Are you looking for a shortcut? Are you not very talented? Maybe you should read this article instead! Five steps to success at high school distance running
Writing the perfect training plan is easy. There are plenty of philosophical debates on the internet about the finer points of workouts, paces, etc., but most coaches can come to some agreement on what works—on paper, at least. The problems arise when you try to take these perfect training plans and start using them in the real world.
Will most 5k and 10k runners improve if they could run 100 miles a week with a 20mi long run, a 10mi tempo run, and a long interval session every week? Sure. But the problem is that most can't do this without getting hurt. Too many runners get caught up trying to emulate a perfect training plan, then get frustrated when it doesn't work out. If you've been in the sport long enough, you've probably known several talented runners who keep banging their heads against the wall, trying the same training plans over and over, hoping that if they can just stay healthy for once, they'll get really fast. Strictly speaking, this is true: if you're really talented, and if you get a long block of great, uninterrupted training in, you're going to get very fit. It's that second "if" that's usually the problem.
Especially if you are a naturally talented runner—and by this I mean that your baseline ability to cover long distances at great speeds is strong, i.e. your natural aerobic fitness is good—nothing matters if you can't stay healthy. If you or a runner you coach is very talented but also very injury-prone, what is the right way to approach training?
Training for the injury-prone talent
The first thing you need to do is forget everything you've learned about normal training. Pretty much any book on training for distance runners (including mine) will lecture you about the importance of a big aerobic base, usually built from high mileage and prodigious high-end aerobic workouts. Again, if you're a very talented runner, you benefit a lot from this, but if you can't run 40 miles a week without staying healthy, forget 80 miles a week for now.
I recall advising a DI runner who had suffered ten stress fractures in five years of running. Despite this, he'd managed to run 2:25 in the 1000m and under 8:30 in the 3k. Though his mileage was, overall, fairly low—50-70 miles a week—he nevertheless suffered constant bone stress injuries. If you were his coach, wouldn't it make sense to try something different after stress fracture #3 or 4? On paper, yes, 50 miles a week isn't enough to run a great 3k or 5k. But no matter how much you run, you'll never race well if you are injured. Is there another way?
A better approach is to return to a first-principles approach to understand what you need to run fast. In a standard training methodology, a large volume of easy to moderate running allows you to do faster high-end aerobic workouts, which in turn enable you to run race-specific workouts and the race itself. Renato Canova illustrates this simply and beautifully:
If I use fast intervals, I train for improving my race performance
If I use long fast run, I train for becoming stronger in my specific training
If I use long run at moderate speed, I train for increasing the base that makes me able to run long and fast
For a very talented but injury-prone runner, you can significantly cut injury risk by using talent as your base—instead of running high mileage and a lot of aerobic workouts, you can very low mileage and only sparingly (but not never!) go for aerobic workouts, relying instead on your high natural aerobic level to enable your race-specific work. You can supplement this by doing easy to moderate cross training workouts, e.g. on the bike or in the pool, to improve your basic aerobic abilities. All distance runners need to be able to run continuously for a long time, but the talented-but-fragile runner don't need to actually do it on a regular basis.
Let's look at a concrete example. A standard sub-16 high school cross country runner's schedule during the early pre-season might look like this: