TR could guess FTP like Intervals.icu: From a max effort of any duration more than 3 minutes. Then people who are not having a good experience with the ramp test number could choose to do a 6 min max effort (or 10m or 20m etc.) instead. Its very easy to implement for TR (they have all the power curve data needed).
Estimated FTP is calculated by placing you on a power curve using a single maximal effort of between 180 seconds and 30 minutes. If your anaerobic (short duration) power is very good you might want to use longer durations to avoid over-estimating your FTP.
for example, my FTP isn’t very high, but for workouts with for example 10-second sprint between 180-200% FTP, my FTP is to low, i think, i can handle a 10 second sprint with 280-300% or, but i am not sure with 380-400% FTP, so my Power Profile in the background could handle this automatic, or?
Also if you go out too hard and blow up before the end it will still give you a good number. If you go too easy and have gas in the tank, just keep gong as long as you can. Any extra seconds will give you a higher FTP so long as you can maintain the power.
With enough data for a rider, this can be reduced to a statistical inference problem, but it likely requires a range of maximal duration tests to make sure the power curve (and its knobs) is accurate.
I think this is still one of the reasons why I almost never use the FTP recommendation I get after a test and instead use the modelled FTP provided by TrainingPeaks, based on the overall shape of the curve and so on. Basically, I use the test as an additional input to WKO5 power curve. Their provided FTP even give me how long they believe I can maintain it (Time to Exhaustion), anaerobic contribution to my FTP (function reserve capacity)
As a side note, I am quite impressed how close interval.icu comes to what TrainingPeaks WKO5 provides ;-).
1 - As a reference figure from which, though a percentual number, we target a power to pace intervals and efforts in general.
2 - To work as an endurance very specific cycling fitness marker we can follow to gauge our evolution.
For me FTP is too much approximative in solving point 1.
What I found in TR workouts is that my ramp test results seem to have overestimated largely my FTP when I need to put out efforts between 0,85 and 105%, they are spot on between 106 and 120% and it underestimates anything above that. Strangely enough they seem ok in racing which is typical done below 0.85.
So this "problem " cannot be solved by multiplying the P60s of the ramp test by 0.72 instead of 0.75 or something like that.
My power profile needs a totally different reference matrix to address the appropriate energy systems while training and direct me to the adequate power targets while racing.
So for me, this is not only about the ramp test capacity to estimate maximum lactate steady state without being a pain.
It is mainly about the flaws of creating a full training philosophy and a product around a single number that in itself is being used much beyond its underlying concept.
In that sense, I have to say that Sufferfest seems quite ahead of the curve here. Maybe their test could be improved but the concept seems much more sound and inclusive.
What a great topic, happy to see a serious discussion.
Have been thinking about the ramp-test for quite a while, and one wonders if we still don’t have anything better after >25 years? Francesco Conconi used the same test in the 80s to help Moser determine (and improve) his anaerobic threshold for the world hour record attempt (oh yeah, and some blood doping as well, different topic). Conconi used the deflection point of heart rate vs power to determine threshold, and the .75 rule from TR is basically a simplification of the original Conconi test.
TR is likely hesitant to use HR as it is affected by hydration, fatigue, temperature, altitude, etc. But as we mere amateurs are unlikely to measure oxygen intake or blood lactate, it still might be one of the only biomarkers that is available.
Based on the many comments on the forum the ramp-test is a serious weak point in the TR approach. Using a combination of the power curve (like Sufferfest), HR and workout compliance as inputs to set training intensity seems to be the way to go. But certainly not trivial, as transparency would be hard to provide and TR might also have to build ‘all out’ efforts into their plans. And that would likely lead to many complaints as well, as many users seem to enjoy the ‘follow the power-line’ approach.
I think they could drop the frequent ramp tests and look at performance in specific ‘test-like’ workouts that are in the plans. We all know that Lamarck is a good check of your ftp for example, I’m sure there could also be a more vo2max type workout. Those could be used throughout the plan to adjust ftp and power at vo2max. The problem is that it would be exceed your current ftp - you would have to deliberatly increase intensity in a testing workout, and that might be difficult for some users to realise.
I’m with @AndrewL here as I’m very aerobically inclined, always have been - even at school I’d struggle with sprints. As a result I find that I underperform on the Ramp Test, typically by 2-3%. When I last did Lamarck I’d already bumped my FTP from the tested 242W up to 250W and my average through Lamarck (which was my twelfth day in a row) was 252W. My last Ramp test was a 'mare, I quit at 17mins as neither my head nor my breathing was in it. I kept the previous FTP value as I recognised that particular test as being a bad day.
I struggle with VO2max workouts such as Bashful+1 but I think that’s mainly down to the very short recovery valleys as I did Pisgah on Monday which is 2mins at 125%, so anaerobic, but with 4mins recovery and that was fine.
