players info to work with. That the escorting team had no idea where the Atlas would go meant they couldn't plan around it as a result.
As I wrote a few times before, there are potentially two other modes that could be built on its' bones:
* A cap mode with a mobile capture target. That means both sides don't want to destroy it (or perhaps you could just have "destruction" halt the target, making it easier to capture for the attackers.
* A "convoy escort" mode where you have to get X periodically spawning convoys to their target locations. That would require some measure of strategy and map control to win, making it potentially more interesting.

[Sjorpha]

Nesutizale, on 20 July 2020 - 01:05 PM, said:

@Sjorpha
I think it would be helpfull to say what makes the good player win more. What is he doing to increase the winning chance of his team?

That depends.

For studying and learning the game it's definitely very helpful to analyze what makes a player good or bad.

But for matchmaking it's completely irrelevant, and you're reducing the predictive accuracy by mixing in secondary stats. So for matchmaking it's actually very unhelpful.

Some of the problems in these discussions arises because people tend to conflate those two topics.

Edited by Sjorpha, 21 July 2020 - 12:07 AM.

[Horseman]

Nightbird, on 20 July 2020 - 01:42 PM, said:

To be clear, no one thinks earning more Match Score = helping the team more.

Hmm, no. There are degenerate ways to farm MS while being of zero benefit to your team, yes. But by and large the score correlates with activities that advance your team's goal.

Quote

Many activities will help the team win, such as good communication (concise remarks, good strategic suggestions) which earns 0 MS,

Wins are rewarded in MS.

Quote

killing enemy efficiently though the CT which earns far less MS,

Kills and KMDDs are rewarded in MS. If you do both, you get the Solo Kill bonus.

Quote

distracting the enemy at an opportune time as a light which earns 0 MS,

Flanking and Hit and Run are rewarded in MS.

Quote

tanking at the right time to keep allies shooting at the enemy alive for longer which earns 0 MS,

Protected and Lance In Formation bonuses are rewarded in MS. If you tanking helps your team, it will come out in your winrate (and wins are, again, rewarded in MS)

Quote

Many MS rewarding activities actually hurts the team, such as inefficiently killing enemies to earn more damage and thus MS but allowing them to do more damage to your teammates,

Being deliberately inefficient is bad, yes. Still, any damage on target is better than no damage on target and may allow a more efficient teammate to land the killshot.

Quote

capping when the match moved into the fight phase making the combat 12vs11 seriously tilts the odds against your team,

Unless your mech is in position to cap and doing so advances your team more than it would if you went and got yourself killed. You forget that capping the base provides a distraction to the enemy team and forces them to move some of their mechs out of the fight.

Quote

hiding and preserving your mech until the second phase of the fight so that enemies are weakened while you are fresh also seriously hurts your team's chances of winning.

Depends on the mech. Brawlers should preserve their mechs while advancing, because if they are getting destroyed from range they won't have a mech left to engage and do their job up close.

Black Caiman, on 20 July 2020 - 05:07 PM, said:

I think whats been missing from the conversation is that a truly good player will make the best out of the entire scenario theyre given. That includes the mech chassis, mech build, team mates, map, game mode, and game score. A good player also knows their strengths and weaknesses and does what they can to minimize their weaknesses getting exploited, and maximizing their strengths. A good player will play the mech/build to its max potential regardless if its laser heavy, streaks, LRMs, brawler, etc, etc. A good player will also attempt to maximize his team whether that means tanking at an appropriate time (as Nightbird mentioned), providing covering fire, or being a distraction. A good player will also make good decisions based on the scenario presented to them. That might mean deciding that fighting the opponent has minimal chance for success, but playing the cap game has a moderate chance for success. That might mean you need to buy time for your team mates to fulfill an objective, or in some cases it may mean knowing you have to sacrifice your teammates so you can get the job done. Piloting and aim is only a small part of the puzzle. Decision making, map knowledge, positioning, knowing your opponents, making sound strategic calls, and many other attributes all play factors.

Amen. Ultimately the goal is to win, and you should do whatever it takes to get there from your current tactical position.

[OneTeamPlayer]

Players who are "good" by virtue of the matchscore system will literally never stop defending the matchscore system.

Heck, you can barely pull enough teeth to get players to admit that Kills is actually a team stat and if we wanted an individual stat we'd list KMDD or Solo Kills.

Not their fault though, a games directives and community are often subtly formed by the stats chooses to prioritize and display.

MWO chooses to display stats that are based on individual "performance" thus- modern MWO.

