277 replies to this topic

[Death Proof]

crashlogic, on 25 September 2015 - 09:58 AM, said:

Or put it this way, its entirely possible to remain a tier 4 player doing 700 damge every time and never going up because your team can't get more the 4-5 kills

THIS is what really bothers me about the system. While being able to lead your team to victory is a boon, it should not be used as the basis for your PSR. It should improve it, but never lower it. In other words, you should NOT be punished for losing because your team mates refuse to get their sh*t together. You should only be punished for performing poorly (i.e. low match score).

When they first announced PSR I was excited because they suggested that it would not be based on W/L ...going so far as to admonish Elo because it did so.

I was disappointed to find out PSR is just Elo dressed up like a pony.

[An Atlas]

Kira Onime, on 25 September 2015 - 10:09 AM, said:

Usually using alt accounts to post things generally paint you as what you're trying to avoid.

I'm not going to worry about whether someone wants to form a baseless opinion about me.

Death Proof, on 25 September 2015 - 10:14 AM, said:

THIS is what really bothers me about the system. While being able to lead your team to victory is a boon, it should not be used as the basis for your PSR. It should improve it, but never lower it. In other words, you should NOT be punished for losing because your team mates refuse to get their sh*t together.

See that the crux of the issue...

You're placed with guys that are as good as you at "being a team player".

It is shocking how derptastic a good portion of "T1" players are though on a regular basis.

Edited by MechWarrior1086091, 25 September 2015 - 10:18 AM.

[Gnume]

Tesunie, on 25 September 2015 - 09:59 AM, said:

It isn't an XP bar, as it can go down (and should once you are in the tier you should more or less be in). They have a good base, but things need to really change for it to work as intended. Read my above statements.

How does that stop it from being an Experience System? There are Experience based games where you can lose experience and thereby lose levels, ability to equip items, and so on and so forth. I know it's an unpopular thing because gamers these days expect to be handed things easily and then in no way expect to be punished by losing experience or stuff but still, it has been a part of experience based games for a long time. Pen & Paper Dungeons and Dragons for example, had monsters that could level drain a person or things that could cause an experience loss. I've already mentioned EverQuest at one point. In that game you lost experience when you died, even losing levels and abilities. Let's not even get into originally you had to retrieve your corpse to get your items back

[Lugh]

CaptRosha, on 25 September 2015 - 10:45 AM, said:

How does that stop it from being an Experience System? There are Experience based games where you can lose experience and thereby lose levels, ability to equip items, and so on and so forth. I know it's an unpopular thing because gamers these days expect to be handed things easily and then in no way expect to be punished by losing experience or stuff but still, it has been a part of experience based games for a long time. Pen & Paper Dungeons and Dragons for example, had monsters that could level drain a person or things that could cause an experience loss. I've already mentioned EverQuest at one point. In that game you lost experience when you died, even losing levels and abilities. Let's not even get into originally you had to retrieve your corpse to get your items back

That's too hard *waaah*

And My favorite UO online dueling PKers while eating foot soup that you made out of their own or your own body parts...

[LordSkyKnight]

Death Proof, on 24 September 2015 - 06:25 PM, said:

No. I know that was humorous analogy, but the fact is, bad players who continue to play bad and do not learn from their mistakes will probably never get past T5 no matter how many matches they play....unless they eventually learn from their mistakes and actually get better.

The real problem with this system is that it's just Elo, redressed as a player skill rating system.

Since it's meant as a means to place players in solo matches (mostly) it needs to not be so dependent on Win/Loss ratio. W/L should be a factor, but not the basis for determining PSR.

It actually isn't. ELO drops one person's rating and raises another's after a match based on who won. With this system the rating gains outweigh the rating losses, so a person who performs exactly average will eventually be raised to tier 1. THAT is the issue. The system should be constructed such that a person who performs exactly average will stay right in the middle of tier 3. Paul confirms this bias, claiming that it is slight. So the system is WAI. Which means that your tier is as much indicative of your time played as it is your actual skill level. The more skilled pilots will just get to tier 1 faster, and if you don't have a consistent upward trend in PSR, you're probably a below average player.

