Should Skill-Based MM really be a thing?

[Destroyer_Zeka]

 This topic is solely my opinion on the matter. If you agree/disagree with me, please constructively discuss down below  

 

In short, my answer is no. While it has come to be apparent, WoWs is kind of starting to become just like WoT with the amount of roflstomp matches going on, especially with most schools starting to get out and all the casual weekend players playing nonstop for a couple months. Anyways, the idea has been thrown to solve this issue by implementing a skill based MM system, which sounds like a good idea on paper, but with a little more thought, actually has lows of major flaws.

 

1.) Long Queue Times: While this one is a little self explanatory, one issue with the idea of SBMM is the significant increase in queue times. Right now I can jump into my Udaloi and hit battle, then get a match in about 10 seconds. if I were to be placed into matchups based on skill, then that 10 seconds would probably stretch into a couple of minutes while the MM looks for enough people of my "skill" to match me up with, and can be even longer depending on time of day, weather, and other things. Would you rather wait 10 seconds for a match or 5 minutes for one?

 

2.) No "Fair" method: Another issue with implementing SBMM is the lack of ways to do it. Lets say WG decided on implementing a "Skill Number" that is the result of a whatever-formula they decide on, the problem is that if people want balanced games, then they suddenly have to worry about that number. They're forced to worry about their stats and performances they wouldn't otherwise care about in order to keep getting balanced games. Otherwise casual weekend warriors suddenly are forced to "git gud" if they want to have a prayer of getting a somewhat balanced game, and to people who just play to have a good time, that is really a very unfair situation. Its a case of screwed if you do, screwed if you don't. This situation is the opposite of fair. So what if WG decides to use WTR as the skill measurement? Same flaw. What if they hide the skill numbers from the player, as a sort of "Out of sight, out of mind?" Another problem with that is the possibility on unintentionally inflated stats. If someone is playing in a division and wins a whole bunch of games, their sudden rise in WR can be interpreted by the game as an "increase in skill" and thus lead that player into a skill level they really aren't ready for.

 

3.) Can games be TOO balanced? In short, yes. If the MM takes player skill into account, the matchups will be quite balanced, which on paper is a good thing. But in practice, the lack of skill variation usually leads to a stalemate; neither team can really gain an edge over the other, and that can draw out games to be a lot longer then it needs to be. While WoWS battles are capped at 20 minutes, not everyone has the time and/or patience to sit there and play for the full 20 minutes every single battle. Take me, for instance. I do consider myself an above average player, and I do enjoy balanced games, but I draw the line at every game lasting half a year before ending by timer. I don't mind a 20 minute match every here and again, but I don't have the patience to sit there for 20 minutes every game while both sides just kinda sit there and neither team gains any sort of advantage, and thus both teams just kinda sit there and do...well....not much of anything. That's just me, but I can guarantee that there are plenty of others out there who don't have the time, patience, or energy to spend the full 20 minutes every single battle. 

 

4.) The game is not competitive: WoWS is not meant to be a really competitive game. It's meant to be a place online where fans of naval warfare can go and sail on historical vessels and fight other players, not to be a place where everyone in every match is a full on tryhards. Skill based MM will only increase tension between teams, as neither wants to lose and risk losing a skill rank or whatever. This sort of thing kind of ties back to #1, but the difference here is that the increased stakes will needlessly and pointless increase tension, which can and will only lead to the salt miners' paychecks being quintupled

 

5.) Newer players will never improve: Think on that, if skilled players are only matched up against other skilled players, then new players are only matched against other new players, and nobody on either team has any clue what they're doing. How do you expect them to learn like that? If nobody in those kinds of matches have any sort of experience, who's there to really teach the newer players how to get better? I know by hard experience that one of the best teachers in the world is getting your butt handed to you. But in a skill based MM, the noobs are only matched against other noobs, and nobody has the slightest clue as to what they're really doing, who's gonna be the one to teach them the "error of their ways" at least for that battle? Knock, knock, who's there? NOBODY!

 

6.) There is already a GM w/ SBMM: It's called Ranked. While the current Ranked system is honestly just a weeee bit flawed, Ranked is probably the closest you'll ever have to a SBMM. Live with it

 

These are just the ones I thought of right off the top of my head. I know there are plenty of others, and I'll be sure to add those later, and I will look at both sides of the coin if you show it to me......

