We introduce a novel indirect method of estimating the pollen dispersal curve from mother–offspring genotypic data. Unlike an earlier indirect approach (TwoGener), this method is based on a normalized measure of correlated paternity between female pairs whose expectation does not explicitly depend on the unknown effective male population density (d_e). We investigate the statistical properties of the new method, by comparison with those of TwoGener, considering the sensitivity to reductions of d_e, relative to census density, resulting from unequal male fecundity and asynchronous flowering. Our main results are: (i) it is possible to obtain reliable estimates of the average distance of pollen dispersal, δ, from indirect methods, even under nonuniform male fecundity and variable flowering phenology; (ii) the new method yields more accurate and more precise δ-estimates than TwoGener under a wide range of sampling and flowering scenarios; and (iii) TwoGener can be used to obtain approximate d_e estimates, if needed for other purposes. Our results also show that accurately estimating the shape of the tail of the pollen dispersal function by means of indirect methods remains a very difficult challenge.