[Death Proof]

LordSkyKnight, on 25 September 2015 - 11:16 AM, said:

It actually isn't. ELO drops one person's rating and raises another's after a match based on who won.

It's based on Win/Loss, just like Elo. That's all I meant.

[Tesunie]

CaptRosha, on 25 September 2015 - 10:45 AM, said:

How does that stop it from being an Experience System? There are Experience based games where you can lose experience and thereby lose levels, ability to equip items, and so on and so forth. I know it's an unpopular thing because gamers these days expect to be handed things easily and then in no way expect to be punished by losing experience or stuff but still, it has been a part of experience based games for a long time. Pen & Paper Dungeons and Dragons for example, had monsters that could level drain a person or things that could cause an experience loss. I've already mentioned EverQuest at one point. In that game you lost experience when you died, even losing levels and abilities. Let's not even get into originally you had to retrieve your corpse to get your items back

Though I get what you are saying, I think we are talking about two different styles of systems.

I'm talking about "experience" in a system where you are suppose to become maxed level and play "end game" content. Weather you lose experience as a penalty is a somewhat irrelevant side point. However, in that comparison, we also need to compare those game formats. MW:O is PvP only. This means that players are expected to die and lose. In the games you describe, they are all "mostly" PvE, where a player can more or less control their engagement levels, and are expected to surpass and survive the engagements. Most penalties are even removed in their PvP sections of the game, as players have to die and lose for others to win.

What we need in MW:O is a "ranking" system that places people of similar skill together. Right now, it seems like an "experience" (by your terms and several other people's terms) bar, where the goal is to become T1. In actuality, we need some system that places players on their individual skill into a tier, and try to get and keep them there as long as they preform at that level. Right now, we have a "team ranking" system, for a game mode that has ever shifting teammates and opponents.


In a pure PvP game, we need to not focus so much in winning (or K/D). Players are going to die. Teams will have to lose for others to win. There should be no punishment (besides reduced c-bills and a naturally lower match score) for a loss. A loss already typically will result in lower match scores. So even if Win and Loss was removed from the equation, it would still be harder to go up on a loss, just like it would be harder to drop down on a win. This would more accurately evaluate the individual player's skill, compared to the skills of the entire match of players.

[Widowmaker1981]

The major problem with Elo was that it based everything on winning or losing. Do people not see the idiocy of replacing it with a system based almost entirely on winning or losing?

[Water Bear]

El Bandito, on 24 September 2015 - 06:18 PM, said:

You are over-exaggerating. Otherwise there wouldn't be player with thousands of matches under their belt who are still in T4 and T5. Yet there are! Plenty of posts from multi-thousand match vets talking about how low their tier is. Hell, Jo Mallan would have reached T1 by your logic.

Sure, not everyone who reaches T1 is gonna be Proton level, but they are sure as hell better than the majority of the lower tier players. That is true especially right now, at the start of the tier system.

Exactly right. I don't want to name any names but I know of people who have been playing for quite a long time who have surprisingly low tiers.

Although it is true that Paul's post is incredibly vague. "Everyone will reach tier 1 or 2 because they're improving." This thread is already full of both of the most prominent interpretations of that sentence. Of course it seems to me that he's just describing a normal Elo type system, but the guarantee that people will reach T1 or T2 is a pretty strong statement.

[Gnume]

Tesunie, on 25 September 2015 - 11:38 AM, said:

Though I get what you are saying, I think we are talking about two different styles of systems.

Gotcha and follow what you are saying. I can agree with that and that we need a ranking system.

The way the current PSR seems to be calculated though, may still end up more like an exp base than a rank though because as long as you keep performing ok, with some good ones thrown in, you WILL advance.

I mentioned this in a previous post earlier in the thread:

I think a better way to keep PSR fluctuating and more like a Rank, is to just take the average PSR score of your most recent X number of matches. However, the issue may then become one where players purposefully tank some games to lower their current PSR and with it using a smaller sample size than your lifetime matches, then that player can tank it faster and raise it faster. They could combat that by only allowing a tier reduction of 1 (from the max tier you ever obtained) so that a T1 player could only ever drop to 2 and a 2 to a 3 and so on.

