1) average earnings over the player base
2) DISTRIBUTION of individual average earnings ... it isn't clear from your 5 to 10% number whether 20% are seeing significantly increased rewards while 80% see a reduction.
3) Earnings by mech class
4) You need to assess whether the role rewards are consistent with actually playing and winning the matches ... if 3 lights go out to scout on one side but die while doing so since they are operating away from the group ... does this behaviour result in the team losing while lights on the other side that "protect" ... end up winning?
5) Are the role rewards accessible to all weight classes? Assaults for example will have trouble getting any bonuses for "hit and run", "flanking", "scouting" ... are there rewards that they CAN receive to make up for the deficit in kills and assists?
There is so much information you need to analyse to determine whether these changes are actually beneficial to the majority of the player base. In addition ... keep in mind that this game does NOT have an economy ... cbills are never transferred between players ... the only thing you are looking at is the average time for players to earn enough cbills to buy new stuff as a competing factor against MC purchases.
If you introduce a change that increases the spread in earnings .... some players make much more and some make much less ... but the average is the same ... all you have done is DECREASED the MC expenditures on the game ... those making more have no reason to spend and those making less will simply quit because the game feels so unrewarding.
It is a balancing act. In a few days you will hopefully have the numbers to make a proper assessment
