Experiential Bias and Reverse Periodization

When a cyclist or triathlete tries to tell me how I should train as a runner, it bothers me because they come from a background that carries baggage with it (if I hear about power meters again, I'll scream). This was true 50 years ago when Jim Ryun's high school coach, a swim coach, had him run "swim" workouts like 50x200m and 20x400m at pace on consecutive days; that it worked for him is remarkable - no one who has tried to replicate his workouts has succeeded. What hadn't occurred to me was that, as someone who ran organized track since the age of 12, I also had similar baggage. When I started running, all distance runners were converted track runners and most tried to train for races like marathons as if they were just long track races. Make it 100 miles on a hilly course and the problems compound.

The standard idea in track for periodization has been that one can only do a little of short, fast workouts before becoming stale or injured, but stamina and endurance take a long time to develop, so one runs a lot of long slow distance and then adds a few weeks of fast intervals before racing. This is still a relevant way to train for middle distances.

Reversing that periodization by putting the speedwork first has become the standard for marathon runners, as the speed work is less specific. There have been many rationales and protocols developed, some of which are disreputable. For example, there are those who would have one perfect a single step at top speed and then try to gradually lengthen the time one can run fast. The best-known marathon plan of Jack Daniels (1996) has six weekly interval workouts 12-18 weeks out, before switching to all threshold and marathon pacing.

Brad Hudson has what he calls "non-linear periodization," where there are not set blocks of training for one specific facet. What he has wedded onto this is starting with short hill sprints and long runs, then working from both extremes to a more specific middle ground. I like this because it looked like what I was doing, namely running a variety of workouts each week, but emphasizing different ones at different times.

In other words, I fell into the trap of agreeing with those who justified what I was already doing: experiential bias.

There are some basic questions about reverse periodization that are left unanswered:

1) It is suggested that highly trained athletes can only improve by pushing each type of training as far as possible and that this cannot be done if one is trying to improve two types at the same time, hence the need for periodization. Even if this is so, does it justify the increased risk of injury that comes from abruptly switching from one type of training to another?

2) It is suggested that the benefits in each type of training accrue to the next. Let's say that you do maximal oxygen uptake intervals, then short threshold runs, then runs at marathon pace. If you raise your VO2max as high as possible, then abandon those workouts, you'll lose some of that ability over time before you race. Do those workouts cause one to be able to do better threshold runs than if one had devoted the entire time for both to doing just threshold runs?

There's no way to answer these questions, but the "start fast" idea has an added benefit. Right now in Minnesota, it's about zero degree Fahrenheit. Running short and fast seems preferable to long slow runs in extreme cold, just as a matter of comfort (and survival). This is also a long time before I'd race, so the timing works for me.