It seems that 4DP is anchored to their “Full Frontal” test and is also prone to error due to a bad assessment.
I already skip mid-season ramp tests because I can easily adjust ride intensity and/or user account FTP based on RPE. If I’m going to do that anyway, then what’s the benefit of 4DP?
I think people spend too much time trying to find out what “optimal” is…the reality is that there probably is no way to find out. Each person’s reponse to training stimuli is different, and moreover, you are never at the same point in conditioning again, so you can’t really compare how one program works vs. another.
Some riders respond well to volume (gello!!) and others to intensity…plus a million other variables. And just because a certain strategy worked one year doesn’t mean that it will be a good choice next year, etc.
research enough ti find a program that has generally provided good results for a large number of people…implement it and follow it. If you don’t see results, adjust. as needed. But don’t worry about finding “optimal” because you will never know if you have ever achieved it. You may achieve “better” but that still may not be optimal.
I think it’s more important to know how much and when you should be doing VO2 workouts vs worrying if they should be at 115% or 120%.
That said, one thing I think TR needs to do better is develop a methodology to allow a rider to “profile” themselves, vs leaving it up to the rider to figure it out.
This could be used to inform training zones, but more importantly inform training plans.
This is one of those things where, a bit like the 8min and 20min FTP tests, you don’t know what level to test at because you don’t know how you’ll perform at any given level. So you need to test to find out that your proposed level isn’t correct. Rinse and repeat.
Unless you regularly do similar efforts in races or club rides then you simply aren’t going to know what FTP+ efforts should feel like. Maybe some workouts outside plans that are low TSS so don’t tire you too much but let you characterise your power curve, a bit like the old 8min test but shorter, a warm up and then a couple of VO2 max efforts that you are encouraged to adjust the level. Maybe do them shortly after the Ramp Test. This value either then gets added to TR’s algorithm for the subsequent workouts or you remember it to up/down the intensity at the appropriate points. It might need doing for several durations, 5sec, 30sec, 1min, 3mins, etc.
Whether the above is self profiling or letting the rider figuring it out for themselves is a moot point.
Agreed. For example right now in vo2max workouts you either need to:
And then once you determine your individual % above FTP, which changes based on interval and fitness, that is where you are left hanging dry on the next workout. For example lets say I drop intensity to 94% during vo2 workouts (120% becomes 113%). Annoyingly TR doesn’t retain the setting for next vo2max workout. You have to either remember it, or trial and error again. Frankly that sucks and I’d really like to see TR get busy and fix that as part of the mobile app update. If fixing “no support for personalized % above FTP” requires using current Plan Builder then it hasn’t been fixed because current Plan Builder is too rigid if you are modifying plans (and I do). The lack of custom TR workout syncing to headunits is annoying as hell but not a deal breaker for me.
I’ve built a lot of custom workouts in TrainingPeaks and Garmin Express. Really prefer searching TR workouts, modifying one in WorkoutCreator, and using TR app on the indoor trainer. And then do analysis in WKO5. Still very happy paying TR annually for that convenience. Proper performance analytics would largely avoid the need for WKO5 in my workflow, however from hint dropping I think TR has set its priorities on some good stuff. Time will tell.
I don’t agree with this idea that if only my workouts were pre-set at just the right wattage I would magically be a bunch faster. And I’m sure as hell not interested in dragging myself through Sufferfest’s 4DP test regularly enough to get any supposed benefit from it. If my workout is a bit hard on the day I nudge the power down a bit, if it’s a bit easy I nudge it up. Just like if I was doing intervals on the road.
If you take FTP for what it is, an estimation of the specific output that causes a specific physiological response, then it’s one of the most valid metrics around all of sport for not only defining aerobic workout intensities, but also anaerobic. If you know the functional ceiling, then you know that going above that ceiling trains a different system (anaerobic). In cycling, anaerobic is less important than, and ultimately still reliant on, the aerobic fitness. Virtually all cycling programs use FTP in some capacity, those mentioned included. They may not even call it FTP, but they use it.
