Let's Discuss Ranked PvP Ideas...

[omaan]

Wait... you just wrote two different and possibly/probably conflicting requirements in one paragraph, and then said "it's not that hard". I can put "it's not that hard" after anything. If there are two mercs that are high rated in the match, and two maras that are low rated... exactly how are they supposed to split that up so that the high rated mercs don't get "punished" with low rated maras, and yet at the same time make sure that both mercs don't end up against both maras?

 

Or are you saying they should not pop a match at all in a case where the ratings are so different, even if they have 8 in queue?

 

Is there is no other highrated classes in queue. Put two mercs against each other and one mara in one team while second mara in another. Both highrated mercs will have a chance to win because both maras are lowrated. Currently the system would make it like this - put 2 lowrated maras in one team with one highrated merc while the second merc will get 2 opers in his team. Guess who will win....

[Lhancelot]

The ELO-system had its chance, and it is clearly not working.

 

Functional ELO requires:

 

 

• Mirror classes and specs and not just DPS vs. DPS e.g. 3 mercs vs. 2 sorcs and 1 oper
  
• All 8 players are within a very small rating margin (10 - 20 points)
  
• Games don't start when imbalanced e.g. if one side have a tank and the other does not
  
• Glitches and bugs are nearly non-existing
  
• Cheaters, trolls etc. are dealt with in-game, which means mods and game admins
  
• Gear has no influence, only skill
  

 

Ranked in SWTOR will never be remotely close to any type of competitive esport. It is a pvp sideshow, and the only focus should be to make it fun and less toxic. I strongly believe that a more "grindy" point system would remove a great deal of toxicity, as players wouldn't rage quite as much when they lose. Yes it can still be exploited, but so will any system they come up with.

 

Right. We can theorize what will be exploited if they went to a system that focused more on number of wins than "quality of wins," however we KNOW how the present ELO system not only is exploited but proven for years not to be an effective way to prove anyone's PVP skills or discourage cheating for rating.

 

The problem of people possibly exploiting a wins based scoring system is a small one because you don't have to be a genius to recognize it's far easier spending one evening and a handful of matches to get top ELO rating in the present rating system. Compared to if people end up actually trying to win trade in a wins based scoring system they will need a significantly larger time investment into accruing a large number of won matches.

 

This in turn will make it easier to detect and track if someone is win trading for most wins in ranked PVP having a far longer time investment to succeed at getting wins.

 

ELO rating is far too easy to manipulate right now. There's no denying that, and we have years of history showing how ranked has only grown more and more corrosive to the point now where it's never been viewed worse. These changes proposed would improve not just ranked but the game imo. If they can get more people interested in this facet of the game that will only make the game that much more fun again.

 

The only part I would reconsider is the rating decay. As others pointed out, forcing people to have to hop on a specific toon to keep it from decaying isn't really necessary, and depending on what kind of scoring system they go with just the contest itself might require players keep track of who is rated highest and continue activity on toons rated high.

 

If BW go off a wins based system for instance, if you had 75 wins which was highest when you stopped playing your PT and someone else gets to 76, this would encourage you to log in and play your PT to get the wins highest again.

 

I think the notes Mike gave are encouraging but still a bit vague. Once more ideas are bounced around and they take more time to consider the feedback and come back with finer details I think so far the proposed changes are going in the right direction.

 

Another point that many people seem to have added is mat farmers being able to do Team Ranked instead of how they are now only able to queue solo ranked. This might be something to consider changing back, too.

Edited March 14, 2019 by Lhancelot

[Banderal]

Is there is no other highrated classes in queue. Put two mercs against each other and one mara in one team while second mara in another. Both highrated mercs will have a chance to win because both maras are lowrated. Currently the system would make it like this - put 2 lowrated maras in one team with one highrated merc while the second merc will get 2 opers in his team. Guess who will win....

 

Ah ok, so the "no class stacking" rule trumps the "don't put goods with bad rule" if necessary. I get it.