[Estimated_Prophet]

More thought out than anything I've considered; which has pretty much been no more complex than; 'based on what others have said; it seems like skilled based MM sounds like it won't work without causing more problems than it solves.'

 

[DeadManxDan]

If they just gave some kind of competitive tier 8 or 10 MM based matchmaking mode or clan battles no one would care about randoms not being MM balanced.

However I will say you will never get rid of the never ending one sided stomps or have legit balanced play without skill based MM in randoms.
 

[Destroyer_Zeka]

 

However I will say you will never get rid of the never ending one sided stomps or have legit balanced play without skill based MM in randoms.

 

I think you missed the entire point of the article 

[DeadManxDan]

I think you missed the entire point of the article 

 

I didn't read it.  I'm just adding my thoughts on skill based MM.

[E25_Syndrome_Aito]

Did you include it or not.

 

Did you know ?, that even with ' skill based MM '

There are still a lot of variables to take in account.

You will never get ' balanced MM ' like that.

 

Depends, but ( most ) Human beings can variable by itself,

so greatly just by its feeling / mood and whatever alone that,

this may lower or even increase the average skill level.

Some may not even try anymore and just trash their battles,

were they skilled or not.

 

There is that, personal thoughts are that ' skill ',

should not really come in to the factor,

when creating teams, due of various of reasons.

Especially in low population servers,

such as these NA servers where I have to be as well.

 

' Skill based MM ' would not work here well, this is the truth.

EU server however, could be possibility. ( all clock times )

 

Also the ' point 6 ' is suitable enough already,

and with that also as you mentioned there are flaws.

 

 

 

 

 

 

[aethervox]

There already is skill based MM.

It is called placing more good players on one side of the battle.

It is called placing more bad players on the other side of the battle.

 It is called not having any accuracy on well aimed volleys.

 It is called the enemy gets accuracy with well aimed volleys.

 It is called so many roflstomp battles that the MM is a joke.

The MM is not random, never has been, never will be.

 WG 'arranges' things.

Isn't this a polite way to say what is really going on?

 

[SergeantHop]

There already is skill based MM.

It is called placing more good players on one side of the battle.

It is called placing more bad players on the other side of the battle.

 It is called not having any accuracy on well aimed volleys.

 It is called the enemy gets accuracy with well aimed volleys.

 It is called so many roflstomp battles that the MM is a joke.

The MM is not random, never has been, never will be.

 WG 'arranges' things.

Isn't this a polite way to say what is really going on?

 

 

 

I've had plenty  of balanced, tense, very close matches. If matchmaking was truly rigged like you say, these wouldn't happen. Where is your proof that random matches are not indeed random?

[m373x]

Bait or not, I'm gonna take it.

 

1.) Long Queue Times: While this one is a little self explanatory, one issue with the idea of SBMM is the significant increase in queue times. Right now I can jump into my Udaloi and hit battle, then get a match in about 10 seconds. if I were to be placed into matchups based on skill, then that 10 seconds would probably stretch into a couple of minutes while the MM looks for enough people of my "skill" to match me up with, and can be even longer depending on time of day, weather, and other things. Would you rather wait 10 seconds for a match or 5 minutes for one?

 

Queue times longer than 10 seconds? The horror.

 

2.) No "Fair" method: Another issue with implementing SBMM is the lack of ways to do it. Lets say WG decided on implementing a "Skill Number" that is the result of a whatever-formula they decide on, the problem is that if people want balanced games, then they suddenly have to worry about that number. They're forced to worry about their stats and performances they wouldn't otherwise care about in order to keep getting balanced games. Otherwise casual weekend warriors suddenly are forced to "git gud" if they want to have a prayer of getting a somewhat balanced game, and to people who just play to have a good time, that is really a very unfair situation. Its a case of screwed if you do, screwed if you don't. This situation is the opposite of fair. So what if WG decides to use WTR as the skill measurement? Same flaw. What if they hide the skill numbers from the player, as a sort of "Out of sight, out of mind?" Another problem with that is the possibility on unintentionally inflated stats. If someone is playing in a division and wins a whole bunch of games, their sudden rise in WR can be interpreted by the game as an "increase in skill" and thus lead that player into a skill level they really aren't ready for.

 

Plenty of fair methods out there. Since when is caring about your performance and trying to improve a bad thing or a thing of the past, this is the single most important thing that drives humans to improve. A good player will have balanced teams, a bad player or a weekend warrior will have balanced teams as well. Solo and party mmr exist.

