No replies to this topic

[David Sumner]

Everyone talks about balancing stuff. But I want to see proof

Here in the forums I mean.
Pretty little 3D graphs updated once a month that show balancing in action
X/Z axes are the two teams
Y axis is a percentage of wins
Every graph the X/Z axes are different pairs
If things are balanced, most of these graphs should be pretty flat at ~50%
Here's what I'd like
Total # of weapon X on each side
AC2
AC5
ML
Etc
For SRM and LRM I think it would need to be total salvo size
Number of ECM units on each side
Number of TAG on each side
Number of BAP on each side
etc
If they're balanced, having more or less on a side should be relatively static
Number of players on each team
This I'd expect to be unbalanced, but it should give an idea of how much effect starting short a player has
Number of players when startup sequence finishes
Number of players that did not disconnect before dying
Number of players who dropped as part of a group
How about
Total ELO (by hundreds band)
Average ELO
Lowest ELO
Highest ELO


Don't tell me you can't track it.
It needs to write once or twice at the start / end of a match
For < 2,000 players at any one time
I play a game that tracks everything your character eats, drinks, visits, makes, wears, and fights
With one of those actions occurring for about every 5 seconds for 1,600 people at a time