[Nightbird]

Horseman, on 21 July 2020 - 12:23 AM, said:

Hmm, no. There are degenerate ways to farm MS while being of zero benefit to your team, yes. But by and large the score correlates with activities that advance your team's goal.
Wins are rewarded in MS.
Kills and KMDDs are rewarded in MS. If you do both, you get the Solo Kill bonus.
Flanking and Hit and Run are rewarded in MS.
Protected and Lance In Formation bonuses are rewarded in MS. If you tanking helps your team, it will come out in your winrate (and wins are, again, rewarded in MS)

Being deliberately inefficient is bad, yes. Still, any damage on target is better than no damage on target and may allow a more efficient teammate to land the killshot.
Unless your mech is in position to cap and doing so advances your team more than it would if you went and got yourself killed. You forget that capping the base provides a distraction to the enemy team and forces them to move some of their mechs out of the fight.
Depends on the mech. Brawlers should preserve their mechs while advancing, because if they are getting destroyed from range they won't have a mech left to engage and do their job up close.
Amen. Ultimately the goal is to win, and you should do whatever it takes to get there from your current tactical position.

Win are rewarded in MS - Yes but specific beneficial actions are not, that's the whole point of MS right? Reward the action rather than the result?

Kills and KMDD you can get regardless, but if you shoot 5 components off and farm an extra 500 damage from one assault mech, you earn a ton more MS.

When you get 4 mechs chasing you, turning the match into 11vs8 at an opportune time, is that help to your team proportional to the measly hit and run MS you earn?

All the other examples are still valid.

Match score is terrible, it doesn't take into consideration when something should be done versus when it shouldn't, and in no way does it give a proportional reward for actions that help the team.

Edited by Nightbird, 21 July 2020 - 06:16 AM.

[Brauer]

OneTeamPlayer, on 21 July 2020 - 05:09 AM, said:

Players who are "good" by virtue of the matchscore system will literally never stop defending the matchscore system.

Heck, you can barely pull enough teeth to get players to admit that Kills is actually a team stat and if we wanted an individual stat we'd list KMDD or Solo Kills.

Not their fault though, a games directives and community are often subtly formed by the stats chooses to prioritize and display.

MWO chooses to display stats that are based on individual "performance" thus- modern MWO.

I do a pretty good job of farming match score AND I've been for a more WLR based system. But do go on.

Also, kills, KMDDs, and solo kills are all somewhat flawed stats. If a player is sandblasting the front armor of a mech to earn the KMDD, but I get the kill by snipping off a ST from behind, who actually did the most to get that kill? It's not uncommon to miss out on solo kills or KMDDs when you are efficiently taking out the opposition. Good players do tend to stack kills, KMDDs, and solo kills more than bad players, but prioritizing KMDDs and solos over kills isn't really going to be the best measure of skill or a of a player's contribution to a win.

[Horseman]

OneTeamPlayer, on 21 July 2020 - 05:09 AM, said:

Players who are "good" by virtue of the matchscore system will literally never stop defending the matchscore system.

Better a flawed system than none whatsoever.
But by and large, MS does revolve around actions that either deny the enemy their assets or preserve those controlled by your team. While it's not a 1:1 correlation with WLR, the two are strongly linked together - almost as if generating higher match score (without deliberately engaging in degenerate farming strategies) does, on average, help your team win.

Nightbird, on 21 July 2020 - 06:10 AM, said:

Win are rewarded in MS - Yes but specific beneficial actions are not, that's the whole point of MS right? Reward the action rather than the result?

Which, given VOIP as a factor, may be occurring entirely outside of the game's systems.

Quote

Match score is terrible, it doesn't take into consideration when something should be done versus when it shouldn't,

The players themselves don't always know for sure, so how would an overgrown pocket calculator judge that? It wouldn't.

Quote

and in no way does it give a proportional reward for actions that help the team.

That, on the other hand, is fixable. Talk to the guys working on the PSR rebalance proposal and see what comes out of that.

Edited by Horseman, 21 July 2020 - 07:27 AM.

[Kubernetes]

My two most often-run heavies these days are the IV-4 (MRMs) and the Top Dog (MPLs). Both are very survivable, can pump out dps, and can reliably kill. The difference is that the IV-4 has to (and can) throw out significantly more damage to achieve the same result as the TD. If I perform similarly in each mech (in terms of killing, surviving, etc), the IV-4 will always put out higher MS. It kinda sucks that you can run through a match hitting CT every time with precision and lose out on MS because you were too efficient-- you can lose out on KMDD and solos too if some teammate is dropping ATM30s on your target while you core him.

[Marquis De Lafayette]

Kubernetes, on 21 July 2020 - 07:34 AM, said:

My two most often-run heavies these days are the IV-4 (MRMs) and the Top Dog (MPLs). Both are very survivable, can pump out dps, and can reliably kill. The difference is that the IV-4 has to (and can) throw out significantly more damage to achieve the same result as the TD. If I perform similarly in each mech (in terms of killing, surviving, etc), the IV-4 will always put out higher MS. It kinda sucks that you can run through a match hitting CT every time with precision and lose out on MS because you were too efficient-- you can lose out on KMDD and solos too if some teammate is dropping ATM30s on your target while you core him.