Weighted historical data/compliance might be useful in adjusting the result of the Ramp Test, or indeed any other, ignore anything older than a month or 42 days or whatever. Something like that might even out “bad” days.
@davidtinker - intervals.icu gives me an eFTP of 256W which I think is low, the Ramp Test gives 266W but then I can’t remember the last time I did a maximal effort of at least 180 seconds, I don’t think any of the TR workouts I’ve done recently have had such an interval, the last one was Sierra+1 ten weeks ago and my notes for that say I found the 3 minute interval straightforward suggesting I’d more in the tank.
From the podcasts TR did quite a bit of beta testing on the Ramp Test and they are unlikely to have begun using it unless it showed at least as good results (compliance in subsequent workouts, etc.) as the 8min and 20min test protocols. Just how much of the bell curve is being missed here? Are we talking about going from 80% to 95% accuracy, 90%-95% or maybe just 95%-97%? If the latter then is it worth it? Those of us in those last percentiles are likely to know where such a simplistic modelling fails and adjust accordingly.
Yeah you have to do a max effort every now and then to get an eFTP number out of Intervals.icu.
I think the ramp test is quite a reasonable choice for TR. People who know what they are doing and who prefer other tests can do those instead easily enough.
And to those of us who have no clue, the ramp test is a good approximation one way or the other. Get you started without unduly taxing the rider, who may not be in optimal shape in the first place, and keep you in roughly a good place while you learn enough (about cycling and about yourself) to make personalized adjustments.
By now, I’m getting to the point where I have enough endurance to complete workouts (for this whole first year, I was failing them pretty regularly). And I’ve learned enough to begin to understand my own profile, maybe make some of my own adjustments. But until now, something simple, prescriptive, and not too demanding, like the ramp test, has been just what the doctor ordered.
I see a possible problem with your example. To many changing variables. Next time try doing a TR ramp test in the morning, then follow that up with a TR ramp test in the afternoon and see if you get different results. With the example you used there is no way of telling if it was the platform or the time of day that is causing the difference. To be honest it seems a better way to evaluate your FTP is to take the ramp test at the same time of day that you normally do your training. This would give you the best chance of having your body and mind in the same place every day to duplicate how you read your RPE.
Chad you may have already covered this but the drawback I see is this…
We also know that rider profiles differ and therefore some people cope well with, say VO2 max workouts, whilst others struggle. I suspect that this would continue to be the case regardless of how an FTP value is obtained.
If this is indeed the case, then changing the way that FTP is determined doesn’t really solve many issues, relative to the amount of time TR would presumably have to invest in implementing your ideas.
Maybe the time and resources are better spent trying to personalise each persons training sessions better. In other words, determine the riders 5sec, 1min, 5 min and 20min powers, and apply this to each workout in such a way that each individual rider gets the optimum amount of training stress at zone 2, sweet spot, threshold AND VO2 max rather the rider having to adjust intensity all the time because they know they’re weak at VO2, for example.
I know Sufferfest’s 4DP covers this in their test but I don’t know anything about if/how this data is then implemented in to an individuals training plan.
This approach may solve both the “is the ramp test right for me?” and “i can’t complete any VO2 workouts” problems.
I’ve thought about this and kinda liked it at first thought. I even thought that the workout text could simply be changed to encourage anyone planning on using that workout as an ftp test/indicator to increase the intensity. This would therefore give them a new FTP, assuming they were successful with the increased intensity.
The trouble is, it could also encourage people to increase the intensity and end in a failed workout and no new FTP… I’m not sure whether this approach would work for the masses or not.
Yeah, I think this is super complex and offer only a minimal suggestion. I am far from informed or knowledgeable enough to make any more specific recommendations. I am a “problem solver” by nature, and try to look for common and repeat issues to “fix”. This one comes up often enough that I see it as being worthy of review. I just don’t happen to know enough to offer any “real” solution specifically.
The “Ramp Test doesn’t work for me” threads and comments are numerous. FTP testing in general is still sort of a black box and an area that can likely be improved. My initial suggestion focused on the potential rider profile, but that data depends on heavily on how and when it was obtained.
There’s some great suggestions above related to alternate or supplemental ways to determine FTP for training. I have had some good luck with the eFTP from Intervals.ICU. That and other options mentioned above all seem like good ideas to consider.
Ultimately, I have no preference on how this is “solved”, only that it is improved in some way.
In my view it is more a challenge of educating the user. There isnt one magic number and whatever protocol you use people have to learn they have strengths and weaknessess. You may need to adjust your intensity for say VO2 workouts vs sweet spot. I know until I developed some endurance long intervals were more of a challenge especially if it had been a hectic day at work.
The workouts themselves still have text in them that indicate what you can do on the intensity and it has been mentioned numerous times on the podcast. There are still those that think its a failure if they drop the intensity.