 

3.) Can games be TOO balanced? In short, yes. If the MM takes player skill into account, the matchups will be quite balanced, which on paper is a good thing. But in practice, the lack of skill variation usually leads to a stalemate

 

There is skill variation in team based skill based matchmaking. We are talking about players in the team being equal or similar in skill against the players on the other team, not all the players being of equal skill. You can have players ranging from 1200 to 1600 for example.

 

4.) The game is not competitive: WoWS is not meant to be a really competitive game. It's meant to be a place online where fans of naval warfare can go and sail on historical vessels and fight other players, not to be a place where everyone in every match is a full on tryhards. Skill based MM will only increase tension between teams, as neither wants to lose and risk losing a skill rank or whatever. This sort of thing kind of ties back to #1, but the difference here is that the increased stakes will needlessly and pointless increase tension, which can and will only lead to the salt miners' paychecks being quintupled

 

You literally get 2 teams fighting each other with announcer reminding you "The enemy is about to win", with a big [edited]sign saying victory/defeat at the end of a match with music playing accordingly and 4 different tabs detailing your solo performance and comparing it to the performance of your teammates and enemies, the game is competitive in mind.

 

5.) Newer players will never improve: Think on that, if skilled players are only matched up against other skilled players, then new players are only matched against other new players

 

This is where I am certain that you don't know how a skill based MM works. Newer players are not immediately allowed into using skill based MM obviously, and when they reach a point where they are allowed they go through calibration matches to determine where they stand.

 

6.) There is already a GM w/ SBMM: It's called Ranked. While the current Ranked system is honestly just a weeee bit flawed, Ranked is probably the closest you'll ever have to a SBMM. Live with it

 

I have to agree with this one, do not expect anything from WG or you will be disappointed.

 

[1SneakyDevil]

Ranked is not a skill based mode, there are plenty of players that get lucky and get carried up to the highest ranks.

[Goodwood_Alpha]

I don't see why there couldn't be a ladder system, similar to how StarCraft does it. Then again, populations would need to be somewhat larger for it to be effective.

 

Which is probably WG's cue to get busy actually promoting this bloody boat game.

[Taylor3006]

 

I've had plenty  of balanced, tense, very close matches. If matchmaking was truly rigged like you say, these wouldn't happen. Where is your proof that random matches are not indeed random?

 

This has not been my experience at all, as a matter of fact, just the opposite. Most matches are lopsided affairs with either my team steamrolled or the enemy. Close matches are fun and I can remember them well but the reason I do so is they are so far and few between. From previous threads on the subject I believe you are probably in the minority of player's experiences with the game. As far as random matches not being random, I wouldn't go that far and agree with you that making a claim like that needs proof. BTW the only exceptions to the steamrolling affairs for me have been in Ranked battles. Almost to a game they have been close, exciting matches that were fun to play whether my team won or not. If you exclude the whole "lose a star" aggravation on a loss, I would say that the most fun I have in this game has been in Ranked battles.

[Mahrs]

I suspect skilled based MM would be viable only if it worked to apportion the same number of like skilled person across the entire skill spectrum on each team.  Trying to stack teams with Captains of equal skill level would be difficult in some instances AND it becomes self defeating over time.  Supposing it works perfectly, than nearly everyone's performance will become "average" within their respective group over a long enough time horizon.  How do you distinguish when everyone becomes average?  Does everyone end up back in the same MM pool we have now.  

 

Part of me suspects there is some balancing that MM already attempts in this regard.  As an idea, to get a taste of this, you could start with Divisions.  Consider this: Divisions would need to be matched against another Division with same (+/- 1%) average WR, and a not greater than 20% difference in total number of games played.  We could then observe how long those divisions queue on average AND the results from those battles.  Since some/many people DIV to farm WR and salty seal pup tears, it would be interesting to see the results of like skilled DIV being constantly pitted against one another.  

 

Somehow, I don't think that'll be a popular idea....

[Taylor3006]

Ranked is not a skill based mode, there are plenty of players that get lucky and get carried up to the highest ranks.

 

Ranked is probably by far the superior test of a player's skill. More people are carried in Random battles than in Ranked I would suppose. The playing field in Ranked is at least fairly level compared to Random play, no stupid -2/+2 matchmaking. There is much more teamwork involved in Ranked, least that has been my experience.