Take W/L out of the equation and instead, determine your Tiers based on PSR scores...say Tier 1 players must maintain an average PSR of X else they drop to T2 and so on. Every match has a greater chance to affect your Average PSR if you use a smaller sample size and one that is constantly updated based on most recent games.

[Chemie]

" The more games you play, the better you should become"

This is where the logic falls flat. Sure, a new player should get better after they play but every person will plateau at a different place. Paul's design on PSR says you will continuously get better and better. Reality is that some people will hit a wall quickly and never get better from there. But with PSR as implemented, they will keep going up. The model is design to prove a false assumption.

The analogy is "If you keep taking the SAT, you will get better". Therefore, just take it enough times and you will get 2400 top score. Obviously wrong.

PS. The other issue is that match score is highly correlated to W/L. If you are on the wrong side of a 12-0, you will be low match score (even as good player) because you will not have any kills and zero assists. Conversely, a bad player who is on the right side of 12-0, could get 12 assists just by passing some lasers on each person. So in the end, PSR as implemented is only counting your W/L and W/L is a team outcome, often independent of individual contribution.

Edited by Chemie, 25 September 2015 - 01:30 PM.

[ArchMage Sparrowhawk]

So...I don't understand how he says we don't play with a team mindset? My KDR is so lopsided because I play a team minded scout most of the time. I always play for the win versus the kill, and I don't know how I could be more team oriented.

does he just mean deathballing? I'm always getting 7+ assists, maybe 1 or 2 kills, 3~600 damage on average. How does that translate into Lone Wolfing it?

[Johnny Z]

There are alot of replies continueing on this topic that misunderstand alot of the issues. Please take the time to read the entire topic and then realize the only one making any sense is me. lol

[Sjorpha]

ArchMage Sparrowhawk, on 25 September 2015 - 03:14 PM, said:

So...I don't understand how he says we don't play with a team mindset? My KDR is so lopsided because I play a team minded scout most of the time. I always play for the win versus the kill, and I don't know how I could be more team oriented.

does he just mean deathballing? I'm always getting 7+ assists, maybe 1 or 2 kills, 3~600 damage on average. How does that translate into Lone Wolfing it?

If you mean Paul, what he's saying is that is if you play in a way that makes your teams win more often it will improve your PSR faster than if you don't.

He's not saying what kind of exact method you should use, that's up to you as long as it wins matches.

[Random Carnage]

Reading these posts I just shake my head.

A little knowledge really is a dangerous thing.

Live it up while you still know everything.

[Tesunie]

CaptRosha, on 25 September 2015 - 12:04 PM, said:

Gotcha and follow what you are saying. I can agree with that and that we need a ranking system.

The way the current PSR seems to be calculated though, may still end up more like an exp base than a rank though because as long as you keep performing ok, with some good ones thrown in, you WILL advance.

I mentioned this in a previous post earlier in the thread:

I think a better way to keep PSR fluctuating and more like a Rank, is to just take the average PSR score of your most recent X number of matches. However, the issue may then become one where players purposefully tank some games to lower their current PSR and with it using a smaller sample size than your lifetime matches, then that player can tank it faster and raise it faster. They could combat that by only allowing a tier reduction of 1 (from the max tier you ever obtained) so that a T1 player could only ever drop to 2 and a 2 to a 3 and so on.

Take W/L out of the equation and instead, determine your Tiers based on PSR scores...say Tier 1 players must maintain an average PSR of X else they drop to T2 and so on. Every match has a greater chance to affect your Average PSR if you use a smaller sample size and one that is constantly updated based on most recent games.

Seen as you really can't go down (or really stay even) on a win, and it's really hard to go up or even stay even on a loss...
Seen as you also seem to gain more PSR on a win than on several losses...
Yes. It is going to probably fill in more like an Exp meter, and as people have stated it will be more likely that more and more players will be Tier 1, rather their actual skill are or not.