Those are the exact two Heavies I have exclusively run this season and your assessment is what I have been seeing as well (not that you need my agreement, your play is better than mine). My MS is definitely better in the IV, but the one thing the TD has is the ability to reliably finish enemy mechs...which prevents some enemy damage from getting pumped onto you or you teammates....this aids in terms of winning and lasting longer in matches (the less damage you take, the more damage you can inflicts....which helps MS). Overall, (imho) they are both strong for the current environment, just the IV will probably still score a bit better. I switched back to the TD primarily went the matchmaker got tightened and more tier 1 piloted lights appeared (I just like the TD better in those matchups). Personally, I am in the camp that thinks the MS calc should slightly nerf damage and buff kills, KMDD and in my view solo kills should get a bigger buff. The top dog could score better with those metrics changed. It’s a good mech that makes a positive difference in matches.

[Gagis]

Winning, winning makes a good player.

All aspects of good performance either add up to winning, or are false.

[Nightbird]

Gagis, on 21 July 2020 - 08:21 AM, said:

Winning, winning makes a good player.

All aspects of good performance either add up to winning, or are false.

Exactly.

[Kubernetes]

Marquis De Lafayette, on 21 July 2020 - 08:18 AM, said:

Personally, I am in the camp that thinks the MS calc should slightly nerf damage and buff kills, KMDD and in my view solo kills should get a bigger buff.

Thing is, KMDD and solo kills themselves are a problem here because they too rely on application of damage. You can hit CT every time with your Top Dog, but if I'm standing off to the side with an ATM Veagle you aren't getting either of those.

Edited by Kubernetes, 21 July 2020 - 08:26 AM.

[Horseman]

Kubernetes, on 21 July 2020 - 08:25 AM, said:

Thing is, KMDD and solo kills themselves are a problem here because they too rely on application of damage.

It's a question of incentivizing accurate damage a bit more than spam. Can't fix everything.

[R Valentine]

Horseman, on 21 July 2020 - 09:21 AM, said:

It's a question of incentivizing accurate damage a bit more than spam. Can't fix everything.

Putting a greater emphasis on rewarding kills isn't the answer. I've seen many a match lost by idiots tripping over each other to kill one mech out of position and completely forfeiting their own position in the process. Sometimes leaving a kill to your teammates is the correct answer, and there's no way the match score system could know that. The game obviously knows what components you've hit over the course of the match, however, and how much damage a target took before it died so those could be factored in. A kill that took less damage = more match score. CT hits = more match score. Those kinds of things.

[Marquis De Lafayette]

Kubernetes, on 21 July 2020 - 08:25 AM, said:

Thing is, KMDD and solo kills themselves are a problem here because they too rely on application of damage. You can hit CT every time with your Top Dog, but if I'm standing off to the side with an ATM Veagle you aren't getting either of those.

My thought here is really that the TD has more opportunity to solo kill with little damage....back CT a heavy, Killing a light/ medium (mean you don’t get much damage if you one shot a light, but it’s valuable to get a good light pilot dead early) . So, if solo kills in particular were buffed...a mech that kills often with lesser gross damage can score better.
Again...it doesn’t fix everything, but could be an improvement.

Edited by Marquis De Lafayette, 21 July 2020 - 09:33 AM.

[Dr Cara Carcass]

HolyGrail101, on 18 July 2020 - 05:34 PM, said:

[mod]Split discussion, origin here: The Great Psr Prophecy (With Graphs!)[/mod]



I disagree, I think that only riding around in the an OP Mech with OP guns, fine tuning your graphics settings outside the UI and or using scripts in order to speed fire your weapons is pretty pathetic and most of the "Competition" level and Faction players I hear boasting tend to do at least 2 out of those 3 things. Being High on the kill zone only matters if you can do it without Macro's, scripts and technical non cheating. Most units only seem to be "Runner's up" to real Comp play until they have to play for real.

As for your high level Comp play BS excuse of coarse 1 Guy controlling an entire unit will be more effective than 1 guy herding 12 cats in PUG play. Duh. It doesn't make other PUG play tactics invalid, everyone can already shoot everyone all game 17000 times or more like I have. I play for fun, feel free to be the 190th best guy in a video game that at over 2500 hours into I admit is kinda niche. You might be a better shot than me but I'm still a better pilot. Piloting doesn't script well I assume.

Edit: I forgot about ping bouncing, some Faction players like that too it seems.

Laughable.....

[VonBruinwald]

VonBruinwald, on 20 July 2020 - 02:50 PM, said:

Someone who enables others to have fun.

Gagis, on 21 July 2020 - 08:21 AM, said:

Winning, winning makes a good player.
All aspects of good performance either add up to winning, or are false.

The difference between a "good player" and someone who is "Good at the game",

Or alternatively, the difference between playing for fun and playing to win.

[My Lord and Saviour Jesus Christ]

OneTeamPlayer, on 21 July 2020 - 05:09 AM, said:

Players who are "good" by virtue of the matchscore system will literally never stop defending the matchscore system.

What metric would you prefer? How long people can sit on a cap? Missiles shot down? How high they can pee?