Maintenance for the week of November 25:
• [COMPLETE] PC/Mac: NA and EU megaservers for maintenance – November 25, 4:00AM EST (9:00 UTC) - 7:00AM EST (12:00 UTC)
• Xbox: NA and EU megaservers for maintenance – November 27, 6:00AM EST (11:00 UTC) - 9:00AM EST (14:00 UTC)
• PlayStation®: NA and EU megaservers for maintenance – November 27, 6:00AM EST (11:00 UTC) - 9:00AM EST (14:00 UTC)
• [COMPLETE] PC/Mac: NA and EU megaservers for maintenance – November 25, 4:00AM EST (9:00 UTC) - 7:00AM EST (12:00 UTC)
• Xbox: NA and EU megaservers for maintenance – November 27, 6:00AM EST (11:00 UTC) - 9:00AM EST (14:00 UTC)
• PlayStation®: NA and EU megaservers for maintenance – November 27, 6:00AM EST (11:00 UTC) - 9:00AM EST (14:00 UTC)
Healthy Leaderboard Implementation & Why the new Battlegrounds are awful.
evLRise
✭✭✭
An accurate Match Making Ranking system for PvP is CRUCIAL in any game with PvP elements, as it influences tons of factors, such as balancing, player experience, fairness and overall enjoyment.
With the introduction of the new Battleground mode, we finally have the opportunity to implement such a fundamental system, one which to this day still eludes every single aspect of the game.
To understand WHY a MMR ( short for match making ranking ) based rank system is so important, we can have a quick look at the current Battleground itteration and it's flaws:
1. Matches are completely unbalanced. From my current experience, about 1/10 matches are actually "close" in terms of win probability, where as 9/10 are completely one-sided due to one team completely overperforming the other.
2. You can encounter pre-made parties when solo-queue'ing. Slightly off-topic from the main idea, but a fatal flaw nonetheless. Considering the popularity of complaints directed towards groups in PvP ( take for example Ballgroups ), I find it difficult to understand why, to this day, neither the vast majority of the player base nor the development team understands this simple idea: A coordinated group will ALWAYS win against an uncoordinated group. There are close to 0 exceptions to this statement.
3. The current leaderboard system does NOT reward player skill. It rewards time investment. As you can never lose rank when you lose a Battleground game, your leaderboard ranking can ONLY go up, favoring time played in detriment of skillful play and winrate.
4. The Leaderboard system is based on medal points instead of actual wins. This issue worsens the aspect described under #3, as players can simply "fish" for actions that generate medal points but don't necessarily help the team. Example: An arcanist shielder that almost single-handedly carries the team through a Deathmatch will get LESS leaderbord points than a Templar from the exact same game, one that ocassionally uses the Rite of Passage ultimate, as shields do NOT count towards medal points.
5. All these issues combined DISCOURAGE players from participating in Battlegrounds, which leads to a very small player pool, where skill discrepancy becomes even more apparent. This issue hurts every single category of players, as completely overwhelming enemies and having them disconnect half-time is extremely unsatisfying, whereas waiting a few dozen minutes in queue to go against a team that you have virtually 0 chance against encourages you to stop playing this mode.
Introducing an actual MMR based ranking system takes care of almost every single one of these issues ( only exception being #2, which is solveable through other means).
A. Getting matched against people with similar rank as you will create a way more balanced PvP experience, where people are actually incentivized to try and win / work together etc. rather than AFKing in base after realizing they have almost 0% winrate probability, or straight up trolling opponents as they can't do anything to fight back. Playing with similarly skilled players provides all teams a realistic winning expectancy and a healthy climb through the ranks, as one needs to actually improve his gameplay above the current MMR to reach higher ranks.
B. A Rank / MMR based leaderboards allows for healthy competition and gives players a sense of achievement, as rather than being rewarded for hunting for medal points, your main objective becomes winning.
C. The possibility to DECREASE in rank is IMPERATIVE to this scenario. Not all players are at the same skill level, and rewarding everyone the same way is consequentially unfair. Remember that PvP is a competitive environment. By definition, it means players COMPETING against other players. Both players and developers need to understand that ranking exists in order to provide balanced matchmaking. Not every single player needs to reach the top #100 leaderboards. In reality, rewarding players for abusing the most rewarding medal point mechanic rather than their team contribution is completely unfair towards the players which dedicated literally hundereds if not thousands of hours to understand the game and improve.
