We all know that experience from using the FTP generated via the Ramp test can and does vary. The 0.75 reduction doesn’t necessarily “work” for everyone with respect to setting appropriate training zones.
There are countless threads related to this discussion, but I wonder if there is a way to move past the “0.75 for all” and add some automatic intelligence in potential adjustments (up or down).
Could using the current or recent power curve of the rider serve as a guide as to whether the the 0.75 is appropriate or a slight adjustment may serve certain power curve profiles better? This may parallel the methods of identifying the “rider type” as done in various test and what we see in The Sufferfest in particular.
I know that there can be pitfalls in judging from past efforts and how that may well be limiting. But we see a large number of posts related to how the Ramp “doesn’t work” for a number of riders. TR knows this and settled on a single value to apply to all riders.
Maybe they have the data in hand to review when that primary value doesn’t work for the riders outside the bell curve they reviewed, and could make a suggestion for a different value based on known past data? The whole analysis of that is well above my knowledge level with respect to the actual physiology and such. Just had the thought and figured it would be worth mentioning to see if smarter people have better suggestions.
This is well beyond my depth, but I wonder if TR can figure out the wattage/length of time where individuals start to fall apart on intervals and start using that to scale the targets of certain types of intervals (sort of like ilevels in WKO, I’d guess). I guess an easy first step would be for the software to auto adjust interval intensity if it detects a dropoff in interval power. It’s an interesting conversation and I think if we get into this and try to dream up all the ways for more dynamically adjusting targets we can really highlight how difficult it is (for all those people who think TR should have had this implemented yesterday lol)
Nate has mentioned that he doesn’t think live / dynamic changes within a workout are something he likes. There may be exceptions along the lines of matching a predefined rider profile (again, The SUF does this) to more closely match their physiology in things like VO2 max efforts.
I’m more interested in the initial setting on a more macro level for FTP, than a micro level of in-the-moment of a workout. I do think those internal workout efforts are important (and partially covered manually here), but I see these as two separate considerations.
I agree, definitely 2 different things. I think there are some overlapping points, as far as maybe analyzing the workouts that correspond to certain zones (do individuals make it through sweet spot without any issues but fall apart at 120% intervals past 2 mins?). I wonder if something like that could help them start to build a picture of what percentage of the ramp to work off as one’s FTP.
I expect TR will definitely have the data to do this. I actually thought they were doing something slightly more complex than x0.75 but perhaps not. At a high level I guess you’d want to group riders by their failure rate for certain threshold sessions we know to be good tests (e.g. 4x10min or 2x20min) and look for commonalities, possibly in their power curve but also general demographics.
Though as Chad mentioned judging riders on their past could be dangerous! I’d like to see something like this incorporated into plan builder. E.g. I recently hit a FTP plateau - I’d been doing a lot of steady state (sustained power build then century, prepping for a fondo that didn’t happen!) and had excellent muscular endurance but no top end. Now I’m training for time trials, but ignored plan builder’s recommendation and did short power build and that got me over the plateau. I reckon that would have been fairly clear from my power curve (perhaps + knowing perceived exertion) and the recommendation could have changed accordingly.
does TR offer an API that can be used to crowd source research? I love analyzing data and it would be great to try to answer questions such as the one you are posing Chad.
The main calculation is the best 1-minute power from the Ramp Test, multiplied by 0.75 (effectively 75%).
There is a “check” in the system that looks at the prior (or best?) 5-minute power to look for poor building, spikes or something. If they see a “problem”, it will de-rate the FTP in some way (but I’ve not seen the specifics about how or how much).
So, short of a “problem”, everyone gets the same 0.75 adjustment.
I think the issue you would run into with this dynamic flow at estimating FTP is that if you go to .7 - .8 depending on the rider you run into an issue of not being able to fully capture outside stressors which would be massive lurking variables.
For example this week I was supposed to do Washington -1 and failed. I don’t chalk this up to bad FTP number but rather other stress from the weekend.