So my argument is this: it doesn’t require continual maximal test data and the resultant nuance, in this case, to have “optimal” anaerobic workouts and get those cherry-on-top gains from higher intensities. Practically, feeding a model with a grueling maximal effort test protocol to keep your highly-specific zones dialed in doesn’t make a lot of sense. That’s not smart training when it’s a model that won’t necessarily represent your physiological capabilities any better than % of FTP can. You can fluctuate 5, 6, 10% in a week, just like you could on the day of a brutal maximal effort, self-paced test. Zones, and TR workouts, account for that just fine, so it doesn’t need to be much more specific. If FTP is roughly 95-105%, then 115% is a great place to start for anaerobic training. TR handles any further required nuance by saying “this is what the effort SHOULD feel like. Adjust up or down if that isn’t your experience within the intervals” rather than giving you a test that may or may not be a true maximal effort every time you do it. It looks fancier, but it’s certainly not always going to result in better training, and it’s certainly a lot of weight to place on a single day of self-paced training every month or so. At the end of the day, it’s an aerobic sport, and you can do more, better, faster anaerobic efforts by being concerned with aerobic fitness, which FTP corresponds to.
I’ll disagree with the words of Sebastian Weber:
If you don’t know what is behind the FTP, you can’t really establish a precise and focused training programme.
Decreasing VLaMax and increasing VO2max would require different training programmes.
Neither SF/4DP nor TR/FTP can test what a rider’s VLaMax and/or VO2max is; at best, they both provide unprecise and unfocused training programmes.
However, most of us are no where near requiring highly precise and focused training; we can get away with simply riding more. 
However, most of us are no where near requiring highly precise and focused training; we can get away with simply riding more
I think you’ve nailed the major question. What is precise and focused, and do I actually need it? Maybe? Regardless, I think it almost entirely relies on feel to truly determine what that is. Even in a lab test, you have to pace or put out the mental effort for a max value. I think % FTP allows the required precision to the extent that the precision and focus makes training more optimal. Knowing what’s behind an effort seems to me like identifying what a zone “feels” like and then adjusting based on that. I think that’s practically not going to change much for an individual even with highly specific measurements or tests.
We don’t test the other systems for the same reason why we aren’t pushing the 20/8 minute test. You need to know your result before you take the test in order to pace correctly and get an accurate result. What you’re most likely to end up with is an imprecise result.
There’s also the repeatability aspect of this, and what your score in the test is not reflective of what you might be able to do multiple times.
We also know that VO2/Anaerobic/Sprint can improve quickly, especially if you haven’t been working them recently. Ramp test gets you close, then we work you through progressions for supra-threshold systems. We’re fairly aggressive with our progression and suggest that people turn it down or backpedal an interval if they are on the edge. This tends to get everyone but we’re going to improve it.
The crazy thing is that people will not hit everything in their first vo2 workout in the plan, but quickly adapt and nail the other, hard workouts later in the plan before the retest.
Yup and the rate of progression are different for people too plus the inaccuracies of testing with a method that benefits from perfect pacing.
I don’t want to spill all the beans now, but adaptive stuff is right around the corner (being implemented now). It will be a series of releases rather than one giant AI that has sentience.
Another issue I have with the 4DP thing is the labelling at the end of the test.
I did the 4DP after my A Race (pre-TR days) and it labelled me as a climber. Yet I had just done 3 months of time trial training (w/ a successful result). And it showed when over the next couple of weeks the only KOMs I could nab were flat/rolling segs.
I can remember one of the <#100 podcasts where the TR team talks about training to be whatever kind of rider you want to be — be a sprinter one season, go be a climber the next season, etc. SF pidgeon-holes you right out of the gate, and if you are a complete beginner and/or believe the hype you could miss out on a heck of a lot of opportunity and fun to try other types of riding. At least TR provides plans for the type of rider you want to try to be instead of the other way around.
$0.02
Bradley Wiggens went from a track star to winning the TdF. I think far too many people label themselves too early.
Do the magic hand thing!
Haven’t Jones, Poole, etc., argued that critical power is a better indicator of the division between intensity domains than any physiological measurement? If so, doesn’t that make TR’s approach better than any lab-based test?
Lol, it will be small steps along the way.
Great.
Thanks @Nate_Pearson for triggering a soon-to-be endless string of excitable memes.
Tru dat.
I think the center of the bell curve is 68%, but point well made.