I had a training partner many years ago who used to torment me by doing workouts like 7x1000 or 13x400. I'd be doing 6 or 8x1000, and 12 or 14x1000 – those numbers just seemed more natural to me. I was thinking about this the other day, thanks to an article about odd and even intervals by David Lowes, the coaching editor at the British track and field magazine Athletics Weekly. Amby Burfoot sent the link to me and a few other friends, noting also that American marathon training plans tend to call for 20-mile long runs while European plans called for 30K. "Which," he asked with tongue firmly in cheek, "is morally superior?"
Lowe points out that, endurance-athlete OCD notwithstanding, you should set your workout based on what your body can handle and what training stimulus you're looking for, regardless of even or odd. There's no arguing with that: clearly your body doesn't care about odd and even. But, just for kicks, I'll offer a partial defense of sticking to "round number" workouts in certain contexts. It's one thing to decide, "I think the best workout for my needs today is 7x1000," and then do it. But the more common situation, at least in my experience, is that people set a rough goal – around 8 reps, say – and then decide when they get to end of the workout whether to do 7, 8, or 9 reps. I think there are some drawbacks to this approach.
One of the essential skills in racing is learning to distribute your effort so that you maintain the fastest possible pace for a given distance without blowing up. This razor's-edge balance is a learned skill, and it's something you develop in training. One way to develop this skill is to have a specific goal in mind for how many reps you plan to do in a workout – and be accountable for the pacing decisions that enable you to reach that goal. If you run the intervals faster than you can sustain and have to cut the workout short, you've screwed up. Same thing if you run them slower than you can sustain (relative to the workout's desired effort level) and feel the urge to add an extra interval. Either way, you're letting yourself off the hook.
Now, like any good rule, there are plenty of reasons to break this one. If you feel an injury coming on, stop. If you're feeling burned out and realize that you're simply not recovered enough to complete the desired workout, stop. These situations will crop up for every runner now and then, and it's important to recognize them. But they shouldn't be a pattern: if it's happening every couple of weeks, there's something else going on. There are also specific types of workout that might have an open-ended number of reps, like seeing how many 400s you can complete at goal mile pace before your pace slows by more than two seconds or something. Again, exception rather than rule.
(There are fewer good reasons that I can think of to add extra reps. Maybe a workout in which your coach says to do 5x1600, and then when you're finished he asks you to do one more – that's a great way to learn about mental and physical limits, but it's not an every-week kind of thing.)
The bottom line, for me, is that by the time you're about a third of the way through a workout, you should have a pretty good idea of whether your pace is sustainable. Pacing studies have shown that perceived effort rises in a fairly straight line from the beginning of an exercise bout to the end, reaching maximum roughly at the finish. As Ross Tucker and others have proposed, you're constantly (and automatically) comparing your effort at any given moment with an internal template of how you expect your effort to feel at that point in the race/workout. If you're consistently failing to correctly judge whether your pace is appropriate in workouts, it doesn't bode well for races.
So that's my argument against "accidental" odd-number workouts. As I said above, intentional odd-number workouts are fine. And, of course, multiples of five count as "even" for intervals!
***
Read Other Hearst Subscriptions, and follow the latest posts via Twitter, Facebook, RSS, or (new!) weekly or monthly e-mail digest.
