Slitherine

I don't understand this.
Here is a unit shooting at full strength, for 214 shots they get 6-22 casualties. Here is a unit shooting on low ammo, for 94 shots they get 0-6. But if you reduce 6-22 by a ratio of 94/214 you would get 3-10, not 0-6.
I don't think this is a display issue, it does seem that these half bow units when shooting on low ammo are shooting at something around 25% rather than the 50% of full firing that I would expect.

I think the issue of 50% bowmen units is a red herring. The way the calculations work would not selectively disadvantage 50% bow units.

It is more a question of fractions being rounded down during the calculations, which does result in low ammunition shooting being less than 50% of full ammunition shooting. (Because the engine only handles integer variables, when an interim "damage" value is halved the fraction is discarded before the overall casualty numbers are multiplied up again to get the actual reported casualties. This probably has more effect on lower overall numbers.)

Here is an example with identical units. The low ammunition casualties are slightly lower than half the full ammo casualties. 43% rather than 50% in this example.

It would be possible to eliminate most of this rounding "error" by doing the calculation slightly differently, but given that some people would like to see low ammunition shooting effectiveness reduced or even eliminated altogether, I don't see the apparent discrepancy as a bad thing.