Background/Question/Methods In population models where all individuals within a given stage are subject to the same rates of growth, mortality, and fertility, the differences among individuals in their longevity and lifetime reproductive success (LRS) are due to luck. We now have well-developed methods for analyzing and partitioning the consequences of multiple types of luck in life histories. Does the importance of luck in lifetime outcomes depend on species traits or phylogeny? Are some life history strategies, growth forms, or reproductive strategies more sensitive to luck than others? We use the COMADRE and COMPADRE databases to investigate these questions across a wide range of plant and animal models. We define the "importance" of luck as the coefficient of variation of LRS and lifespan, and investigate covariates such as life form, generation time, position on the fast-slow/reproductive strategy continua, habitat or biome type, and phylogenetically-corrected functional traits.
Results/Conclusions The COMADRE and COMPADRE databases provide information on mean fecundity, but not variance. We find that our choice of variance/mean scaling relationship (e.g. Poisson or Bernoulli distribution) for reproduction is very important for our results. We find that luck plays a large role in longevity and LRS across many groups of taxa. The results of our meta-analysis have implications for population ecology and life history evolution.