

Weight Matching Tolerance To Zero
#1
Posted 18 April 2013 - 08:10 AM
#2
Posted 18 April 2013 - 08:20 AM
#3
Posted 18 April 2013 - 08:28 AM
#4
Posted 18 April 2013 - 08:33 AM
Acid Phase, on 18 April 2013 - 08:20 AM, said:
Completely untrue. I've been in more than a few random drops where there was no obvious premade, and it was still 8-0. It's going to happen.
On the other end, having been in more premades than solo drops, I can count more than a few drops where there was no obvious premade on the other team, and we still lost despite our best efforts. When this happens, all that can be said is "Well, we got a bad draw of PUGs on that one".
#5
Posted 18 April 2013 - 08:35 AM
The Lonegopher, on 18 April 2013 - 08:28 AM, said:
In this most recent patch, didn't they say they made it so they can modify stuff like this without having to put out a patch?
#6
Posted 18 April 2013 - 08:38 AM

#7
Posted 18 April 2013 - 09:23 AM
I doubt there are enough players in each skill level and chassis weight class at any given time to achieve that.
If, as I suspect, they can't both be implemented concurrently, then which takes precedence?
#8
Posted 18 April 2013 - 09:25 AM
Appogee, on 18 April 2013 - 09:23 AM, said:
I doubt there are enough players in each skill level and chassis weight class at any given time to achieve that.
If, as I suspect, they can't both be implemented concurrently, then which takes precedence?
incorrect
AND (&&) over OR (||)
Edited by CapperDeluxe, 18 April 2013 - 09:25 AM.
#9
Posted 18 April 2013 - 09:25 AM
#10
Posted 18 April 2013 - 09:50 AM
Roadbeer, on 18 April 2013 - 08:33 AM, said:
On the other end, having been in more premades than solo drops, I can count more than a few drops where there was no obvious premade on the other team, and we still lost despite our best efforts. When this happens, all that can be said is "Well, we got a bad draw of PUGs on that one".
Unpossible.
If I lose, someone HAS to be cheating or gaming the system. Period.

#11
Posted 18 April 2013 - 10:27 AM
Roadbeer, on 18 April 2013 - 08:33 AM, said:
On the other end, having been in more premades than solo drops, I can count more than a few drops where there was no obvious premade on the other team, and we still lost despite our best efforts. When this happens, all that can be said is "Well, we got a bad draw of PUGs on that one".
That is arrogant. Why is it the pugs that failed? It could be that you were just outplayed.
#14
Posted 18 April 2013 - 10:41 AM
armyof1, on 18 April 2013 - 10:39 AM, said:
And all 4-man are equally great in skill right?
Nope. But we do have an equal advantage that is instant communication. Whether we use it properly or not is a different story.

Edited by Joseph Mallan, 18 April 2013 - 10:41 AM.
#15
Posted 18 April 2013 - 10:50 AM
Joseph Mallan, on 18 April 2013 - 10:41 AM, said:

Ain't that the truth, and that's why it is kinda arrogant to assume it's all the fault of the pugs in your team because you don't win. I've seen a fair share of 4-mans just march over the ridge in unison on frozen city, thinking they'd just clean up by themselves while some slow assaults in our team haven't even caught up yet. Not surprisingly they get focused on from all directions and melted, costing our team the match. 4-mans are truly not created equal either.
#17
Posted 18 April 2013 - 10:53 AM
Appogee, on 18 April 2013 - 09:23 AM, said:
I doubt there are enough players in each skill level and chassis weight class at any given time to achieve that.
If, as I suspect, they can't both be implemented concurrently, then which takes precedence?
They can be implemented together.
If they put Weight matching within say +-50 tons for the team, then increase the variance in elo until they achieve a full drop, with an outer limit on the ELO that moves the weight balance out ( say +-100 tons), and repeat the elo variance till full drop... rinse repeat all wth outer limits on the variances, and a time limit on the overall process.
set Weight (of team): +-50T
set ELO target : X , expanding in 50 point increments in both directions until the total variance is > (maximum elo differential)
reset weight to +- 100t
set ELO target : X , expanding in 50 point increments in both directions until the total variance is > (maximum elo differential)
set a maximum weight differential, set a maximum elo differential, and set a maximum length of time to process.
Edited by Barnaby Jones, 18 April 2013 - 10:58 AM.
#19
Posted 18 April 2013 - 11:38 AM

#20
Posted 18 April 2013 - 11:45 AM
Barnaby Jones, on 18 April 2013 - 10:53 AM, said:
What you went on to describe though is in effect partly implementing weight and partly implementing ELO concurrently. That is, you are compromising on weight matching and on LEO-matching in order to have a bit of both.
That's workable, but it's not what I meant as ''implementing both concurrently''.
Your suggestion would be a good way of compromising between the two approaches.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users