Predictions of evolutionary outcomes based solely on genome sequence data is limited in experimental evolution studies. Gene-level parallelism that results in differing fates across populations and mutational cohorts formed via hitchhiking obscure our ability to assign likelihoods as to which lineages will survive. Furthermore, while initial fitness has been shown to be a strong predictor of final fitness in adapting laboratory populations of microbes, it is unknown how this phenomenon arises from a system's underlying genomic and physiological architecture when significant sequence-level stochasticity is observed. Here we explore the relationship between genotype and fitness by examining genealogies from a previous long-term evolution study in S. cerevisiae, in which whole-populations were whole-genome sequenced at various time-points over 1000 generations of evolution. We selected a spread of populations representative of observed dynamics in genomic sequence (hard sweeps, clonal interference, etc) and parallelism in mutation identity. We then isolated clones from whole population samples and performed whole genome sequencing and fitness measurements. Using the resulting genealogies and their associated fitness data, we examine the various population genetics factors resulting in the lineage dynamics of adaptation in this system.