[aethervox]

 

I've had plenty  of balanced, tense, very close matches. If matchmaking was truly rigged like you say, these wouldn't happen. Where is your proof that random matches are not indeed random?

 

ROFL.

 As if.

 The reality is the majority are one sided 'roflstomps'.

Balanced, tense, close matches are much rarer.

Unless, of course, you, Sgt Hop, are getting more balanced battles than average, by 'arrangement'.

[RivertheRoyal]

 

ROFL.

 As if.

 The reality is the majority are one sided 'roflstomps'.

Balanced, tense, close matches are much rarer.

Unless, of course, you, Sgt Hop, are getting more balanced battles than average, by 'arrangement'.

 

Didn't you know? Each unicum level player enters into a deal with WG. We give them something of value—kidneys, time, souls, WG accepts anything—and in exchange, they give us better matchmaking and luck. So, unicums have closer games more often because of this deal. Their shells also disperse into citadels more often, set more fires, break more modules, incoming shells will overpen more, their citadels are never hit, and fires never stick on their ships.

All in all, a good deal.  

[VGLance]

You guys keep rehashing the same wrong talking points about skill based MM because you're using the same ridiculously juvenile skill based MM concepts.  I've gone to great lengths to deliver a solution that addresses every single one of your points making them null and void.

 

For example, my algorithm doesn't reduce queue times at all because it shuffles players AFTER the 24 players have been selected.  You need to remove that image in your head that skill based MM has to achieve a perfect 50/50% chance and adopt a healthier paradigm of an algorithm that simply shuffles players to within a 60/40 threshold thus removing the overwhelming 90/10% imbalances where the match is over before it started.  And if you're thinking it's impossible to achieve a 60/40 threshold when you factor in divisions, not a single person has yet to provide me with a realistic scenario of two teams where my algorithm didn't improve on the MM.  That is because it also has a built in fail safe where it will continue to shuffle solo players towards the 60/40 threshold but stop short of it if there are no other options (still resulting in a more even distribution of skill between teams).  And for the record, there has yet to be a scenario presented (again realistic, not one with like 4 3-man divisions on one team) that hasn't been able to get within that 60/40 threshold.

 

Also, another critical factor you need to wrap your head around is that you are not forcing potatoes to only get on teams with other potatoes and purples to only be on teams with other purples so that the potatoes will never get the opportunity to learn from better skilled players.  The mix is just the same as it is now.  My algorithm simply ensures that the majority of reds and/or purples aren't ending up on the same side.

 

And lastly there's this fear that since a skill based MM would drive everyone's win rate closer to 50%, the skill based MM would fail to distinguish who are the skilled players and who aren't thereby defeating the purpose.  While win rates will gravitate toward 50% a new paradigm will naturally be adopted in the fact that WTR will remain as the more accurate determination of expected performance.  Regardless of the outcomes of the matches, the skilled players will be more successful at capping, defending, dishing out damage and protecting their own health vs. those with far less skill.  And the skill based MM would be based off of efficiency rating, not win rate.

 

If you're going to make a post about why skill based MM won't work, you have to do the mature and intellectually proper thing and say SPECIFICALLY that the skill based MM you are referring to is the idiotic one that most sane people would agree is stupid.  But then again, do you really need to waste time making a post about it?  It's like acting like you've reached a higher level of enlightenment few humans will ever reach and in doing so, you need to inform the masses of your new found wisdom, and that wisdom is telling people that putting their hand on a red hot stove is a bad idea.

[anonym_OlK73e329bMY]

It's stressful too. I used to be a Top 8 Masters Starcraft II player and let me tell you about that ladder anxiety..

[aethervox]

 

Didn't you know? Each unicum level player enters into a deal with WG. We give them something of value—kidneys, time, souls, WG accepts anything—and in exchange, they give us better matchmaking and luck. So, unicums have closer games more often because of this deal. Their shells also disperse into citadels more often, set more fires, break more modules, incoming shells will overpen more, their citadels are never hit, and fires never stick on their ships.

All in all, a good deal.

 

I like how you show up to disparage any contrary opinion to the 'WG Glee Club' boosters. Still, River, a 'sort of funny' retort, lol.

[RivertheRoyal]

 

I like how you show up to disparage any contrary opinion to the 'WG Glee Club' boosters. Still, River, a 'sort of funny' retort, lol.

