Yes, we should in this case because it is not arbitary.
Consider that the 30th April 1864 monthly report. There was an argument between Humphrey and the Secretary of War as to the strength, and Drum dug the following out of the files:
View attachment 520441
Per regulation, the PFD is the sum of first two lines, which the SECWAR quoted. It gives 121,964 Present for Duty, including all officers and men on duty whether they be assigned to special, extra or daily duty or not. However, the OR lists only the first value.
The PFDE, after excluding logistics etc. is 97,273. That is approximately 80% of the PFD, or almost exactly what McClellan said should be removed to account for logistics and all minor details that were extra duty etc. (deduct 1/6th for extra duty etc, and 1/5th for all-up). This represents the maximum possible combat strength, assuming no casualties, no details, no straggling etc. You will note that the PFDE of 1864 has already removed these. In comparison, those of 1862 had not done this, and the PFD and PFDE figures are thus calculated differently and represent two different things.
The PFDE of 1864 is basically calculated the same was as PFD on rebel ANV returns, and are directly comparable. Those of 1862 are not, and 1/5th or 1/6th needs to be removed to account for extra-duty etc. This gives a figure comparable to 1864 figures, and rebel figures.
The actual number brought into battle will be a figure lower than this. How much lower depends on the situation. Armies on the move, and especially those making forced marches will have a much lower figure, whilst those defending a static position will have much better retention.
For the most part, the rebels gave us their "carried into battle" figures for the Seven Days - about 90,000 with officers and arty.
We don't have the same value for the Federals. If we make reaonable estimates, then the numbers would be around 70-75,000 in the same category.