One way to approach this question is to think of chromatography as a competition between the solvent and the analyte for binding to the stationary phase. In reverse phase chromatography water (being the most polar commonly used solvent) is less able to compete against the analyte than acetonitrile is. Therefore, the analyte stays stuck to the stationary phase most of the time. I realize that this model ignores the interaction between the solvent and the analyte, but it works for many situations in both reverse phase and normal phase chromatography, although I can think of one situation that it does not explain well. If someone with more knowledge of chromatographic theory wanted to offer further thoughts, that would be great.