Fishing selectively removes phenotypes, altering phenotypic variability. The loss of phenotypic variability can arise through direct removal of fast-growing individuals, known as the Rosa Lee phenomenon, or through indirect, evolutionary changes, known as fisheries-induced evolution (FIE). It is necessary to disentangle these processes because they have different implications for population resilience and fisheries management. The southern sand flathead (Platycephalus bassensis) is Tasmania’s most recreationally fished species; however, fishing pressure varies across the state. In southern Tasmania, fishing mortality is 5 times natural mortality, and the observed size-at-age is significantly smaller than the comparatively lightly fished Flinders Island. It remains unclear whether this difference reflects Rosa Lee phenomenon, FIE, or just different environmental conditions. In this study, we reconstructed individual growth trajectories from annual otolith increments of ~1000 sand flathead. Using nearly 15 years of data across four Tasmanian populations, we examined inter- and intra-population patterns in growth and by extension size-at-age variability. We developed a novel statistical growth model to describe individual growth trajectories and showed clear inter-population differences in growth patterns. The three heavily fished populations showed faster declines in growth from age 3 onward, resulting in overall slower growth relative to the lightly fished Flinders Island population. At the intra-population level, age-at-capture in all four populations was negatively correlated with initial growth rates, indicating that faster-growing individuals were more likely to be captured earlier in life. Rapid decline in somatic growth in intensively fished populations is consistent with expectations from FIE, but environmental effects cannot be ruled out. Yet, the quantified Rosa Lee effect was weak even in intensively fished populations. This may reflect our model’s limited ability to reconstruct expected growth trajectories under unfished conditions due to the lack of juvenile sand flathead samples. Nevertheless, our novel individual-level growth model benefits recreational fisheries management well beyond Tasmania.