Agreed. That’s partly why I was surprised to see a value flagged like that.
At least some of this probably comes from the longer than usual break I took in Oct/Nov, which lead to big drops in actual and eFTP. So this trigger here may be a result of no other significant efforts in that timespan.
For reference, here is what my fitness graph looks like over the last 3 months, if it’s useful:
Considering how the ramp felt and how my training was going leading into this test, I actually think the 240w is quite close for me vs other tests where I knew something was “off”. I will get a better feel with workouts this week, and evaluate just like I usually do.
Ditto, I had some inflated estimates when it was at the 3m default. But that is at least partly due to my apparently good short power. So might not be a problem for everyone.
Again - I am sure someone from TR will give more info at some point but from Nate’s description they won’t be directly extrapolating FTP from specific submaximal efforts (i.e., you did X watts for Y duration therefore your FTP is Z).
From Nate’s description this is a predictive method based on looking at your complete training and FTP history (and that of all the other TR users) to come up with an ML-based prediction of what your FTP will be at a given point. Remains to be seen how accurate that would be but as this thread shows you can argue about what an “accurate” FTP figure is till the cows come home; from a TR perspective the estimate just needs to be good enough to serve you appropriate workouts under the AT framework.
They’ve also mentioned a feature where you get a prediction of what your FTP could be after completing training blocks (which amounts to the same thing except forward-looking in the calendar).
If anyone can do it, TR has the data to support it. But I guess we’ll see how good it is
Interesting My SS PL is 5.3 so if it’s right my estimated FTP is near bang on. My 20 mins FTP is higher but my ramp is lower, so I have went with what I can consistently deliver and tweaked it to make training zones feel right
counterpoint: my SS level is 8.9 (and prob 9.2 after today) and I still wouldn’t say my actual FTP is higher, I really don’t think equating levels to FTP really works
Completely agree - there is not enough specificity around FTP, lacking an agreed time frame creates all sorts of challenges. I also think someone earlier touched on something quite important. The focus on testing and FTP in cycling is actually holding some people back from potential performance gains. In other endurance sports the primary focus is progression - ultimately by training through a progression you will be getting stronger and faster. What does testing really add to that over and above what TR have implemented in progression levels?
Well, easy to test. If your SS PL is 9.2, increase your FTP by 10% in Trainerroad. This should drop your SS PL to ~5.5. See if you can handle the training at the new FTP and level.
I’m aware that the levels drop, I raised my ftp manually 5w (2 percent) in ssb1 and my levels dropped by 0.6. But I’m just saying that achieving high progression level shouldn’t automatically mean to increase ftp. Because if I went up 10 percent I’d basically be doing vo2 stuff at supposedly sweet spot workouts
Yes, that is the interpretation of FTP and PL’s from Trainerroad. And no, see no reason to do ramp-tests for the purpose of setting the correct training levels. Three to five trainings a week with consistent feedback are much more reliable than one ramp-test every 4-6 weeks.
Possibly but not likely. For the lower progression level SS workouts intervals are done at <90% of FTP. For the highest levels this goes up to 95%. So when you increase FTP and effectively lower PL, the workout intensity relative to FTP decreases and you will likely stay in the correct range. And keep in mind that for SS workouts the total kJ roughly stays the same before and after FTP change.
Regardless, you’re wrong in this approach, level 9 or 10 sweet spot should not automatically mean one should increase ftp and speaks to the dangers of using the wrong ftp setting, even with progression levels. If I couldn’t get past level 7 or 8 sweet spot, it would be because of a grossly overestimated ftp, folks should be able to do a 7.9 sweet spot (3x30 @ 88% as an example). If someone is plateauing sooner I would say it’s a target problem
I totally agree and I would even say a 3x30 should not be barely possible but right in the middle of „hard“ or below
That being said, if you can do a 3x30 at 88% you can express your ftp pretty well and you can decide to
a) progress further in your tempo/ss abilities
b) bump ftp and start again with 2x15 ss
what is the better way depends on the athlete I guess
I’d restate that as, if you can knock out 3x30 @ 88% then you can:
keep focusing on extensive aerobic and go for something like 4x30 to increase fatigue resistance
switch to intensive aerobic focus and push up power under lower durations (say 20-30 minutes, depends on your power curve)
retest if you think FTP has increased
Well first off, a lot of people need to test because they just want help training without having to understand how it all works.
Secondly, like many other companies the TrainerRoad plans have always used linear periodization to progressively develop fitness. That was true before PLs, and true after PLs.
The purpose of estimating FTP is to establish power training zones, and update them as your fitness changes. There are many ways to estimate FTP, and TrainerRoad is hoping to add their own estimation method that doesn’t require a ramp test, or more traditional 8-min or 20-min tests.
Good thoughts. A little bit similar to me.
Based on some races I did last year, my FTP in Garmin Connect is estimated way higher than my Ramp test Result. That’s fine, but I would fail nearly all workouts on TR using this number.
So I think maybe it is possible to skip maybe some Ramp Tests, But 2 or 3 Times a Year, it will be nessesary.
That is correct. Intervals.icu doesn’t currently interpret a ramp test (using the last minute etc.). It just so happens that using the average power for the last 6 minutes or so to pick an FTP often produces very similar results.
The default min duration for eFTP has been 300s for a while now. The original 180s was definitely too short!
The progression levels for all the zones are linear. I am not using Threshold much so cannot test this, but for VO2 max an increase of +1 PL can be translated to +2.1% FTP (instead of 2.8% for SS). So you can use the same approach for all the zones, the gradient of the curve is just slightly different.
for anyone who listened to the latest podcast, there was a ton of discussion of this feature and other related features, with remarks that it is “very close”
I’m definitely interested in seeing this as it relates to my own approach to FTP, which has been incremental 5w adjustments if things feel good and start to feel a bit too easy on some workouts. I think if the TR team really wants to make a big splash, at some point you may want to consider having some validation with lab tests compared to the machine learning. I know there are a lot of people who are skeptical, might help to win over some folks if there’s some independent data to validate.
My SS PL is 5.3 so if it’s right my estimated FTP is near bang on. My 20 mins FTP is higher but my ramp is lower, so I have went with what I can consistently deliver and tweaked it to make training zones feel right 