 

I'm not sure I would order them that way. I mean, I've been with and against "bad" mercs, and they can go down pretty fast. And a dead merc is worth exactly as much as a dead PT for the rest of that round. But in general, yeah, I agree that the class stacking gets old. As someone who seems to attract "nets", being against 3 mercs is just a nightmare.

 

But I might go so far as to agree with another suggestion just one or two above this... if there's no one in the queue at my level, don't pop. Personally, for me, since I'm right in the middle, that's not a problem. But I could see how the very high or very low rated players would not like it. Although, on the other hand - that might be part of a solution for win traders. Once the "winning" side of a win trader pair gets a bit of elo advantage, they won't pop with their other half anymore? Maybe?

[Akushii]

One of the system that I saw in a different game that I thought was really great and encouraged play was the following.

 

* Remove rating decrease for a loss until you reach a certain rank (bronze). Start rating from match 1 and start at 0.

* Once you reach bronze rank, you can lose rating for a loss but win should give higher rating gain than losses.

* Once you reach the next tier (gold) increase the rating loss for losses.

* Have an additional incentive to keep playing. Have a bar that you can fill up that earns you a mount. For each win, 2% of the bar is filled, for each loss 1% of the bar is filled. This way even if you are losing, you are encouraged to keep playing because you know you will get to the end goal. Where as currently if your first 10-20 matches don't go well, you pretty much give up on that toon and try a different one.

 

I think that if you have a 50/50 win loss ratio, you should be able to reach the silver rank at the end of the season. Many people are discouraged from doing ranked in the fist place because a few losses (specially early on) can mess up the entire season.

[Screaming_Ziva]

• 
   
  
• Adding a rating decay over time
  

I'd have to say a big old NO to this one.

 

If I play (and win) enough to achieve the rank I want to be at, why should I be punished for no longer playing to make sure I keep that rank?

[Yezzan]

I believe the objective is twofold:

 

1. Increase Ranked WZ participation across the player-base
   
2. Eliminate Devious Behavior
   

 

If you want to increase ranked activity then much of what you suggest will not help. The first consideration ( reducing win rating) will not incentivise anyone to play. The lower tiers should not result in a loss of rating in the event of a loss. The limiting factor for tiers should be number of matches played. For example, to advance from tier 3 to tier 2, a notional rating of 3000 is required and 10 matches played. Hence, the punitive effect of a loss is that you would have to play more matches, as opposed to you losing rating.

 

The rating decay is a terrible proposition as it penalizes players who have alts, and players who for various reason may not be 100% active during the season.

 

match-making is the biggest issue as it is not prioritizing trinity group compositions, as it should.

 

I think rewards for season participation should be more attractive so that all the emphasis is not totally placed on top tier players. The idea is to increase wide-spread participation. Obviously these rewards could be tied to specific number of matches, as opposed to rating.

 

if everything is locked behind Tier one rating then most people not going to bother cause they know they are not good enough to achieve a tier one rating.

[JediMasterAlex]

Rating decay is the single most important, realistic change that you can make to ranked that will deter bad behavior. Wintraders will be forced to do it much more, and will therefore be much easier to catch. Patterns will be easier to discern, and they will probably get sloppy when they have to do it so much. For this reason alone, rating decay would be worth implementing.

 

For those complaining about decay, the system would still only take into account your highest earned rating while calculating all reward tiers except top 3, so I'm not sure what the issue is. If you make it to the tier that you want on an alt and stop playing on it, why do you care if you decay? If you care about your spot on the leaderboard, well, the leaderboard should reflect the top players of each class, not who got lucky in their first 20-30 games (and I say this as someone that went 20-8 and got over 1500 within 30 games this season). In other words, winning a majority of games within a small sample is not at all a reflection of skill, but of luck in this game. Luck shouldn't be rewarded; it should take a consistent demonstration of ability

[sharkfishman]

Rating decay is the single most important, realistic change that you can make to ranked that will deter bad behavior. Wintraders will be forced to do it much more, and will therefore be much easier to catch. Patterns will be easier to discern, and they will probably get sloppy when they have to do it so much. For this reason alone, rating decay would be worth implementing.

