17 to 18 F climbed the most while E dropping, while A and B stayed static.
18 to 19 F, B and A climbed while E dropped.
19 to 20 F, E climbed while A and B dropped.
20 to 21 F, A and B dropped while E climbed.
21 to 26 F and E climbed while A and B stayed the same.
25 to 27 E climbed, the rest remained static.
27 to 28 F and E climbed while A and B stayed the same.
Has been static since 28
See the influence changes of Faction F and E? That is likely pushing the influence around as you watch them climb.
The evidence is in how factions A and B don't climb at the same rates as E and F - and those fluctuate. Its not a steady climb as influence is drained.
If it was a "war tax" as you claimed, A and B should have received more and the influence from E and F should have received the reversed amount.
E had more of the system's share and should have climbed the most - instead it was F. Then the numbers were not static nor consistently even with a steady drain.
I've not put any sort of label on this - it's just a phenomenon that I have noticed on several occasions that performs consistently in many more examples than the one I have given.
What one has to explain is the standardised behaviour that occurs between the 21st and the 28th. The two explanations are as follows:
1 For seven days a player is active in the system and performs exactly the same tasks. Each day faction E pulls exactly 0.3% from the conflict and F pulls exactly 0.1%. This activity stops on exactly the same day that the two warring factions drop to 1%.
2 There is no player activity from 20/12 to 4/1 (actually, nothing changes until the 7th). During the conflict, factions C and D daily lose 0.2% each. This 0.4% is proportionally distributed among the other factions in exactly the same way that a retreating faction's remaining influence is distributed when it leaves, with the leading faction getting the most and the trailing faction the least. We're talking very small fractions at the bottom end of the table - not enough to round up the decimal.
I'm in a very quiet sector with very few visitors, usually itinerant traders/mission takers. When I first saw this I also thought it was down to player activity, but having seen it on more than one occasion, sometimes lasting more than 14 days, I drew different conclusions. Unfortunately, I only keep records for the past 28 days and this example is the only one in the current data set out of 13 violent conflicts.