They have talked about .75 not being perfect for all but it is a good start for most
I’m going to use myself as an example for a sec, I definitely do poorly on the ramp, in fact a couple of weeks ago I did a TR ramp in the morning, bailed at 17min (so nominally an FTP in the 250 range when I had finished SSB1 and 2 HV at 285), then in the afternoon I did another ramp test (on a different platform) and got 285.
I knew and felt I had more in me, I’ve been training at 290 for short power HV, 2 weeks in I’m nailing the workouts (knock on wood lol). Granted, I’m not doing the 2-3min efforts that really suck for me, but it’s an example of the ramp not matching the user completely (in my case, I’m probably not emptying the tank like I should, so that’s not the assessment’s fault).
That’s why I am suggesting a broader review with respect the power curve of the rider, specifically in the recent history. I am not talking about missed or problematic workouts from outside issues.
I am talking about a broader, systematic inability of some riders to perform workouts with the 0.75 derived FTP. Many people “overtest” and have to reduce their FTP based on repeat issues with multiple workouts following a Ramp test and the new FTP.
There is some level of introspection needed for any test. We see the same considerations for the 1x20m and 2x8m tests, where people may need to fudge the final FTP up or down a bit. They do it with respect to the test itself, but also from their prior knowledge of similar tests and the workouts that followed.
We need to do this even with the Ramp test. My suggestion stems from the apparent large number of people where the 0.75 value is not the best one. People have either ditched the Ramp, or determined a “fudge” factor that they use on the final assigned FTP.
My thought was to consider the rider power profile, as at least one factor that may point to a better prediction of an appropriate multiplier, when the 0.75 is not “right”.
My bet is that the avoidance of pacing is a bigger factor. Doing a proper test effort in the 1x20m or 2x8m tests is far from easy and something of a learned skill.
New riders have little to no idea how to do one properly, and end up with a “bad test” in many cases until they get a few under the belt. Those bad tests require guessing to interpret an FTP when they aren’t performed well.
Really cool idea Chad and I will happily pass this on! With the work we are doing into “smarter” testing and progression, this could certainly be something we look into .
I don’t mind the idea of a “Full Frontal” Test and would love to see if TR wizards could figure out good ranges from it.
The issues I see when it comes to best 5 minute and 1 minute power is that they are all relative to other work. I mean one could go all out get a 5 minute PR but then immediately get punted out the back.
I think the potential for messy and incomplete athlete power curves may make implementing such a system difficult.
I actually think The Sufferfest might be on the right track with their “Half Monty” test, which there was a big thread about recently. It gives you a similar ramp test to TR, then asks you to cycle for 20 minutes at a particular heart rate (which will be a % of the max you achieved on the ramp). It then uses the power you achieve at that heart rate to measure your steady-state ability and adjust your FTP accordingly.
I don’t know if they have the data or the science to back up the number it produces, but the theory seems fairly sound to me.
Edit: alternatively, go the other way and follow the ramp test with a max 1min effort to determine your anaerobic capacity and repeatability under fatigued conditions…
I would love to see someone like Lionel Sanders do the Half Monty SufferFest test. On the bike my HR rarely ever tops over 170. I understand that know test is perfect just would be interesting to see how it does.
Yeah for sure. There’s a lot of factors which can influence test results as much, if not more, than a rider’s power profile. Given that, I feel like it’s more appropriate for the individual to make their own adjustments as needed based on their own judgement and prior experience. Obviously that doesn’t help on the first few tests, but I’m yet to hear of a way of determining FTP that doesn’t become at least somewhat more accurate if the rider has more experience with the testing protocol.
I also think it’s really common for people to second-guess a ramp test result if they’re weak at a particular sort of workout, which more often than not is the one they haven’t trained as much because it’s not as enjoyable. I feel like these situations are often more indicative of an area that needs to be worked on than an inaccuracy in the FTP value, and that in itself can be helpful.
.