With　the　continuous　development　of　society,　the　cooperation　of　different　dimensions　is　urgently　needed.　Analysis　and　modeling　of　team　cooperation　model　and　performance　evaluation　are　especially　important　for　competitive　sport.　In　this　paper,　a　football　team＇s　attacking　mode　and　the　team　performance　were　assessed　using　network　science　methodologies.　The　match　process　was　analyzed　by　using　the　data　of　Team　A　(given　in　the　form　of　attachment　due　to　excessive　file　size)　and　the　method　of　complex　network　science.　Each　player　was　regarded　as　a　node　in　the　network,　and　the　interaction　among　players　was　considered　as　the　connection　to　the　network.　This　method　directly　reflected　the　favorable　formation　of　the　team　and　the　interaction　frequency　among　members.　Then,　a　team　performance　evaluation　model　was　established　using　the　backpropagation　neural　network　(BPNN)　and　the　uncrossed　analytic　hierarchy　process　(U-AHP)　method　based　on　the　factors　including　the　number　of　passes　and　successful　pass　rate.　The　team　performance　was　comprehensively　rated　from　two　levels:　member　and　team　level.　Analysis　from　established　models　indicated　that　Team　A　had　a　higher　probability　of　winning　when　using　the　＂4-4-2＂　offensive　strategy　and　performance　evaluation　analysis　indicated　that　more　passes　and　higher　pass　success　rates　were　more　beneficial　to　win　the　game.　Following　the　model　developed　in　this　study,　some　suggestions　were　given　from　the　perspectives　of　team　strategy,　attack　mode,　cooperation,　and　incentive　mode.