 

For those complaining about decay, the system would still only take into account your highest earned rating while calculating all reward tiers except top 3, so I'm not sure what the issue is. If you make it to the tier that you want on an alt and stop playing on it, why do you care if you decay? If you care about your spot on the leaderboard, well, the leaderboard should reflect the top players of each class, not who got lucky in their first 20-30 games (and I say this as someone that went 20-8 and got over 1500 within 30 games this season). In other words, winning a majority of games within a small sample is not at all a reflection of skill, but of luck in this game. Luck shouldn't be rewarded; it should take a consistent demonstration of ability

 

So the way you envision rating decay is that you would still retain the highest rating you obtained, even if your current rating would drop?

 

I.e. - I played 50 matches and got to 1600. That's good enough for me so I stop on that toon. My highest rating stays at 1600, but if I start playing again two weeks before end of season my rating would now be at 1200 or something, so I would have to fight a lot if I now wanted to try and beat my 1600 score?

 

I could get behind that. As long as your highest achieved score is what counts.

[sharkfishman]

I think rewards for season participation should be more attractive so that all the emphasis is not totally placed on top tier players. The idea is to increase wide-spread participation. Obviously these rewards could be tied to specific number of matches, as opposed to rating.

 

if everything is locked behind Tier one rating then most people not going to bother cause they know they are not good enough to achieve a tier one rating.

I agree.

 

When even gold tier players don't have enough rewards to purchase an armor set, there's probably a problem, lol.

[bladech]

Ah ok, so the "no class stacking" rule trumps the "don't put goods with bad rule" if necessary. I get it.

 

I'm not sure I would order them that way. I mean, I've been with and against "bad" mercs, and they can go down pretty fast. And a dead merc is worth exactly as much as a dead PT for the rest of that round. But in general, yeah, I agree that the class stacking gets old. As someone who seems to attract "nets", being against 3 mercs is just a nightmare.

 

But I might go so far as to agree with another suggestion just one or two above this... if there's no one in the queue at my level, don't pop. Personally, for me, since I'm right in the middle, that's not a problem. But I could see how the very high or very low rated players would not like it. Although, on the other hand - that might be part of a solution for win traders. Once the "winning" side of a win trader pair gets a bit of elo advantage, they won't pop with their other half anymore? Maybe?

 

no it's just ur example of issue was easy to solve lol. like 2 lowrated maras and 2 highrated mercs ofc it would be fair to put both lowrated maras in different teams and let the best merc to win. such fight is fair since both maras are same rated. Ofc we can't afford to make highrated players to fight always only other highrated players it is UNREAL. however teams must be made RANDOMLY without class stack and if there are two HIGHRATED players ofc they must be put against each other.

[JediMasterAlex]

So the way you envision rating decay is that you would still retain the highest rating you obtained, even if your current rating would drop?

 

I.e. - I played 50 matches and got to 1600. That's good enough for me so I stop on that toon. My highest rating stays at 1600, but if I start playing again two weeks before end of season my rating would now be at 1200 or something, so I would have to fight a lot if I now wanted to try and beat my 1600 score?

 

I could get behind that. As long as your highest achieved score is what counts.

 

Yeah, i mean, that's how it works now. I'm currently below 1500, but I'll still get gold at the end of season (if they keep gold at 1500) regardless of how much I play between now and then. The only rewards that are tied to your current rank are top 3, and that is where decay would have its biggest impact. We'd see actual movement at the top of the leaderboards throughout the season, and barring last-second wintraders (another issue), we'd actually see the most deserving get those rewards.

[snave]

Please stop suggesting points for wins, that's literally the same as the currency system we have now but with added ego compensation.

