The best pacing strategy for an ITT on a steady 4 hour climb?

Hello, as some of you guys have heard, the climb of Letras in Colombia is one of the longest in the world. It is 80km at 4% average, you go from 500 meters above sea level to 3700m. The climb has some flat/downhill sections (that’s why is 4% gradient instead of 6-7%). Here is the Strava segment: Alto de Letras completo desde Mariquita | Strava Ride Segment in Mariquita, Tolima, Colombia

I have two conflicting ideas about the fastest way to go up the climb.

  1. Try to maintain an average power from bottom to top. Let’s say that I can maintain 260w average for 4 hours, then it would just be to maintain that power the whole climb (assuming there were no downhill sections to make it easier…)
  2. Go higher power at the bottom and decrease it steadily as you go up… Because the bottom of the climb is at 500m, there is a lot more oxygen than at 3000m, and we know that the power you can maintain drops substantially the higher you go (even if the person is acclimatized). I live and train at around 1800m, just in case you need that assumption. I was thinking of starting at 90% of my FTP (at 1800m) and decreasing it steadily to about 80% on the final kilometers.

Is there any research on this, or is this just down to “physics logic”? Which is it?

The guys touched this topic on podcast #124, but did not have the answer to the question. They just wondered about which would be the best and talked about their experiences going up Mouna Kea.

Thank you

1 Like

Have you checked what Best Bikesplit says would be the optimum way?

For that much altitude change you will definitely need to adjust your threshold power (and power levels) accordingly. For long hillclimbs I will break up the ride into elevation bins and write what my effective threshold is at those altitudes on tape and stick it on my stem. Both Bassett et al and Peronnett et al have described the power change relative to elevation. My experience is that they’ve been pretty accurate, at least as far as setting a ceiling for my efforts.

1 Like

Hi Daniel, That looks like a great event! I’d love to do it!

The “fastest” way to ride it is one that optimizes around workload (i.e. kilojoules). Best Bike Split is the best tool for that. And with regard with altitude, once you have a BBS model for the race, you would adjust at different elevation points. After reviewing the elevation profile, you would adjust suggested Normalized Power about 5% downward at the 1/3 point and another 5% at the 2/3 point.

I created the route in Strava, exported the .gpx file, and imported the course into BBS. Using my own #s (3.6WpKg), and a conservative ave .65 IF (approx 0.70 lower, .65 middle, .60 upper), and other data BBS requires, I’d do the the course in just under 6 hours (excluding rest stops) . . . perhaps a bit faster after a deeper analysis.

If you want, I could create a BBS “race” for you on this course. I’d need some #s and info. Let me know if you’d like me to do that for you. You can either provide it here and/or private message me.



Yes, but it looks like it doesn’t care about elevation, just gradient. The power in the first climb is the same as the last one, I think it doesn’t take into account elevation (even when I activate it in the advanced options)

I actually just put it into BBS to see the results, but was kind of disappointed it doesn’t take altitude into account, just the usual gradient/wind etc, so the power in the first climb is almost the same as in the last one.

Here’s the link to the race I created: Best Bike Split

I actually already did this climb, will do it again (probably next year though), but was discussing the best way to do it with some friends and coach… that’s why I wanted some input from the internet and other people. It was actually a similar time than the one predicted by BBS, here it is in case you want to check it out:

Thank you

Very impressive results! Holding an IF of 0.81 for 4 hours takes a lot of stamina. The fatigue resistance component of your training really paid off.

Yes, I agree that it is disappointing that BBS does not include an altitude adjustment. However, it is likely because it varies widely person to person (although they could allow the user to input it - you should consider sending a note to their support people).

There’s a lot written about the effects of altitude. I’ve done Haleakala on the island of Maui in Hawaii several times (sea level to 10kft over 36miles) It’s approximately the same gradient as the climb you did and hence where my power adjustment thoughts came from taking into consideration where you live/train at and the elevation of the climb.

Another suggestion: As I mentioned, if I were to do it, I’d do a deeper dive analysis that might include a BBS of 3 segments of the course, adjusting the IF of each to account for altitude. So if you are able to maintain an average of 0.81 for the entire course, you may consider looking at something like 0.85, 0.81, 0.77. Musculature-wise, they’d all be the same relative to the concentration of oxygen (e.g. NP Watts/O2).