One main idea guided this research article: can feedforward neural networks (FNN) be used by investors to cross-validate their investment decisions and stock market strategy entry-signals. Therefore, the research question of this study was: how profitable is a break-out strategy if the strategy is cross-validated by a FNN for U.S. stock markets? The study followed a quantitative, quasi-experimental design, regarding in-sample and out-of-sample tests. For the method of analysis, five hidden layers of a FNN were computed, a sigmoidal function, and break-out strategy conditions as well as entry-signals. As a result, the break-out strategy was profitable to trade with a money management strategy; however, the transaction costs had an effect if for the exit strategy the assets were sold at the end of the day. The FNN could only provide cross-validating results if a dichotomies entry-signal variable was added to the model. The study is relevant for portfolio managers and investors, who are interested in a second assessment of their data or market entry decisions through a cross-validation performed by a FNN.