The behavioral test described by Porsolt in 1977 for screening potential antidepressant drugs is extensively used both in basic research and in the pharmaceutical industry. The measured behavior is the immobility time during the swimming test (preformed in rodents), which decreases upon acute antidepressant treatment. Several research groups have suggested some modifications on the original Porsolt paradigm and its analysis. Nevertheless, there are still inaccuracies resulting from either undefined intermediate behaviors or from considering the movement of the whole body as one unit without analyzing the motion of the limbs. Herein, we propose a novel and simple scoring method, based on continuous measurement of the limbs motion, using a joystick, a computer screen and simple software. We validated the method, using antidepressant drugs and studied examples of false positives and false negatives of the traditional Porsolt paradigm. The proposed method is easy to use, it accounts for all range of movements and the analysis is relatively fast. Moreover, the results obtained using this analysis method show a normal Gaussian distribution in a population of rats (while the traditional Porsolt analysis does not) which allows selective breeding of 'motivated' and 'depressed' lines of animals.