Fatigue exposed

In my Honu race report I described how tired I felt on the bike. How, even after a reasonable warm up period, I was unable to produce the watts necessary to follow my plan. At first it just felt weird. Then I began to understand that my fatigue was very real and the result of an hour long high power swim. Had I not been so aggressive in the water my race would have ended on the beach.

In the days that followed I did what any athlete would do in a similar situation, I began to question my preparation. Had I prepared adequately? Was I right to expect that I could produce the necessary power? In short, what went wrong? Fortunately Training Peaks makes available a ton of data, and I decided to look there for answers. I found them, too.

First, a quick review of what the data means. These are my definitions.

FTP - Functional Threshold Power. The maximum sustained power a cyclist can produce for an extended period of time, typically one hour. Equivalent to the power generated at Lactate Threshold Heart Rate (LTHR).

NP - Normalized Power. An average power, expressed in watts, for a period of time, adjusted to express the impact on the body. A ride on rolling hills at an average power of 120 watts will feel harder, and require more energy from the rider, than riding a flat course for the same duration at an average of 120 watts. In that case the flat ride will have an NP of 120 watts while the NP for the rolling hills ride will be higher.

IF - Intensity Factor. The average percentage of FTP produced for a period of time. IF = 1.0 means the ride was done at an NP equal to FTP. If the goal is to ride at an average of 85% FTP, IF will be 0.85.

VI - Variability Index. I am still not completely clear about this one. VI expresses how much the power varied from an ideal. Typically a time trial is done at a smooth, continuous power level. On a flat course VI should be close to 1.0, with anything between 0.97 and 1.03 being excellent. I am not clear how hills affect this number. A good strategy is to work harder on climbs and recover while descending. Seems to me that this will increase VI, unless the software also takes into account elevation changes.

The remaining columns in my analysis should not require explaining.

What I did was go back and get the most recent useful ride data and compare them with my Honu ride. To a certain extent this comparing apples to oranges, because none of the rides have the same profile. By focusing on a twenty minute continuos climb I was able to find something significant.

On May 8th I did a workout on Pineapple Hill -- Hwy. 99 from Haleiwa going up towards Wahiawa, which consisted of three ascents of twenty minutes each. On May 15 I did the Honolulu Triathlon, which has no hills, but I did ride at my planned Honu intensity, the goal being to sustain 85% FTP as smoothly as possible. On May 22 I did Pineapple Hill again, but this time only one ascent followed by a long flat ride with some all-out intervals.

The results of these rides and my Honu ride are given in the table below. (Click to see a larger version.) There are two sets of columns. First, the data for the entire ride. Then, for all except the Honolulu Triathlon, data for a twenty minute continuous climb. Note that I include all three ascents from the 15th.

There is a lot to look at, but one comparison in particular stands out, Honu and the second ascent on the 15th. On Pineapple Hill I was able to generate 172 watts for twenty minutes at an average heart rate of 135 BPM, while on the climb to Hawi I was making 115 watts at a heart rate of 137. What these data do not reveal is that in both cases I am very close to LTHR. Any attempt to go faster was unsustainable. There is a huge difference between 172 watts and 115 watts, and I could feel it. Even on the second day at Pineapple Hill I did 147 watts. On the first day I was going all out, On the second day I was trying to hold back as if it were a race. (The goal is to be able to run a half marathon after the bike.) For Honu my overall IF goal was 0.85. I planned to do the climb to Hawi a little above that; here I did 0.89 and followed the climb with a hard set of intervals and could still run off the bike. One more number. On the twenty minute climbs on Pineapple Hill I gained over 500 feet. For the twenty minute Honu segment -- not the entire climb but most of it, I only gained about 300 feet.

These data show that in the weeks leading up to Honu I had the strength to do well. The one thing I did not take into account was the effect of the long, intense swim. I think this confirms the conclusion I arrived at in my race report, that to do any better ay Honu I need to improve my swim efficiency. Of course there is plenty of room for improvement in my bike and run performance.