Estimates have been made of 'attainable production' and 'production without protection' for a range of crops and the assumption that these could either be fully protected or not protected at all. The difference between the two then meaures the full extent of 'potential loss'. Actal production is also know (which always fall between the two limits as expected).
Prof Oerke adn co-authors at the University of Bonn in Germany has been doing this work for decades. See the book: Crop Production and Crop Protection: Estimated Losses in Major Food and Cash Crops by E.-C. Oerke, H.-W. Dehne, F. Schönbeck, A. Weber, Publisher Elsevier, 2012, 829 page; ISBN 0444597948, 9780444597946.
See also related work for example www.nrel.colostate.edu/ftp/conant/SLM-proprietary/Oerke_2006.pdf
It has to be borne in mind that the level of potential loss rates (and any measure of effectiveness of crop protection that stems from it) could be misleading. It does have a economic equivalent meaured as the physical output that is saved, but the important issue is whether that saving is worth achieving given the labour, expenditures for crop protection and other costs of the application.
The goal of crop protection is to prevent economic loss (not potential loss). If the anticipated physical yield is low for instance because it is limited by abiotic factors such as shortage of water it might be uneconomical to apply crop protection. There is an economic optimum between the intensity of crop protection and the remaining crop loss. Therefore the economics of crop loss protection in different production systems cannot be concluded from the actual loss rates. See e.g. Wossink & Rossing (1998) Journal of Environmental Management.