Things That Don't Matter Much
Notice that I said 'Much.' There may be differences but, according to the research, said differences are likely to be marginal or negligible, hardly significant for the average trainee:
Rep Tempo: I can't say that I've seen any evidence that fast reps make you faster, and they may be potentially more injurious. For my money, if you want to run faster, do barbell squats and practice your sprinting technique. Keep it simple.
Number of Sets: Nope, doesn't matter that much, this based on a broad compilation of studies comparing multiple vs. single sets. The only thing I'd add here is that if you're going to opt for the single-set format, make your one set count!
Rep Range: You've heard it said that low reps are best for strength and high reps best for muscle endurance. This may be generally true. But as Jim Wendler points out, if you go from being able to deadlift 315 for 5 reps to 315 for 10 reps, guess what ? You've gotten stronger! Also, think you can progress from five reps of one-armed assisted pushups to a one-armed pushup ? Probably not. You're probably going to have to build up reps at one stage before you can progress to the next.
Exercise Choice: While certainly true that some exercises are more productive than others, there is no exercise that the average trainee "Can't do without." Or do you really think that you're going to suffer if a rack pull is a better choice for you than deadlifting from the floor ?
Workout Frequency: Honestly, I've seen people thrive on one workout a week and I've seen people thrive on doing the same damned workout five days a week.
Variety: Probably overrated. Perhaps even vastly overrated.
Food Choices: Hear me out on this. For those wanting to lose weight, the most 'unhealthy' aspect of the American diet BY FAR ... is that it contains too many calories!