[omaan]

Hey All,

 

First off, thank you all for your participation in the broader PvP thread talking about win trading. I appreciate all of the feedback that was given and hope that I was able to shed some light on the topic. The goal of this new thread is to specifically focus on gathering ideas for potential systemic changes to Ranked PvP systems.

 

To start, I’d like to share the high-level goals for any ideas gathered in this thread:

 

 

• 
   
  
• Change rating distribution to better encourage prolonged, honest play throughout the season
  
• Reduce or eliminate mechanics that incentivize exploitative behavior
  
• Make matchmaking feel more balanced
  

 

There have been a lot of specific ideas posted on the forums and discussed internally about how we could improve our current rating system, our matchmaking system, and to better deter exploitative behavior throughout. Please know that I have gathered (and continue to look for) as many of those ideas as I can find, and have been sharing them with the broader team in order to build some sense of how viable they each may be.

 

Before we dig too deep into any of these ideas, I need to clearly set expectations. These are early steps in an effort to build a broad consensus towards as many of these ideas as make sense to potentially pursue. Depending on what we discuss here, follow up conversations will need to be had with our dev team to figure out exactly how much time each of the ideas would take to implement. With that information we'd need to find them a place on our schedule. Fair warning, this is not necessarily a short process.

 

With all of that in mind, here are some of the specific ideas that we feel would best work towards the goals stated above:

 

 

• 
   
  
• Reduce both the rating gained from a win and removed from a loss
  
• Adding minimum requirements for games played to achieve each rank
  
• Adding a rating decay over time
  
• Adjusting the matchmaking system to better prevent large class imbalances between teams
  

 

We understand that there are obvious sensitivities to changing the rules of the game in a season that is already underway. For that reason, and the logistical hurdles mentioned above, I cannot promise that any of these changes will go live in season 11. It is possible that some ideas may be easier to implement than we expect or that obtaining one may lessen the need for another, but I want to be as transparent as possible regarding timelines.

 

I realize that some of you may have previously put forth suggestions that are not captured in the list above. If anyone has a proposal that you feel is not covered by any of the above ideas, please share it in this thread (or link to where you have explained it elsewhere), and we can all discuss the pros and cons. I’m looking forward to digging in here. Thanks for making it through this mountain of text!

 

Until next time…

 

 

Pls do not change elo system into grinding points system. Watching those who are unskilled getting gold tier just because they are able to sit all day and ruin solo ranked during the whole season wiil definitely force me to unsub. Only those who understood their class, got the proper experience and skills should be able to reach highest tier. Not just all-day queuers and points farmers

[sharkfishman]

Pls do not change elo system into grinding points system. Watching those who are unskilled getting gold tier just because they are able to sit all day and ruin solo ranked during the whole season wiil definitely force me to unsub.

 

So you believe that less participation is preferable?

 

Or that there should be less people playing ranked?

 

Or that no one should be rewarded for playing except the top-tier players?

 

I'm not sure why you're suggesting this for solo ranked (I mean I would understand it for granked where you actually have control over the game's outcome.)

 

Please answer this: If you play the game at the top of your skill, knowing your class, playing it in and out, are you guaranteed a win, or even a string of wins, in solo ranked?

 

So why are you suggesting that the current ELO system is somehow a measurement of skill?

[Zurules]

Thank you Mike for the post.

 

I have a couple of questions that I hope you might be able to answer me that I think would really help all of us players with our suggestions to help improve ranked.

 

 

What is Bioware's definition of what is ranked?

 

What do you want ranked to mean for your players who choose to play ranked (Bioware's perspective)?

 

 

The reason I ask these questions (I know they are somewhat similar) is because I think that most of the ideas so far that have been presented comme from players with different answers to the questions above.

 

 

As an example, my definition of ranked would be something like this: Ranked is a place for competition and it should be competitive. It is meant for players who they themselves believe they are GOOD at what ever they choose to queue with (different specs from all classes and roles) and to find out roughly where they compare to others.

