Physical exercise has many benefits, but it might also have a negative impact on the body, depending on the training level, length of workout, gender, age and fitness. The negative effects of physical exercise are commonly attributed to an imbalance between the levels of antioxidants (both low molecular weight antioxidants and antioxidant enzymes) and reactive oxygen and nitrogen species due to excessive production of free radicals during physical exercise. In this critical review, we look for answers for three specific questions regarding the interrelationship between physical exercise and oxidative stress (OS), namely, (i) the dependence of the steady-state level of OS on fitness, (ii) the effect of intensive exercise on the OS and (iii) the dependence of the effect of the intense exercise on the individual fitness. All these questions have been raised, investigated and answered, but the answers given on the basis of different studies are different. In the present review, we try to explain the reason(s) for the inconsistencies between the conclusions of different investigations, commonly based on the concentrations of specific biomarkers in body fluids. We think that most of the inconsistencies can be attributed to the difference between the criteria of the ill-defined term denoted OS, the methods used to test them and in some cases, between the qualities of the applied assays. On the basis of our interpretation of the differences between different criteria of OS, we consider possible answers to three well-defined questions. Possible partial answers are given, all of which lend strong support to the conclusion that the network responsible for homeostasis of the redox status is very effective. However, much more data are required to address the association between exercise and OS and its dependence on various relevant factors.