Given the current state of research, I'd say that optimizing exercise for its effects on longevity is as much a fool's game as optimizing diet - if you want to take it on as a hobby, then by all means, but don't expect to beat the scientific community in terms of finding a better way, or to know how well you're doing. Obtaining significant benefits to life expectancy is easy: just exercise as recommended by physicians, the standard 30 minutes of aerobic exercise a day. The tricky question is whether there is a reliable way of gaining more expected years of life than are provided by that 80/20 position. But the research keeps rolling in, so perhaps one day there will be sufficient weight of evidence to say in confidence that one way of exercising is significantly better than another: "A study conducted among cyclists in Copenhagen, Denmark showed that it is the relative intensity and not the duration of cycling which is of most importance in relation to all-cause mortality and even more pronounced for coronary heart disease mortality. The [study] concluded that men with fast intensity cycling survived 5.3 years longer, and men with average intensity 2.9 years longer than men with slow cycling intensity. For women the figures were 3.9 and 2.2 years longer, respectively ...The groups were adjusted for differences in age and conventional risk factor levels. Current recommendations prescribe that every adult should accumulate 30 minutes or more of moderate physical activity in leisure time, preferably every day of the week. The optimal intensity, duration and frequency still have to be established. ... this study suggests that a greater part of the daily physical activity in leisure time should be vigorous, based on the individuals own perception of intensity."