The study introduces Tumor Immune Dysfunction and Exclusion (TIDE), a computational method to predict the response to immune checkpoint blockade (ICB) therapy in cancer patients. TIDE models two primary mechanisms of tumor immune evasion: T cell dysfunction in tumors with high cytotoxic T lymphocyte (CTL) infiltration and T cell exclusion in tumors with low CTL levels. By analyzing large tumor cohorts, TIDE identifies gene signatures that predict patient survival and ICB response. The method also reveals new candidate regulators of ICB resistance, such as *SERPINB9*. TIDE outperforms other biomarkers like *PD-L1* level and mutation load in predicting the outcome of melanoma patients treated with first-line anti-PD1 or anti-CTLA4 therapies. The web application and source code for TIDE are available at <http://tide.dfci.harvard.edu>.The study introduces Tumor Immune Dysfunction and Exclusion (TIDE), a computational method to predict the response to immune checkpoint blockade (ICB) therapy in cancer patients. TIDE models two primary mechanisms of tumor immune evasion: T cell dysfunction in tumors with high cytotoxic T lymphocyte (CTL) infiltration and T cell exclusion in tumors with low CTL levels. By analyzing large tumor cohorts, TIDE identifies gene signatures that predict patient survival and ICB response. The method also reveals new candidate regulators of ICB resistance, such as *SERPINB9*. TIDE outperforms other biomarkers like *PD-L1* level and mutation load in predicting the outcome of melanoma patients treated with first-line anti-PD1 or anti-CTLA4 therapies. The web application and source code for TIDE are available at <http://tide.dfci.harvard.edu>.