Power data analysis – Glocknerkönig

Note: My race report in German (with nice pictures instead of numbers ;) ) can be found here!

As a reminder, here is the beast in all its glory:


The race finishes at the Fuscher Törl after 27.7km at an altitude of 2428m.

Background info

I use a power2max (classic) powermeter. Unfortunately I do not know my exact body weight on race day; my lowest the week before was 52.6kg and the day after I weighed 53.1 (after eating quite liberally post-race ;) ). Let us assume a body weight of 53kg on race day. My bike weighs 6.7kg.

The start

The first 10km or so are a false flat, which we completed quite fast (~39km/h); I managed not to waste too much energy there (134w average, obviously with some spikes but my old GC software won’t show NP for intervals, sorry…).

Then the road ramps up quite suddenly; here is where the first selection is made and it is very tempting to go with the pack, but well, not everyone is in for a 1:20h time… ;) . I did the first two minutes of the climb at 260w, corresponding to about 274w (5.17w/kg) at sea level – see discussion below. This is NOT my one hour power hahaha…

I backed off quite quickly as you can see in the power graph below. Before the toll station Ferleiten there are some short flat parts where I recovered in the pack. After Ferleiten the climb is quite steady, 13km/1278m at about 10%, it only flattens off a bit in the corners. I took about 63min to complete this part; therefore the data lends itself to making an estimate for my maximal one hour power (FTP).


Power graph in Golden Cheetah (GC): purple = watts, red = heart rate, green = speed, light blue = cadence, dark red = power balance left/right

Last hour

I could simply tell you that my average power was 209w (3.94w/kg). This is very little; the deceiving thing about this climb is the altitude, which – at least if you are not used to it – results in a severe drop of aerobic power. I will follow Jürgen Pansy’s approach to take this fact into account: We divide the section from Ferleiten till the finish line into four segments of equal distance (3.25km) and use the percentages of available aerobic power cited in Friel’s blog for the average altitude of each segment to calculate a height-adjusted average power for each segment:

Now the number there (234.74w) is quite a different figure than the initial 206w!! (Obviously it is only an estimate; we do not know how much I really lose due to altitude – at any rate I am not acclimatized at all since I live at almost sea level!)

At a body weight of 53kg this means an estimated height-adjusted 4.43w/kg for the last hour of the race. A very satisfying number for me, considering that (a) I hadn’t really spent the 30min before that picking my nose and (b) I didn’t ride with the aim of achieving a high average power but a good time and therefore backed off from time to time due to tactical reasons (drafting).


Power profile according to Allen/Coggan. There’s still room for improvement haha!

(Btw, it just occurred to me that last year in May I could barely hold 210w over 20min at sea level with 3kg more body weight!! What a change!!! :))) Beginner’s luck paired with hard work and a good trainer!! Thanks Carles! ;) )

Final sprint

At the end of the race the influence of altitude is extremely noticeable (at least for me). I was mainly focusing on breathing; my brain and legs were rivalling for oxygen and any slight increase in power made me feel afraid of losing my balance. The final sprint was absolutely all-out, but I only did 251w during 32s. Given the altitude of 2428m (7965ft) the loss of power according to the table above is over 15%!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s