D. An actual rank system provides huge opportunities for balancing, as developers will be able to see which class / sets / skills / party compositions have the highest winrates / best performance. Balancing PvP aspects around top performing players is so important that I can't even stress enough about it. Different classes / sets / skills are easier / more difficult to play than others. If for example a certain aspect ( be it skill, set etc. ) is strong enough that it performes above average even when used at an average skill level, a player or group with a deep understanding of the game can use the said mechanic to undisputedly dominate with it. Because of this, being able to identify top performing groups and individuals and using that data as standard is a key factor when it comes to creating an engaging and fair PvP environment balance wise. In this scenario, we can finally employ the understanding of the most knowledgeable players, rather than ending up nerfing certain aspects on a whim or buffing something to the point where it becomes visually overperforming.
E. Slightly off-topic to address the issue marked as #2 - Players need to have the possibility to avoid pre-made groups in matchmaking while solo-queue'ing. This can be implemented as simple as a checkbox that one could validate or not, based on choice, before entering the queue.
F. When it comes to implementing MMR based ranking, the phylosophy is rather simplistic. Should one influence his team in a positive way ( winrate > 50% ) his MMR should go up. Should one be a negative influence when it comes to the team's overall performance ( winrate < 50% ) MMR should go down, until each player reaches the level where they get matched against evenly skilled opponents. As such, gaining or losing rank should NOT take into account elements like medal points, stats etc. but simply winrate.
Not only is it a VERY complicated process to take personal stats into account when determining rank gain, but it can also lead to dozens of errors. Handful of random examples:
1. If I play healer against a very burst damage oriented enemy group, my heal will generally be low, as rather than focusing on DPS, they do concentrated damage and there's not much to heal outside of their damage drops.
2. Should I play a Nightblade DD, my damage will be significantly lower compared to for example a damage over time oriented DK or Stamina Sorcerer, as I will mostly only deal a fixed amount of burst damage when my ultimate is up.
3. Should I play an arcanist shielder, the game won't recognize my contribution, as shields are ignored.
4. Should I play a necromancer healer, which does not invest into critical healing and chance due to passives, a healer that might end up healing significantly less than me will get awarded more Medal Points, as critical healings are considered more valuables. Therefore, due to my Near-Death Experience passive ( While you have a Living Death ability slotted, your Critical Strike Chance with all healing abilities is increased by up to 20% in proportion to the severity of the target's wounds. ), I will actually get punished medal points wise for keeping my team mates at full HP all the time.
G. Ranking NEEDS to be separated between solo battlegrounds and group battlegrounds, as the solo experience SHOULD be fundamentally different gameplay wise from the group experience ( but it's not due to the issue mentioned under #2 ).
A very basic MMR system could look something like this:
Won a BG? +1 MMR
Lost a BG? -1 MMR
Only get matched with team mates / opponents who are no further appart than ~20-40 MMR than me.
Edited by evLRise on 31 October 2024 19:22 With the introduction of the new Battleground mode, we finally have the opportunity to implement such a fundamental system, one which to this day still eludes every single aspect of the game.
To understand WHY a MMR ( short for match making ranking ) based rank system is so important, we can have a quick look at the current Battleground itteration and it's flaws:
1. Matches are completely unbalanced. From my current experience, about 1/10 matches are actually "close" in terms of win probability, where as 9/10 are completely one-sided due to one team completely overperforming the other.
2. You can encounter pre-made parties when solo-queue'ing. Slightly off-topic from the main idea, but a fatal flaw nonetheless. Considering the popularity of complaints directed towards groups in PvP ( take for example Ballgroups ), I find it difficult to understand why, to this day, neither the vast majority of the player base nor the development team understands this simple idea: A coordinated group will ALWAYS win against an uncoordinated group. There are close to 0 exceptions to this statement.
3. The current leaderboard system does NOT reward player skill. It rewards time investment. As you can never lose rank when you lose a Battleground game, your leaderboard ranking can ONLY go up, favoring time played in detriment of skillful play and winrate.
4. The Leaderboard system is based on medal points instead of actual wins. This issue worsens the aspect described under #3, as players can simply "fish" for actions that generate medal points but don't necessarily help the team. Example: An arcanist shielder that almost single-handedly carries the team through a Deathmatch will get LESS leaderbord points than a Templar from the exact same game, one that ocassionally uses the Rite of Passage ultimate, as shields do NOT count towards medal points.
