Track and field events place diverse demands on athletes’ physical fitness. Traditional training methods have limitations, including fixed patterns, low intensity and neglect of individual differences, which make it difficult to cope with dynamic changes in events. In this study, a time series model was used to analyze the fluctuation of physical fitness of track and field athletes in different training cycles, and to explore the scientific methods to optimize the training effect and athletic status. The study selected 22 athletes at level 2 and above in track and field sprinting, jumping and hurdling events from a university, randomly divided into 11 athletes each in the experimental group and the control group, and implemented an 8-week comparison experiment. The data were analyzed by paired-sample t-test and independent-sample t-test to compare the changes of athletes’ physical fitness parameters before and after training, and time series correlation analysis was applied to investigate the relationship between coaching quality and athletes’ development. The results showed that the experimental group was significantly better than the control group in the standing long jump, 100 m and high jump (P<0.001), in which the average performance of the experimental group in the standing long jump amounted to 2.89±0.68 m, the average performance in the 100 m was 12.32±0.47 s, and the average performance in the 110 m hurdles was 20.07±1.04 s. The time-series analysis showed that the average performance of the new athletes in the 4th year of the team with the retired allocation numbers showed the highest correlation (0.653), while the effect of highly educated coaches on the level of athletes' grades also peaked at year 4 (correlation -0.689). The study shows that scientific cycle training combined with time series analysis can effectively improve athletes' physical fitness level, while the optimization of the training system needs to take into account the professional quality and practical ability of coaches, and at the same time, the cycle of athletes' success is about 4 years, which provides data support for the training planning of track and field.