 

If that's what you think I do, then you haven't known me well enough. 

 

 

[madgiecool]

TL;DR

 

As you get better, and face better players until you start to lose more.  Result 50%

 

As you lose more, you face worse opponents.... and then you will win more.  Result 50%

 

IMHO if will all have a 50% win/loss what will be the metric for skill base MM?

[SergeantHop]

 

ROFL.

 As if.

 The reality is the majority are one sided 'roflstomps'.

Balanced, tense, close matches are much rarer.

Unless, of course, you, Sgt Hop, are getting more balanced battles than average, by 'arrangement'.

 

Again, I see them all the time. Not every game, but most games I play are not particularly one-sided. Some, obviously, because it's random. Conspiracy theories are silly, and you still haven't provided any proof. 

 

Also, I've asked this I don't know how many times, but, if you hate WG so much, why the [edited]are you still here?

[VGLance]

TL;DR

 

As you get better, and face better players until you start to lose more.  Result 50%

 

As you lose more, you face worse opponents.... and then you will win more.  Result 50%

 

IMHO if will all have a 50% win/loss what will be the metric for skill base MM?

 

If you're too lazy to read the very information that answers your question, don't embarrass yourself by asking the question.  One of the basic laws of how not to be a derp.

[Business6]

 This topic is solely my opinion on the matter. If you agree/disagree with me, please constructively discuss down below  

 

In short, my answer is no. While it has come to be apparent, WoWs is kind of starting to become just like WoT with the amount of roflstomp matches going on, especially with most schools starting to get out and all the casual weekend players playing nonstop for a couple months. Anyways, the idea has been thrown to solve this issue by implementing a skill based MM system, which sounds like a good idea on paper, but with a little more thought, actually has lows of major flaws.

 

1.) Long Queue Times: While this one is a little self explanatory, one issue with the idea of SBMM is the significant increase in queue times. Right now I can jump into my Udaloi and hit battle, then get a match in about 10 seconds. if I were to be placed into matchups based on skill, then that 10 seconds would probably stretch into a couple of minutes while the MM looks for enough people of my "skill" to match me up with, and can be even longer depending on time of day, weather, and other things. Would you rather wait 10 seconds for a match or 5 minutes for one?

 

2.) No "Fair" method: Another issue with implementing SBMM is the lack of ways to do it. Lets say WG decided on implementing a "Skill Number" that is the result of a whatever-formula they decide on, the problem is that if people want balanced games, then they suddenly have to worry about that number. They're forced to worry about their stats and performances they wouldn't otherwise care about in order to keep getting balanced games. Otherwise casual weekend warriors suddenly are forced to "git gud" if they want to have a prayer of getting a somewhat balanced game, and to people who just play to have a good time, that is really a very unfair situation. Its a case of screwed if you do, screwed if you don't. This situation is the opposite of fair. So what if WG decides to use WTR as the skill measurement? Same flaw. What if they hide the skill numbers from the player, as a sort of "Out of sight, out of mind?" Another problem with that is the possibility on unintentionally inflated stats. If someone is playing in a division and wins a whole bunch of games, their sudden rise in WR can be interpreted by the game as an "increase in skill" and thus lead that player into a skill level they really aren't ready for.

 

3.) Can games be TOO balanced? In short, yes. If the MM takes player skill into account, the matchups will be quite balanced, which on paper is a good thing. But in practice, the lack of skill variation usually leads to a stalemate; neither team can really gain an edge over the other, and that can draw out games to be a lot longer then it needs to be. While WoWS battles are capped at 20 minutes, not everyone has the time and/or patience to sit there and play for the full 20 minutes every single battle. Take me, for instance. I do consider myself an above average player, and I do enjoy balanced games, but I draw the line at every game lasting half a year before ending by timer. I don't mind a 20 minute match every here and again, but I don't have the patience to sit there for 20 minutes every game while both sides just kinda sit there and neither team gains any sort of advantage, and thus both teams just kinda sit there and do...well....not much of anything. That's just me, but I can guarantee that there are plenty of others out there who don't have the time, patience, or energy to spend the full 20 minutes every single battle. 

