Showdown: Nightbird Vs Jayz Vs Cluster Fox - A Psr Comparison
#61
Posted 09 September 2020 - 07:37 AM
#63
Posted 14 September 2020 - 04:32 AM
#64
Posted 14 September 2020 - 07:18 AM
I O O percent KongLord, on 14 September 2020 - 04:32 AM, said:
Another poster who doesn't read that the graphs on the first post is based on the CURRENT POPULATION.
#67
Posted 15 September 2020 - 11:19 AM
I O O percent KongLord, on 15 September 2020 - 10:00 AM, said:
I live in the world where math exists, and people use it to test ideas and build things. You, the "community" working group, and probably PGI, are living in a world where math is optional, feelings is proof, and the first idea on any subject is the best. I don't have to claim which path is better, what you get from life will prove that many times over.
#69
Posted 15 September 2020 - 02:21 PM
#73
Posted 16 September 2020 - 06:09 AM
I O O percent KongLord, on 14 September 2020 - 04:32 AM, said:
KongLord, Nightbird is right, this was made with the current population.
Heck I could even release player data with their PSR under each system and stats before/after the simulation.
Now, if your point is : You can't have a match with only Top 1% players in Combined Queue. Well, you're right.
However, the discussion is not about all players of exactly the same PSR playing along. It's about balanced matches.
if the right metrics and logic are used, it's a matter of how many 2% players is that 1% player worth.
Maybe he's worth 1.2, 1.3, 1.5, 2.0? Then maybe he's worth 3 mid-pack players. Once you balance this out, you can even out the match outcome by pairing the top players with weaker players.
If this is done properly:
Will the Top 1% player be thrilled by the skills of his team? Probably not.
Can he carry/coordinate them? Maybe.
Will the match be an even win chance for both teams? Most of the time.
Edited by Cluster Fox, 16 September 2020 - 06:12 AM.
#75
Posted 16 September 2020 - 11:24 AM
Cluster Fox, on 16 September 2020 - 06:09 AM, said:
KongLord, Nightbird is right, this was made with the current population.
Heck I could even release player data with their PSR under each system and stats before/after the simulation.
Now, if your point is : You can't have a match with only Top 1% players in Combined Queue. Well, you're right.
However, the discussion is not about all players of exactly the same PSR playing along. It's about balanced matches.
if the right metrics and logic are used, it's a matter of how many 2% players is that 1% player worth.
Maybe he's worth 1.2, 1.3, 1.5, 2.0? Then maybe he's worth 3 mid-pack players. Once you balance this out, you can even out the match outcome by pairing the top players with weaker players.
If this is done properly:
Will the Top 1% player be thrilled by the skills of his team? Probably not.
Can he carry/coordinate them? Maybe.
Will the match be an even win chance for both teams? Most of the time.
That seems to be the most hard thing to understand, a good MM should try to build a balanced match, not necessarily a 1 % match.
#76
Posted 16 September 2020 - 12:00 PM
This discussion is very interesting, but in the end moot in my opinion. PSR doesn't really mean anything anymore when it comes to the MM. Before the reset I was T2. Like everyone, I started mid-T3 after the reset and thanks to a combination of my middling skill and playing for event objectives I now have a lot of c-bills and swag and find myself in T5.
Despite this, I still regularly drop with individual players from the top comp teams like EmP, 228, and JGx and it is not at all rare to encounter a four-player group from these types of top-tier units. In a way, it is a lot like online combat flight sims now. Everyone goes into the same pot so you might end up against a rookie, or the Red Baron.
Good hunting,
CFC Conky
Edited by CFC Conky, 16 September 2020 - 12:02 PM.
#77
Posted 16 September 2020 - 02:37 PM
CFC Conky, on 16 September 2020 - 12:00 PM, said:
This discussion is very interesting, but in the end moot in my opinion. PSR doesn't really mean anything anymore when it comes to the MM. Before the reset I was T2. Like everyone, I started mid-T3 after the reset and thanks to a combination of my middling skill and playing for event objectives I now have a lot of c-bills and swag and find myself in T5.
Despite this, I still regularly drop with individual players from the top comp teams like EmP, 228, and JGx and it is not at all rare to encounter a four-player group from these types of top-tier units. In a way, it is a lot like online combat flight sims now. Everyone goes into the same pot so you might end up against a rookie, or the Red Baron.
Good hunting,
CFC Conky
Even if T1 thru T5 players are in the same game, the MM will make more balanced matches, to the point that 3 times more players will have 1 WLR and the rest will be pulled closer to it.
#78
Posted 16 September 2020 - 02:41 PM
Nightbird, on 16 September 2020 - 02:37 PM, said:
Thanks Nightbird,
I wasn't complaining, it was just an observation based on my matches since the reset. The pendulum swings both ways and if I stuck to my fully-skilled meta-build mechs my PSR would be higher.
Good hunting,
CFC Conky
#79
Posted 14 October 2020 - 06:43 AM
Removing cap - stable means it doesn't need one (or it can be very high).
New players start at 500 PSR. Arbitrary, but near the bottom of the distribution so they have room to move.
Edited by Cluster Fox, 14 October 2020 - 06:48 AM.
#80
Posted 14 October 2020 - 06:49 AM
Cluster Fox, on 14 October 2020 - 06:43 AM, said:
Removing cap - stable means it doesn't need one (or it can be very high).
New players start at 500 PSR. Arbitrary, but near the bottom of the distribution so they have room to move.
I don't think any of your win brackets graphs are correct, could you post your methodology for calculating the win rate?
https://mwomercs.com...cy-with-graphs/
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users