 

 

If we can know Bioware's philosophy on ranked, we could save time and focus on idea's that leads towards that philosophy. (I think it might also be a good idea to create a poll where the players can select different aspects that would lead towards one philosophy or another so that Bioware can see the players preferences)

 

 

Regardless of the philosophy that anyone has, I still believe that the most effective way to make ranked more ENJOYABLE for the players would be to have frequent SMALL balance changes to shift the meta faster which will make every class and spec at least feel strong for a period of time.

 

However, one problem I still foresee is that I still believe the balance design for the game is around 8v8 and not 4v4.

 

Because of this, smaller scale skirmishes (especially when there are no supportive roles involved like tanks or heals, so 4 dps vs 4 dps) skew the power of balance to certain classes and specs that are more self sufficient. These days Merc/commandos are probably considered the strongest by most players because they have good damage, good anti-focus dcds and the option to offheal. On the other side we have the 100% consensus that VG/PT are the weakest because they are the least self-sufficient and rely the most on their teammates if they are to be successful.

 

 

This is why I truly miss the old 8v8 ranked days because every spec for every class was VIABLE in ranked. Of course some specs seemed better than others which is normal, however, the gap between what people called best and the worst specs was way smaller compared to what we have now.

 

To conclude for now, the most important thing that I feel needs to be done to improve ranked is to close the gap between the most self-sufficient and the least (find a way to keep things UNIQUE, do not give every or atleast most the same type of ability, aka do not give a reflect to pt/vg) and then do small balance changes OFTEN without re-increasing the total gap to make every spec feel good to play in 4v4 ranked.

[omaan]

So you believe that less participation is preferable?

 

Or that there should be less people playing ranked?

 

Or that no one should be rewarded for playing except the top-tier players?

 

I'm not sure why you're suggesting this for solo ranked (I mean I would understand it for granked where you actually have control over the game's outcome.)

 

Please answer this: If you play the game at the top of your skill, knowing your class, playing it in and out, are you guaranteed a win, or even a string of wins, in solo ranked?

 

So why are you suggesting that the current ELO system is somehow a measurement of skill?

 

I have 7 gold characters and three top 3 characters (for season 10). So yes i can speak of current elo system. It is far from perfect yet it allows to draw the line between those who TRYHARD, learn his class, gets proper skill and "damn iam playing 2 days for mara Why iam not gold yet?". With current system gold tier can be received by a skilled person within few weeks (or even days if playing fotm classes).

 

As some players mentioned there are a few things which could be done to improve current system such as fix class stacking (this is a real problem which i have observed for a long time), stop punishing those who has high elo (currently those who try to reach top 96 rewards and are on high elo are forced to monitor best moments to queue, dodge throwers/wintraders which is really hard and time-consuming). The matchmaking system must be random and neutral to player's elo and it should avoid class stacking.

 

In addtion to that, fix instant death bugg (which still occurs sometimes), combat removal bugg (yeah sometimes you can run away too far from everyone, remove all buffs which were given by your team mates and still not get combat reset).

 

There is really few things which could really improve current system without any need of changing it to dumb grind and mindless queue.

Edited March 15, 2019 by omaan

[sharkfishman]

I have 7 gold characters and three top 3 characters (for season 10). So yes i can speak of current elo system. It is far from perfect yet it allows to draw the line between those who TRYHARD, learn his class, gets proper skill and "damn iam playing 2 days for mara Why iam not gold yet?". With current system gold tier can be received by a skilled person within few weeks (or even days if playing fotm classes).

 

As some players mentioned there are a few things which could be done to improve current system such as fix class stacking (this is a real problem which i have observed for a long time), stop punishing those who has high elo (currently those who try to reach top 96 rewards and are on high elo are forced to monitor best moments to queue, dodge throwers/wintraders which is really hard and time-consuming). The matchmaking system must be random and neutral to player's elo and it should avoid class stacking.

