Profiled sheets are widely used in modern steel structures, either as cladding or as casing in composite structures. Their strength calculation represents a complex task because one must deal with thin-walled structures that have complicated cross-section shape. Manufacturer’s catalogues provide data about their strength, mostly for continuous surface load. These data are mostly obtained by testing. EUROCODE EN 1993-1-3, i.e., its Annex A2, regulates the testing procedures for profiled sheets, allowing two main approaches regarding load application: uniformly distributed load and equivalent line load (in four locations). In addition, the mentioned code proposes roller supports that simulate pinned joints at the ends, neglecting the fastening conditions, which unavoidably are present with these structures. Aim of this research was to prove if the two proposed load patterns produce identical results, and to reveal how fastening devices affect the structural strength. In this research, the Finite Element Method (FEM) analysis with geometrical and material nonlinearity and contact analysis in the support zones were applied for the strength calculation of one typical profiled steel sheet. The analysis was conducted for ultimate load strength and for maximum load at standard deflection of L/200. In order to provide the testing conditions as real as possible, the support conditions were set in two ways: a) with support width B=40 mm and N=2 fasteners at the support, i.e., one fastener in each outer trough; b) with support width B=200 mm and N=8 fasteners at the support, i.e., two fasteners per every trough. The first support case is intended to simulate behavior close to a simply supported beam, and the second should simulate a beam fixed at its ends. The analysis encompassed the two load patterns provided by EUROCODE standard, and the results were compared. The results reveal that the surface load approach gives higher strength values than the four-point-load, both for ultimate load and for maximum load at deflection of L/200. This stands for both analyzed support conditions. The results of the research indicate that the codes should much more precisely define the testing conditions for such structures and make them closer to reality, for the purpose of more reliable and economic design.