The Long Road Home - Experiment

Imagine you've been running the same distance at the same pace every day and you decide to train for a marathon. You know one of the challenges is running the distance, so once per week you run longer, adding a mile per week. If you run at the same pace, you're going to have a problem going further, so you run slow enough to get your long run done. If you do these slowly enough, with bathroom breaks and/or walking stretches, you might actually lose fitness by your weekly average getting worse. [I've done this, training to run 100 miles.] If your fitness is actually improving, you should find that you're running faster on the other six days, or at least running the same pace more easily.

Now imagine that you've heard about doing short hill sprints and you want to know if that will improve your marathoning. You need something measurable, not just "run short hill sprints," so you use the same hill, with landmarks for the start and finish and measure the time it takes you to finish, say, 10 repeats, including the time spent going downhill. Assuming nothing goes wrong (bad weather, injury), you'll probably improve the second and third time you do them, just from learning how to do them.

These hill sprints have reportedly been very beneficial for some marathoners. Let's consider why they might help. 50% of cases could be coincidence or placebo effect, but that still means that it works for some. Brad Hudson has said that a 1% improvement in 100m sprint time means a 1% improvement in marathon time (all other things being equal). This is not true. As in all things, a few runners will improve greatly, most will see a very small improvement and some will see no improvement or even get worse. There are mechanisms that might apply, however. High mileage runners after a number of years tend to develop a slower step cadence, lessened knee lift and less flexion of the ankle at push-off, all of which have to be corrected to sprint well uphill. Less seasoned runners don't seem to respond as well.

But will it work for you? If after doing the first few sessions of hill sprints your improvement stalls, it probably won't. But maybe you keep improving, maybe dramatically; even then, there's a problem. Consider this: there are multiple barriers that keep you from being a world-class marathoner, but it's the one you hit first that keeps you from improving and it's probably not the same one that keeps you from being a world-class sprinter; you know this intuitively, as there's no overlap between world-class marathoners and world-class sprinters. So, maybe improving your hill sprints just makes you a better sprinter. The way you know is the same as what I said about the long run - if your other runs get faster too, this was your barrier. If they get worse, it's probably because you're improving at this one thing at the expense of your other runs, either by inadvertently taking it easy the day before and "keying yourself up" for the sprints, and/or running slower the next day or two, trying to recover.

When you find what works for you, stick with it until it stops working. Then experiment with something else.