 

In addtion to that, fix instant death bugg (which still occurs sometimes), combat removal bugg (yeah sometimes you can run away too far from everyone, remove all buffs which were given by your team mates and still not get combat reset).

 

There is really few things which could really improve current system without any need of changing it to dumb grind and mindless queue.

You didn't answer my questions.

 

Does skill guarantee a win, or a win streak?

 

Can you 1v4 the other team without problems?

 

How long did you play on each of those gold rank characters to get your rating? How many total wins? How many total losses?

[sharkfishman]

I have 7 gold characters and three top 3 characters (for season 10). So yes i can speak of current elo system. It is far from perfect yet it allows to draw the line between those who TRYHARD, learn his class, gets proper skill and "damn iam playing 2 days for mara Why iam not gold yet?". With current system gold tier can be received by a skilled person within few weeks (or even days if playing fotm classes).

 

As some players mentioned there are a few things which could be done to improve current system such as fix class stacking (this is a real problem which i have observed for a long time), stop punishing those who has high elo (currently those who try to reach top 96 rewards and are on high elo are forced to monitor best moments to queue, dodge throwers/wintraders which is really hard and time-consuming). The matchmaking system must be random and neutral to player's elo and it should avoid class stacking.

 

In addtion to that, fix instant death bugg (which still occurs sometimes), combat removal bugg (yeah sometimes you can run away too far from everyone, remove all buffs which were given by your team mates and still not get combat reset).

 

There is really few things which could really improve current system without any need of changing it to dumb grind and mindless queue.

 

Also, a quick question. Are you talking about group ranked or solo ranked for those characters?

 

And, if solo ranked, did you queue alone, or did you queue sync with friends?

[DanNV]

• 
   
  
• Make matchmaking feel more balanced
  
• Adjusting the matchmaking system to better prevent large class imbalances between teams
  

 

I can't really make better suggestions for the rest of your list. However, for this part it seems pretty obvious to me. Stop making teams and then trying to match teams. Instead, match players to other players with close ratings. Then make teams from 4 players who have closely matched opponents. Further, match healers with healers, tanks with tanks, DPS with DPS. NEVER put out teams with 3 healers and a DPS v 2 tanks and 2 DPS, or whatever. Keep team composition similar between teams. This will go a long way towards keeping teams even instead of some of the horrible mismatches that happen now. (Then extend that matchmaking to regs.)

[Sertar]

A thing also is the rating gain based on how many players you face. Even if you have 2000 rating and you rig a match where the 3 people on your team decline que and you win 1v4 (where the 4 enemies dont fight you ofc) you still gain 20+ rating. This also needs to be fixed cause it again helps people gain rating fast with little matches. On the otherhand you have to consider matches where 1 or more people leave, out of whatever reasons, and the reduced team faces an obvious loss. In such a case their rating loss should be minimal due to things being "out of their hand".

Thank you, I completely forgot about this. Yes, it is very important that matches that are not 4v4 DO NOT count any rating. No + but also no -. Obviously this has to go together with punishment for leaving, declining and AFKing.

 

 

So to repeat myself and summarize, here are 5 easy to implement changes, that don't require a full rework of the current system. And they will be a good start to work further on matchmaking after:

 

 

1. Make Elo visible from the first game on! Every game should have the same Elo calculation, being it the 1st or the 101st game on a character.

 

2. Add minimum requirements for games played to achieve each rank: 30 games for Bronze, 50 games for Silver, 100 games for Gold / Top 3.

 

3. Punish leaving, declining and AFK queueing with a legacy wide queue ban of XY minutes.

 

4. Don't give any Elo +/- for matches that aren't 4v4. This protects players from losing Elo in a 3v4 because of someone leaving or disconnecting, but also not gives the other team the "free" Elo gain. It also eliminates the possibility of 2v4 or 1v4 botting / win-trading.