 

4.) The game is not competitive: WoWS is not meant to be a really competitive game. It's meant to be a place online where fans of naval warfare can go and sail on historical vessels and fight other players, not to be a place where everyone in every match is a full on tryhards. Skill based MM will only increase tension between teams, as neither wants to lose and risk losing a skill rank or whatever. This sort of thing kind of ties back to #1, but the difference here is that the increased stakes will needlessly and pointless increase tension, which can and will only lead to the salt miners' paychecks being quintupled

 

5.) Newer players will never improve: Think on that, if skilled players are only matched up against other skilled players, then new players are only matched against other new players, and nobody on either team has any clue what they're doing. How do you expect them to learn like that? If nobody in those kinds of matches have any sort of experience, who's there to really teach the newer players how to get better? I know by hard experience that one of the best teachers in the world is getting your butt handed to you. But in a skill based MM, the noobs are only matched against other noobs, and nobody has the slightest clue as to what they're really doing, who's gonna be the one to teach them the "error of their ways" at least for that battle? Knock, knock, who's there? NOBODY!

 

6.) There is already a GM w/ SBMM: It's called Ranked. While the current Ranked system is honestly just a weeee bit flawed, Ranked is probably the closest you'll ever have to a SBMM. Live with it

 

These are just the ones I thought of right off the top of my head. I know there are plenty of others, and I'll be sure to add those later, and I will look at both sides of the coin if you show it to me......

 

1 - There are games that do SBMM where you'll get a few great, some good, some average and a few bad players in a given match.  That's really the only way it would work with Warships for a number of reasons though I'd say population is the primary reason.  How the matchmaker works in the high population times is a lot less of an issue compared to when the population declines at night and the parameters need to start being a little bit more relaxed.  

 

2 - If casual weekend players are actually casual weekend players then why would they care about how a SBMM is implemented?  They're casual and don't care about improving.  Is it more unfair to have people not interested in improving (seemingly even participating in any way in the match at all, for that matter) being combined with people, regardless of having better or worse stats, who are actually trying to sharpen their skills?

 

3 - Incorrect because we already have stalemates without it.  Just because teams are balanced doesn't mean it will be a stalemate and I really don't think there's any correlation between balanced matches and stalemates given the amount of variables involved by the time the first contact is made, let alone anything afterwards.  

 

4 - While I don't think it can be competitive (not without changes, at least) in the manner of Tanks or the like your idea of what that means seems to be confused.  It's a multiplayer game and thus it's competitive.  That's entirely different from being a competitive, pro-league type of game.  Moreover you're alarmingly defensive on this point:  who are you to say what the game is supposed to be as far as why other people play it?  If they want to play on historical vessels and do whatever, that's fine---as is if they want to be "tryhards" as you so delicately put it.  You seem to want co-op or single player experience with 23 other players somehow.

 

5 - See #1.  Also, new players have to go on youtube, twitch, forums, etc in order to learn even some of the most basic information because they game just does not explain enough, if at all.  So I'd call this one moot.

 

6 - That's patently false.  It's mostly luck as to whether or not you advance especially at a given point once the pool you're working within narrows down.  It's unfortunate that it's called Ranked because it implies there's skill related to the Rank you're at as opposed to just going through the motions.  Ranked is simply about throwing enough crap at the wall to see what sticks, nothing more.  Being good or bad doesn't guarantee anything.

 

Overall I don't think there's a way to implement a skill-based matchmaker effectively for Warships even if I want there to be one.  Mainly because, if you look at overly simplistic metrics like WTR, you can manipulate the hell out of it.  A game like this would require something much more complex than that which I think just puts it further from the realm of probability.  

[VGLance]

Overall I don't think there's a way to implement a skill-based matchmaker effectively for Warships even if I want there to be one.  Mainly because, if you look at overly simplistic metrics like WTR, you can manipulate the hell out of it.  A game like this would require something much more complex than that which I think just puts it further from the realm of probability.  

 

You're right on all fronts except this last point when it applies ~specifically~ to my skill based MM algorithm because my SBMM is based off of a combination of overall WTR and WTR in the specific ship you're launching with.  That way, while you can do things to impact WTR, so can other people in that same ship so the inflated WTR cancel out.  And besides, if someone is skilled at manipulating WTR to turn purple, they have to have a decent amount of skill and commitment to carrying more than their weight anyway.  It's not like the player I ran into today with over 500 games played and averaging less than 6k damage is going to be manipulating his WTR to purple unless somehow the very bricks that hit his head to put him in his current state magically hit him again and raised his IQ by 50 points.