This document presents a proposal of structuring the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) twice normalized. Instead of using Euclidean distances, TOPSIS was reckoned by the ellipse length because it was necessary to observe a distance without the basic assumption about to know the way where the “pulse” might go it. The computation was changed at the distances, using the elliptical distance. The dataset was observed by other works without grouping as it was possible to observe the distances towards the decision occurred and also the concentration of alternatives at same ranked place, forming a kind of “group” at this proposal, when TOPSIS-2NE compared to TOPSIS-2N showed grouped decisions while TOPSIS-2N kept different distances and places for ranking compensatory alternatives.