Compare performance only when wear conditions are reasonably similar. Record projection and persistence in separate fields, then reject a ranking if sessions differ too much in application, environment, or timing. Context mismatch should remain a limitation, [https://harryWww.mainArcadetimecapsule.jpcom:443/mediawikiwiki/index.php/Affordable_Tuscan_Leather_Alternatives:_IMIXX_No.03_Fragrance_Review Build_A_Multi-Candidate_Leather_Shortlist_Around_Texture,_Warmth,_And_Wear parent resource] timing. Context mismatch should remain a limitation, not be hidden.