First, on individual rotation which we employed during WW II, Korea and Viet Nam. It is probably noteworthy that two of the best divisions in Europe, the 3d ID and the 82d refused to accept any replacements above the grad of Private or 2LT; they promoted from within. In Korea, individual replacements were a detriment to full combat effectiveness as units effectively turned over 10% per month; Viet Nam was even worse because there was also the infusion program which meant an effective turnover of over 120% per year for most combat units.

Your suggestion of packet replacement has also been tried with the COHORT process -- that didn't work even in peace time.

Due to the above cited experiences, we have since gone to unit rotation and that needs to stay in place. Contrary to your assertion, the Army and Marines have not made a huge mistake in rotating units; they have improved combat effectiveness by an order of magnitude. Yes, there are personnel changes within units between rotations but the unit does get locked for its final trainup and, far more importantly, there is little personnel turbulence while committed. I'll acknowledge that not putting units back in the same AO is questionable but the rotation between AOs is designed to enhance flexibility in planning, assimilating and working with other units; it does that at only slight cost to combat capability. The enhanced capability it provides as a training vehicle for other wars in other places -- and there will be some -- is significant.

Having deployed to two wars, in each case once as an individual replacement and once with a unit; I have absolutely no doubt that individual replacements should be avoided at all costs.