Collaborative biostatisticians in academic health centers are confronted with a unique set of challenges to implementing best biostatistical practices. These challenges yield gaps between ideal and actual practice with regard to designing studies, conducting reproducible research, formulating and implementing analysis plans, applying optimal analytic tools, and disseminating findings. These gaps may represent logistical and practical challenges of carrying out a set of experiments, a collaborator's comfort level with proposed tools, and/or unexpected findings that arise during the study. Training the next generation of collaborative biostatisticians in how to minimize these gaps is critical for allowing biostatisticians to engage in clinical and translational research that has the utmost integrity and to ultimately lead with statistics. Solutions to these issues involve not only understanding the implications of the gaps, but also flexibility and compromise, that are much more likely in the presence of a deep level of iterative collaboration. This proposed panel will include expert guidance from collaborative biostatisticians confronted with these challenges.