The aim of this exploratory study was to identify the most influential training designs during the final six weeks of training (F6T) before a major swimming event, taking into account athletes’ evolution over several seasons. Fifteen female and 17 male elite swimmers were followed for one to nine F6T periods. The F6T was divided into two sub-periods of a three-week overload period (OP) and a three-week taper period (TP). The final time trial performance was recorded for each swimmer in his or her specialty at the end of both OP and TP. The change in performances (∆P) between OP and TP was recorded. Training variables were derived from the weekly training volume at several intensity levels as a percentage of the individual maximal volume measured at each intensity level, and the individual total training load (TTL) was considered to be the mean of the loads at these seven intensity levels. Also, training patterns were identified from TTL in the three weeks of both OP and TP by cluster analysis. Mixed-model was used to analyse the longitudinal data. The training pattern during OP that was associated with the greatest improvement in performance was a training load peak followed by a linear slow decay (84 ± 17, 81 ± 22, and 80 ± 19 % of the maximal training load measured throughout the F6T period for each subject, Mean ± SD) (p < 0.05). During TP, a training load peak in the 1^st week associated with a slow decay design (57 ± 26, 45 ± 24 and 38 ± 14%) led to higher ∆P (p < 0.05). From the 1^st to 3^rd season, the best results were characterized by maintenance of a medium training load from OP to TP. Progressively from the 4^th season, high training loads during OP followed by a sharp decrease during TP were associated with higher ∆P.