The Bell curve of cylists - how fast are the average TR users?


Yep, I’d say you’ll be able to roll in under 10. But TR probably has hundreds of potential data points on this, and I just have a few!


Agreed with @DaveWh. 3.7 sea level should get you under 10 at Leadville well executed.

Nate was just under 4w/kg at 5000’ and had a week of acclimatization and went sub 9.

in 2017, I was just over 4w/kg at sea level, had a week of acclimatization and had almost exactly the same first half splits as Nate, but had to limp home during the second half due to a knee that went bad on me. My moving time was 9:45.


Take a look at best bike split. You can run a simplified race analysis on one course using the ‘try it free’ link and it will give you an idea as to how you should optimally pace your ride also.


Ha! I ran it with a 180# rider weight, 275 FTP at a 300 foot training elevation, on a 22.5 lb bike, and the response was 8:32! I must have input something a little off on that one.


Maybe if you could sustain that power throughout, had a clear track and didn’t stop for water/food etc…

It’s more optimised for road riding and more accurate specifically for TTs


I don’t think BBS has a good model for leadville. Another way to say what willball12 said; they don’t have suspension/dirt/MTB tires dialed.


Agreed with Nate. I tried to get BBS close to known good numbers at Leadville but it just didn’t work out.


I just tested 1 week apart, once using Virtual Power based on a Kurt Road Machine with Wahoo Speed sensor and the second time on a Wahoo Kickr. Both Ramp Tests were conducted on about 48 hours rest.

Virtual Power yielded an FTP of 289. Consistent with past performance on this protocol.

The Kickr placed me at 251. Intuitively this fits better with my real life experience riding with people who train with power year round.

All to say protocol and equipment can yield substantially different FTP for the same rider. We all sort of know this but all the same I found the experiment educational for me personally.


@Nate — is there data showing workout compliance rates per FTP level?

I.e. — are some workouts non-compliant on the left of the curve but RH compliant?



I haven’t looked at that yet but that’s on my list. I want to look at watt/kg, raw FTP and 5 year age buckets.


Awesome! I think you guys (TR) are on the verge of developing a really powerful training tool…if done right.
I know there’s bit a bit of hubbub about the change in training plans, it would go a long way in your…popularity(?) if you could offer customers TR prescribed metric-matched plans/workouts rather than just doing a wholesale alteration to accommodate the masses. Totally understand there’s a business model behind all this and you aren’t in business to lose money…maybe a ‘TR Premium’ platform down the road?


Allen Lim Rice Cakes and home made energy drinks in all my over 60 min workouts have made such a big difference in my performance and ability to smash the rest of the weeks workouts. And with all that still loosing body fat so I am trying to see if I can eat even more on the bike :slight_smile:


Awesome thread… I was a bit demotivated after my last ramp test (beginning SSB high volume II) was virtually the same than the previous one and approximately the same numbers than a year ago but seeing that my 312W-4,5W/kg is in the upper part of the bell curve (at 40 and training late at night after full time job) has given me some extra encouragement!

Comparing not only FTP but complete power curves with other TR user groups would be great


You’d have to do this based on averages and by aggregating you would lose a lot of the fine detail. I’d be willing to bet that the curves will be almost identical.

Power curves can also be misleading. If a rider only logs their indoor workouts and sticks to the plans they are unlikely to register a maximal effort. TT/Tri based riders are also unlikely to get sprint power into their curve.


I see your point, but still I think it would be useful to be able to compare my 5s, 20s, 1m, 5m, 20m, 1h… power with others.

To be really useful, I think TR would need to guide you through a series of tests aimed to measure your best power for different time durations and ‘build’ and accurate picture of your power profile (not only the estimated 1h power or FTP like now).

Then, if you could select different groups to compare with that would be killer (age group, triathletes, road, mtb, pros vs amateur competing vs enthusiasts, racing category, sprinters vs tt vs climbers, etc.)

And finally… a training plan recommendation based on the “section” of your power profile that you want to focus.


This already happens when you choose a training plan based on your discipline.


I had a KK on virtual power and my tests were very consistent and after I bought a Kickr my ftp dropped by 30 watts. IIRC 250 to 220.


Agree. This would be great. Something like Sufferfest’s 4DP is a more complete profile for a rider.

I also think TR should add a test to determine your aerobic capabilities. I don’t know the best way for this, but one way to do it is a steady ride at, say, 65% of FTP and look for Power/HR decoupling.

The more TR can help you understand your profile as a rider, the more they can help tailor training to your needs.


You don’t necessarily need to test. Starting in March our weekly Wed night race produces maximal efforts for the 5 sec to 20 minute timeframes. Because I do a lot of riding on flat courses, my preference is Xertonline rankings which show how you stack up in terms of both absolute power (flat courses) and also W/kg climbing courses that favor GC guys.

Seeing my 5 min power (“breakaway specialist”) is 77th percentile for my age group is encouraging enough that I plan to enter some crits this year. And because of my size/weight, you can see that W/kg rankings like GC and Climber (are below 40th percentile which nicely aligns with where I end up on Strava climbing segments.


I certainly wouldn’t be demotivated if I were you, that’s a strong power and FTP any way you look at it. Be proud of what you’ve achieved :blush: