Employing the stochastic approach to chemical kinetics, an improved statistical technique is presented for the experimental design and analysis of receptor-mediated adhesion. Explicitly utilizing the randomness of transient pausing of cells in flow due to receptor-interactions, we apply the method of maximum likelihood to develop a robust statistical point estimate of the rate constant for the dissociation of receptor tether-bonds. The method additionally suggests an exact approach to the calculation of confidence intervals, and offers a direct methodology for the testing of hypotheses regarding the effects of force applied to tether bonds as well as the effects of multiple bonding.