If we were broadcasting 200 & 200 ahead of corn, I'd agree. However we broadcast both P  & K separately, prescription spread, based on  4 yr soil test intervals and yield expectations. We spread on 2 yr intervals and always ahead of corn crop, with consideration for the following year bean crop. 

Some field areas receive no P, some no K, some areas relatively high rates. I'd guess that 75% of 0 would still equal 0. I'm familiar on how to divide the total amount of fertilizer products by the entire field acreage for an average, assuming it was blanket spread. The only thing I blanket spread is AMS or gypsum in the spring, also incorporated prior to planing. All P & K is fall applied, then incorporated via chisel plow (what we call a chisel plow in IL) and then field cultivated in the spring. I expect almost no amount of fertilizer remains on the surface. We have 0 no till acres. I also spread lime via Rx.

I'm not sure how those no-spread areas of one or both elements would work in the strip till environment where I believe one is counting on the "starter effect" of the constant banded product. Could be wrong though. Isn't the starter effect of strip placed fertilizer one of the goals?