The primary reason for baseline deficiencies in both the general population and in serious athletes is that it's difficult to get proper amounts of zinc and magnesium solely through diet alone. USDA studies show that 68% of self-selected diets contain less than two-thirds of the RDA for zinc  and 39% contain less than two-thirds of the RDA for magnesium.  While zinc and magnesium are contained in a wide variety of foods, it's been my experience that athletes don't acquire sufficient quantities through their normal diets. One reason may be that foods high in these minerals aren't necessarily the most desirable. For example, the best food sources for zinc include oysters and beef liver. These foods aren't consumed by most athletes, nor should they be.
Modifications of enzyme activities (creatine kinase and its B subunit; adenylate kinase; hexokinase; phosphofructokinase; lactate dehydrogenase; malate dehydrogenase, isocitrate dehydrogenase; citrate synthase; acetylcarnitine transferase; beta-hydroxyacetyl-CoA dehydrogenase; cytochrome c oxidase) in gastrocnemius muscle and myocardium were reported after two forms of training with or without administration of anabolic steroid. Endurance training was on a horizontal motor-driven treadmill, 2 km X hr-1, 5 days a week for hr per day for 5 weeks. In the case of power endurance training there was a slope of 45 degrees. Enzyme activities in controls and treated guinea pigs, as well as treatment-induced enzyme activity changes are time dependent. Some of these activities correlate linearly with one another; such correlations characterize the effect of adaptation. Endurance training and power endurance training in this study induce similar modifications and seem to differ essentially in the daily work load. The anabolic steroid methandrostenolone (dianabol) induces modifications which training does not bring about but which training at least partially eliminates.