5. All these issues combined DISCOURAGE players from participating in Battlegrounds, which leads to a very small player pool, where skill discrepancy becomes even more apparent. This issue hurts every single category of players, as completely overwhelming enemies and having them disconnect half-time is extremely unsatisfying, whereas waiting a few dozen minutes in queue to go against a team that you have virtually 0 chance against encourages you to stop playing this mode.
Introducing an actual MMR based ranking system takes care of almost every single one of these issues ( only exception being #2, which is solveable through other means).
A. Getting matched against people with similar rank as you will create a way more balanced PvP experience, where people are actually incentivized to try and win / work together etc. rather than AFKing in base after realizing they have almost 0% winrate probability, or straight up trolling opponents as they can't do anything to fight back. Playing with similarly skilled players provides all teams a realistic winning expectancy and a healthy climb through the ranks, as one needs to actually improve his gameplay above the current MMR to reach higher ranks.
B. A Rank / MMR based leaderboards allows for healthy competition and gives players a sense of achievement, as rather than being rewarded for hunting for medal points, your main objective becomes winning.
C. The possibility to DECREASE in rank is IMPERATIVE to this scenario. Not all players are at the same skill level, and rewarding everyone the same way is consequentially unfair. Remember that PvP is a competitive environment. By definition, it means players COMPETING against other players. Both players and developers need to understand that ranking exists in order to provide balanced matchmaking. Not every single player needs to reach the top #100 leaderboards. In reality, rewarding players for abusing the most rewarding medal point mechanic rather than their team contribution is completely unfair towards the players which dedicated literally hundereds if not thousands of hours to understand the game and improve.
D. An actual rank system provides huge opportunities for balancing, as developers will be able to see which class / sets / skills / party compositions have the highest winrates / best performance. Balancing PvP aspects around top performing players is so important that I can't even stress enough about it. Different classes / sets / skills are easier / more difficult to play than others. If for example a certain aspect ( be it skill, set etc. ) is strong enough that it performes above average even when used at an average skill level, a player or group with a deep understanding of the game can use the said mechanic to undisputedly dominate with it. Because of this, being able to identify top performing groups and individuals and using that data as standard is a key factor when it comes to creating an engaging and fair PvP environment balance wise. In this scenario, we can finally employ the understanding of the most knowledgeable players, rather than ending up nerfing certain aspects on a whim or buffing something to the point where it becomes visually overperforming.
E. Slightly off-topic to address the issue marked as #2 - Players need to have the possibility to avoid pre-made groups in matchmaking while solo-queue'ing. This can be implemented as simple as a checkbox that one could validate or not, based on choice, before entering the queue.
F. When it comes to implementing MMR based ranking, the phylosophy is rather simplistic. Should one influence his team in a positive way ( winrate > 50% ) his MMR should go up. Should one be a negative influence when it comes to the team's overall performance ( winrate < 50% ) MMR should go down, until each player reaches the level where they get matched against evenly skilled opponents. As such, gaining or losing rank should NOT take into account elements like medal points, stats etc. but simply winrate.
Not only is it a VERY complicated process to take personal stats into account when determining rank gain, but it can also lead to dozens of errors. Handful of random examples:
1. If I play healer against a very burst damage oriented enemy group, my heal will generally be low, as rather than focusing on DPS, they do concentrated damage and there's not much to heal outside of their damage drops.
2. Should I play a Nightblade DD, my damage will be significantly lower compared to for example a damage over time oriented DK or Stamina Sorcerer, as I will mostly only deal a fixed amount of burst damage when my ultimate is up.
3. Should I play an arcanist shielder, the game won't recognize my contribution, as shields are ignored.
4. Should I play a necromancer healer, which does not invest into critical healing and chance due to passives, a healer that might end up healing significantly less than me will get awarded more Medal Points, as critical healings are considered more valuables. Therefore, due to my Near-Death Experience passive ( While you have a Living Death ability slotted, your Critical Strike Chance with all healing abilities is increased by up to 20% in proportion to the severity of the target's wounds. ), I will actually get punished medal points wise for keeping my team mates at full HP all the time.
G. Ranking NEEDS to be separated between solo battlegrounds and group battlegrounds, as the solo experience SHOULD be fundamentally different gameplay wise from the group experience ( but it's not due to the issue mentioned under #2 ).
A very basic MMR system could look something like this:
Won a BG? +1 MMR
Lost a BG? -1 MMR
Only get matched with team mates / opponents who are no further appart than ~20-40 MMR than me.
1