What beautiful curves! (2/2)

Last week I started looking at power output over time and what it means for our Ironman quest. I mentioned Alan's post on fatigue curves over at Endurance Corner, which gave me a great perspective of what's important for Ironman racing.

The 5 and 20 minute tests I looked at last time give you a good idea of your aerobic and anaerobic capacity. But does that help you predict performance over a race that lasts about thirty times as long as the test? It turns out that it's only of limited value in doing so. Alan shows some data on the fatigue curve over longer durations: by how much your power output drops when the duration of the race doubles. Elite Ironman athletes have extremely "flat" fatigue curves - their output drops as little as 5% as duration doubles. Very good age group athletes are in the region of 7%.


But what difference does that make to power output over time? If I take my best power output over 5  minutes and start looking at the predicted power over longer durations, we see that the fatigue curve starts making a huge difference. An athlete with a 5% fatigue curve should be able to hold 70% of his 5 minute power over the duration of an Ironman, whilst a 9% fatigue curve athlete would only be able to hold 52%! For Alain the difference would be between 270w and 200w, and for me between 346w and 257w. A caveat here is that for big athletes like me, fueling high power outputs becomes really difficult.

A 5% curve is definitely out of the question, but let's look how Alain and I fare against our short tests: Alain has a 5 minute best of 5.36w/kg and a 20minute best of 4.59w/kg. Over longer durations, he has produced well: he has put out 3.79w/kg over 2.5 hours, and 3.31w/kg over 5 hours - only 2 and 5% less power than predicted by a 7% fatigue curve.

I, on the other hand, have slightly higher 5 and 20 minute power numbers at 5.82 and 5 w/kg for 5 and 20 minutes. My longer numbers, though, are further from the 7% fatigue curve: I put out 3.79w/kg and 3.33w/kg over 2.5 and 5 hours, 7 and 9% less than predicted by the curve. A caveat: the long numbers might be a bit low because neither Alain nor I did specific all out efforts over 2.5 or 5 hours in the last year.

Looking at our training history, the data makes sense. In the last two years, when I worked with Marc and first concentrated on half Ironman, he had me do a lot of VO2Max and threshold work, which have boosted the short numbers. It also means that I have to do a lot more endurance work to bring my long numbers in line with my short numbers, since this is what ultimately makes the difference in Ironman racing.

To illustrate this, here are some numbers from Paul Linck, who is also trained by Alan. He races extremely successfully in the 40 to 45 age group, and just recently came second in Ironman Florida in his age group, despite a running injury. Paul has lower short numbers (4.65w/kg for 5 minutes), but he puts out amazing long numbers, in line with a 5% curve. His best demonstrated 5 hour effort of 3.40w/kg - higher than what Alain or I produced. No wonders he biked the course in 4h48 in Florida!

Now I'll try to figure out how best to increase those long numbers. Putting out 347w for 2.5 hours and 312w for 5 hours is extremely scary - but that's what it takes to bring the long numbers in line with the short ones.