Macromolecular association commonly occurs via dynamic engagement of multiple weak bonds referred to as multivalent interactions. The distribution of the number of bonds, combined with their strong influence on the residence time, makes it very demanding to quantify this type of interaction. To address this challenge in the context of virology, we mimicked the virion association to a cell membrane by attaching lipid vesicles (100 nm diameter) to a supported lipid bilayer via multiple, identical cholesterol-based DNA linker molecules, each mimicking an individual virion–receptor link. Using total internal reflection microscopy to track single attached vesicles combined with a novel filtering approach, we show that histograms of the vesicle diffusion coefficient D exhibit a spectrum of distinct peaks, which are associated with vesicles differing in the number, n, of linking DNA tethers. These peaks are only observed if vesicles with transient changes in n are excluded from the analysis. D is found to be proportional to 1/n, in excellent agreement with the free draining model, allowing to quantify transient changes of n on the single vesicle level and to extract transition rates between individual linking states. Necessary imaging conditions to extend the analysis to multivalent interactions in general are also reported.