Statistical Analysis Reveals . . . . Overtraining!

It’s at about this point that I lose most of my readers. But here goes anyway . . .

Toward the end of last week, I was feeling pretty low. Basically, I was just permanently tired, and really easily fatigued in training - not a good sign. Well, for the last few weeks, I’ve been taking daily measures of my weight and heart rate, with the intention of applying an engineering technique called Statistical Process Control (SPC) to analyse the data. I know - pretty nerdy, but there you go.

Well anyway, tonight I decided that I probably had enough data to be able to make meaningful inferences, so I went ahead and used the handy e-zine article from SPC for MS Excel to help me do the cunning math to convert my columns of numbers into something meaningful - a graphical picture:

Here we see the chart for my Weight. It’s an X-Bar, R chart, but I’ve only shown the X-Bar part here.

The two red lines are the allowed variation (upper and lower control limits) - anything outside of these indicates something other than just natural fluctuations. The black line is my ‘target’ - based on the history represented in all the data, I should see variation either side of this. The blue line is derived from my actual data - it represents a rolling average of three days’ readings. i.e. each point represents the average of that day, and the previous two days’ weights.

There are certain things to look out for on these control charts. Amongst others:

• Any ONE point outside of the control limits. Oh look, I have out of control weight readings. Actually, on this one, I’m not too worried - it’s not a straight line plumet, and I did expect to be losing a little weight - my target is 73kg. There’s also the fact that my ‘target’ and control limits were set based on only ten readings, and I should be recalculating these every twenty readings.
• Grouping of seven points all either above, or all below the ‘target’. Yep - got that too, but as I said, I am expecting this kind of picture on this graph.
• Seven points in a row - all either ascending or descending in value. Well at least I’m in the clear on that one.

So despite my reassurances about expecting to see some weight loss, it looks like things might not be totally normal. What about my heart rate? This graph is constructed in the same way:

What does it show?

• Every reading’s between the control limits. Check!
• Grouping either side of the ‘target’. Not so good - the first 11 points are above the line . . . and the the last seven are below.
• Seven in a row? Oh dear! Look at what happens from 25th Jan to the 2nd of Feb! There are TEN points in a row

Now, while I would expect to see my heart rate coming down as a result of training . . . I would only expect to see a movement in the long-term average - not a consistent downward shift in short term readings like we see here. In other words if we drew a long term trend line over several weeks or months, then the short term averages should fluctuate around this - not follow it exactly.

SO what do I need to do?

1. Listen to my body - in the second half of last week, I had clear indications that I was getting myself into a bit of bother.
2. Keep recording the data and plotting it! I could be mistaken in how I feel . . . but the data’s more than just a little compelling.
3. Joe Friel’s Triathlete’s Training Bible says that the only option is REST - 48 hours of rest
   • Now, I took Saturday off . . . well kind of. It was supposed to be the last day of the Life Centre’s open-air ice rink, so I had an hour’s ice skating with my daughter - she can’t yet skate, so I spent an hour holding up 24kg around a small, crowded ice rink.
   • Um . . . and since our friend Von SmallHaussen was up for the weekend, I did a spin session with her at the gym on Sunday. (That’s her on the LEFT of the picture, BTW. You just don’t want to know about the ‘Lady‘ on the right.)
   • And today . . . I went for a ride - 18 miles on my sturdy, non-aero bike in a shade over an hour.
   • And tomorrow . . . well, I’m booked for the 06:45 spin class . . . perhaps I should cancel?
4. Thinking beyond tomorrow though, I DO need to re-think my training a little. Overtraining stems from workouts being too long (excess duration); exertion too high (excess intensity), or too many workouts (excess frequency). I think I’ve been guilty of the last two - cycling very frequently, and to farily high strain level. Some thinking to do on this though - I’ve that 3,000 mile target for this year, and I just can’t let THAT go - imagine the ribbing that ZappoMan would give me . . . I’ve been thinking of converting spin-time into bike miles to reduce the pressure on my road time - something conservative, like 10 miles for a 45 minute class? What do you think?

More on all this later - the comments about WHEN to eat that I got from Friday’s sob-story post need more reading and thinking about - I’m fairly sure I messed this up too!

Filed under: Fitness, Triathlon