Patients with sepsis often experience reductions or increases in platelet counts, but the implications of these temporal patterns on prognosis remain unclear. The aim of this study was to investigate the impact of changes in platelet trajectories on the clinical prognosis of sepsis. This study was a retrospective analysis using data from the Medical Information Mart for Intensive Care (MIMIC)-IV database. Patients with sepsis were identified from the database, and their platelet trajectories were categorized into four distinct models based on the changes in platelet counts over a period of 14 days post-diagnosis of sepsis. The effect of these trajectories on patient prognosis was subsequently evaluated. A total of 15,250 patients with sepsis were included to construct a model, and the following four distinct platelet count trajectories were identified: normal platelet levels (phenotype 1); persistently low platelet levels (phenotype 2); gradually increasing platelet levels exceeding the normal range (phenotype 3); and consistently significantly elevated platelet levels (phenotype 4). Statistically significant differences were found in the 28-day mortality, in-hospital mortality, and 90-day mortality among the four phenotypes. Multivariate regression analysis showed that compared to the group with normal platelet levels (phenotype 1), the group with persistently low platelet levels (phenotype 2) had higher in-hospital mortality (odds ratio [OR] = 1.34, 95% confidence interval [CI]: 1.16-1.54), 28-day mortality (OR = 1.69, 95% CI: 1.47-1.94), and 90-day mortality (OR = 1.50, 95% CI: 1.32-1.69). There was no difference in in-hospital mortality between phenotypes 3 and 4 compared to phenotype 1, although phenotype 4 showed an increase in 28-day mortality (p < 0.05), and phenotype 3 showed a decreasing trend in 90-day mortality (p < 0.05). The results of inverse probability weighting adjusted by regression were basically consistent with the above findings, except that there was no statistical difference in 28-day mortality between phenotype 4 and phenotype 1. In the subgroups based on age, weight, and antiplatelet drugs or therapies, there was an interaction between platelet levels and these factors. In patients with sepsis, a decrease in platelet count is associated with increased mortality, while a moderate increase in platelet count can reduce 90-day mortality. However, for patients with persistently elevated platelet counts, caution is advised when using antiplatelet drugs or therapies, as it may increase mortality.