Double strand breaks (DBSs) promote different repair pathways involving DNA end joining or homologous recombination, yet their relative contributions, interplay and regulatory interactions remain to be fully elucidated. These mechanisms give rise to different mutational processes and their propensity for activation directly affects genomic instability with implications across health and evolution. Here we present a new method to model the activation of at least three alternatives: non-homologous end joining (fast), homologous recombination (slow) and alternative end joining (intermediate) repair. We obtain predictions by employing Bayesian statistics to fit existing biological data to our model and gain insights into the dynamical processes underlying these repair pathways. Our results suggest that data on the repair of breaks using pulse field gel electrophoresis in wild type and mutants confirm at least three disjoint modes of repair. A density weighted integral is proposed as a tool to sum the predicted number of breaks processed by each mechanism from which we quantify the proportions of DSBs repaired by each. Further analysis suggests that the ratio between slow and intermediate repair depends on the presence or absence of DNAPKcs and Ku70. We outline how all these predictions can be directly tested using imaging and sequencing techniques. Most importantly of all, our approach is the first step towards providing a unifying theoretical framework for the dynamics of DNA repair processes.