 

5. Fix Backfills! A leaving/declining player MUST ALWAYS be replaced with a player of the same role (T/D/H). If there is absolutely noone of that role in queue at that time, start the match as a 3v4, so that it does not count any Elo for the players in that match.

 

 

Once again, thanks for working on Ranked PvP.

[DavidAtkinson]

You didn't answer my questions.

 

Does skill guarantee a win, or a win streak?

 

Can you 1v4 the other team without problems?

 

How long did you play on each of those gold rank characters to get your rating? How many total wins? How many total losses?

 

I answer with another question:

 

How can people talk about skill when players are matched with randoms in their team ? If you get the same bad healer 4 times in a row vs the top rated healer and tank you will lose 4 times no matter how " gud" you are.

[merovejec]

in a wins based scoring system they will need a significantly larger time investment into accruing a large number of won matches.

 

The people that cheat do have the time, just watch streams, and chech the names in ranked pvp every day, they are always there. One would think that after some times they would grow up and stop playing the game but no, even after so many years these people still play like there is no tomorrow

 

So a system based on matches played will not hinder them.

[merovejec]

What is Bioware's definition of what is ranked?

 

WOW! If I could vote for a post this one would get 5 stars. It actually never occured to me (I am a manager, but never thought to apply this logic to the game) that Bioware should really define what this part of the game should be about, its goals and aims.

 

Should it be a competitive environment? A way to grind gear? A way to grind UCs, mats...? If you really define what it is then you can start by coming up with the criteria for it.

 

Very good Mr. Zurules!

 

I have 7 gold characters and three top 3 characters (for season 10).

 

Please, you need to go out more

Edited March 15, 2019 by merovejec

[omaan]

You didn't answer my questions.

 

Does skill guarantee a win, or a win streak?

 

Can you 1v4 the other team without problems?

 

How long did you play on each of those gold rank characters to get your rating? How many total wins? How many total losses?

 

You ask pretty stupid questions. Ofc i did it alone; yes all chars except 2 of them were played in solo ranked; skill doesnt guarantee a win in solo ranked and it shouldn't since solo ranked matchmaking must be random and sometimes you are lucky sometimes not. The feature is that if you are skilled your win ration outweighs loose ratio which allows you to climb up with time. I already said how long a skilled player can play to obtain gold depending on the class he plays. This also includes me.

 

Btw skill cant guarantee you a win in tr neither. Much depends in your team mates who could be much less skilled than you and on setup. For instance, decent 2 marauders, jugg and a healer wont win skilled skilled engi sniper, dot mara, pt tank and a healer since aoe setup will always stack and create a huge pressure.

[Eveleria]

- I would strictly separate the ladder from any rewards apart from having it made to a certain rank.

For example in dota 2 you will get the highest medal you achieved for your profile for the previous season as well as your current medal. This should discourage wintrading to some degree, since reaching a rank at one time in the season is enough to get you your medal/titel or whatever you want to give out.

 

- How you calibrate the players at the start of the season idk, that's something you should be able to asses through your collected data.

Give out the same amount of points for wins as you deduct for losses, that's it you are done for ranking players as long as your matchmaking system is ok and the classes are balanced.

 

- If you want to reward players based on numbers be very careful with that. This will only lead to players farming statistics vs playing for the win. it doesn't matter how your numbers look as long as you did your part to make your team win.

 

- Give out a currency for winning and a smaller amount (maybe half the amount) for loosing to reward players for playing the game so they can buy deco stuff, different looks etc. It's always nice to get something on top of having fun, just don't make it something mandatory that forces you to play something although you dislike it.

 

- Gear aquisition should be done via daily / weekly quests or just use the command xp system. Don't use the ranks for that, it just gives further incentive to wintrading.

 

- Last but not least, a good pvp players doesn't need any badges or titles to show he is good. If someone is really good others will come to realize this through his plays and his ability to enable his team to win. What good does a shiny title do if it gets awarded to the wrong players through wintrading and a botched system.