As it is, I'm finding the system very easy to manipulate, and I'm not even trying to do it. If I play X mech, my PSR tends to go up. If I switch to Y mech, it probably will go down. Right now, I can adjust my PSR between Tier 3 and Tier 4 at just about any given moment on a whim. (Not saying I am, just that I can.) I also have full confidence that I could even, through such manipulations, even get up to Tier 2 if not Tier 1 if given enough time playing with X mech. Even though I don't believe for a moment I'm Tier 1 worthy skill level or material. At best, I gauge myself as upper Tier 3, maybe lower Tier 2 if I really pushed myself in a "normal" manner. I started off with a very high Tier 4 rank, almost Tier 3... (Sadly, with this statement, I don't see anyway to avoid players being able to game their tier rank if they wanted to. I can't think of any stop gap or rule that could prevent this abuse if someone wished it.)


PSR I believe should be a score system over all matches played, not just the last X matches. But I don't mean as taking from an average match score across all matches, but as a score that is earned and can move up and down based on each individual match (like how it is currently). For an example:
Each tier has a score value it is assigned. If you are within those values, you are that tier.
Each player accumulates and loses points based on their performance.
If a player does poorly, over time they will lose enough PSR to drop a rank. Same in reverse.

However, to calculate this PSR number, I think wins and losses need to be removed. It should be based on individual match scores. As a loose example:
- Score 100 points or less, PSR drops dramatically.
- Score 150 points (or some other lower than average consideration), PSR drops slightly.
- Score 200-250 points, no change. (Or what is considered an "average" match score.)
- Score 251-300, a slight positive PSR increase.
- Score 301+, a larger PSR change.

Then, win or lose, players will drop if they preform poorly for too long, stay the same if they are "average" for the match, or improve if they did good or exceptional in many matches. Thus, if you are in the tier you basically belong in, you end up preforming more average match scores to remain in that tier. (And, between the occasional good matches being balanced by the occasional bad matches, it should remain within the tier bracket if their overall skill isn't improving.)


PSR should require no "sample size", and instead should be a score that is continually added and subtracted to with each consequential match. Pure number of matches should get you no extra benefit, as once you hit your skill level (for that moment at least), your match scores should average out overall, leaving you with very minor changes to no changes. (Recall that, in theory of course, as you progress up in tiers, your competition and game play should become more challenging. This should lead to a harder time to maintain an "average" match score if your actual skills are not up to that tiers challenge.)

(For PSR to accurately work as well, Match Score probably would need to be revamped to reward other actions besides just damage related activities. Such as base capturing on those mission objective modes, scouting, ECCM, Spotting, etc. Right now, none damage related skills don't really add up for me. I think if you get most all of the non-damage related actions done, it equals around 100 match score. Where as damage (which I admit is important) can (in theory) give you gains of over a 1,000 to your match score.)


I'd just like to add one final statement. Your Tier should be just an indicator of your game play style and skill level. Should not have any negative intonations based on your Tier. (AKA: Tier 5 are not "the steering wheel under hive". More than likely, they are the "party brigade" who are just out to have fun in their game, and they don't care who happens to wear the lamp shade at their party. ) Tiers just represent a different level of game play, as much as it should represent level of player skill. Low tier doesn't always mean low skill exactly (just maybe a more fun play style and less "optimized" mechs). (There is probably a reason I started in upper Tier 4... I like mech experimentation, and I've been working on leveling new mechs for a while now.)

Random Carnage, on 25 September 2015 - 06:56 PM, said:

Reading these posts I just shake my head.

A little knowledge really is a dangerous thing.

Live it up while you still know everything.

But...I do know everything. Just ask me, I'll tell you.
(No he don't. Don't believe a word this guy says.)
(Quite you. I'm not talking to you.)
(I'm yourself... You're showing your crazy again...)
(>.>)

Edited by Tesunie, 25 September 2015 - 08:32 PM.

[Jon Gotham]

Yeonne Greene, on 24 September 2015 - 05:34 PM, said:

On the other hand, linking it so closely to wins does provide an incentive to play as a team, since the best way to raise your PSR is to win and the best way to win is to play together.

Which is ironic as in the solo q nearly no one knows what a team is even!