Friday, April 10, 2009


I was talking to an athlete last weekend about structuring their weekly training plan, and remembered that I never finished the final blog post on microcycle structure. So here we go....

A few notes to start:

1) I've written this as a three-part post. See Part 1, and Part 2, for some background.

2) I found a major error in the Part 2. I had written that Foster (1998) calculated Monotony as the mean daily training load multiplied by the weekly standard deviation of the daily training load. In fact, Monotony is mean training load DIVIDED BY standard deviation of the daily training load (now corrected in part 2).

3) Session RPE is not without is flaws, not the least of which is that it definitely underestimates the training load of intense workouts. But I still find it a quick and easy way of approximating load (more below).

4) I was reminded last weekend that there are few absolutes in coaching. These thoughts/ideas are just one way of doing things; the concepts are based on some good research, but theory and practice are not always the same thing. I happen to think this approach would work well for many athletes, but most importantly, it works well for OUR PROGRAM. If it appeals to you, give it a try, and see what happens.


If you've read (or remember) Part 1 and Part 2, you'll recall the training study with the horses in Part 1. The researchers designed a training protocol to induce overtraining, in order to better understand the process. In a nutshell, a group of racing horses were trained every day, one hard day, one easy day, with an increasing training load on the hard days, and no change in training load on the easy days. For 260 days, the horses thrived on the program. On the 261st day, the load on the easy days was increased; the time of the session was held constant at 20mins, but the intensity was increased from a speed associated with a heart rate of 140bpm, to a speed which elicited a heart rate of 180bpm. Within 12 days, the experiment was stopped, because none of the horses could manage the training. BOTTOM LINE: When the easy days got harder, the horses failed to recover sufficiently to handle the training load.

In Part 2, I did a quick review of some of Carl Foster's work on Session RPE, and the concepts of Training LOAD, MONOTONY and STRAIN. Foster studied 25 endurance athletes for 6 months to 3 years, and found that athletes displayed individuals thresholds for LOAD, MONOTONY and STRAIN, which when exceeded were associated with increased incidence of illness. Training MONOTONY is a measure of the difference in training load from one day to the next.

So how do we put this info together to improve the way we program?

It's actually quite easy, but we like to make it complicated. As with most things related to physiology, I think it's important for coaches to have a firm understanding of the fundamentals and the relevant and emerging research, but I think that the implementation of these ideas has to be simple, context specific, and sustainable. For example, it's a good idea to have a handle on the time required to recover from various types of workouts, but triathlon schedules will frequently break these 'rules'. Similarly, there are lots of gadgets and software available to (arguably) measure training load more accurately than Session RPE (Training Peaks' WKO+, PhysFarm's Race Day, and TRIMPs are three examples). But in my opinion, software can't beat daily observation and interaction with athletes, and my time is better spent coaching than analyzing data and endlessly editing programs to make sure we don't break a dozen rules. In my context, I see my athletes at least once a day, 5-6 days a week, so I can make very informed decisions about their training based on my observations. In order to keep things running smooth, I just need a simple set of guidelines when I'm writing or editing the schedule.

Which brings us back to the studies from Part 1 and Part 2.

You've probably realized by now that the horses flourished for 260 days as their weekly training load increased, and monotony DECREASED week after week (if the load on the hard days increases, and the easy days stay the same, the monotony decreases). When the training load on the easy days increased, the training monotony also increased, and the horses quickly failed.

Consider the two sample graphs of weekly training load below. Both weeks have a value of 5200 units.

Graph 1: Training Load = 5200 units, Monotony = 1.6, Strain = 8507

Graph 2: Training Load = 5200 units, Monotony = 3.9, Strain = 20305

Both weeks have the same training load (5200 units), but Graph 1 has much harder and easier days - or less monotony and less strain. In Graph 1, the athlete has days with almost double the average daily training load, but there are also days of reduced training, or no training at all. By contrast, Graph 2 has the same weekly training load as Graph 1, but it is more evenly distributed across the week - no hard days, no easy days.

The MONOTONY and STRAIN are higher in Graph 2. Based on Foster's (1998) findings, there is a higher chance of illness with the program outlined in Graph 2, even though the weekly training load is identical.

You can plug all of the numbers into Excel, or you can just tell by looking at the graphs that Graph 1 has less monotony than graph 2.

Putting this into practice.....

In my opinion, all of this info about horses and speedskaters distills down to two basic principles:

1) Do the highest training load that is sustainable and repeatable, to obtain the most adaptation.

2) Deliver the training load in a manner that best minimizes monotony.

In practical terms, we simply do what smart runners have been doing for years - we go hard on the hard days, and easy on the easy days. That means our threshold swim on Tuesday morning is followed up with a threshold run on Tuesday night. Our easy swim on Thursday is followed up with an easy run. It also means that one of my fundamental rules for athletes is: if you missed a workout, don't try to make it up, just get back on the program asap. I've seen many athletes miss a workout and try to make it up the next day - usually going hard on what was supposed to be an easy day.

For a few months I planned and tracked daily training load for our program to make sure that we were actually doing what I thought we were doing. It turns out we were very close, so I stopped tracking the load because I was satisfied we were on track. Now I just program with those two principles in mind: keep the weekly load high, keep the monotony low. Any casual observer can see the difference between a week with low monotony and one with high monotony.

As athletes get fitter and we increase weekly training load, we increase the volume or intensity on the hard (and moderate) days first. As athletes develop over the years, we begin to string together 2, 3 or 4 big days in a row. This increases the monotony, but we do it gently, with long term progress in mind, and we monitor the athlete response. At the international level, elite athletes rarely take a day off, but remember that they have accumulated enough experience and fitness over their career that an easy 3k swim and 60min spin makes for a very easy day. Compared to the work they do on a hard day, there is a large enough difference between the two that monotony stays relatively low.

So it took me three long posts to get to the point: Hard on the hard days, easy on the easy days. Hopefully the journey was worth the effort for the coaches out there.

Train smart. Have fun.

No comments: