I do not fully understand the outstanding obsession with validating / proving / disproving the Ramp test. Nate has essentially suggested that they have programmed the Ramp test to provide you with an estimated FTP which is like 0.72 of your highest power achieved during a Ramp test.
They’ve taken an existing protocol and made a few changes to provide their customers with a FTP esitmation that they feel will best work with their product.
I do not belive that they are going to be sharing their data from their test as it has significant commerical value to their product. Typically any time one of these companies make anything and publish the data they get ripped off.
The human body and our training of it on the bike is a complex and multivariate system where there is an addition of significant variance do to environmental factors. So, like any complex system it is typically not worth the effort to make a model which is statistically perfect through the user base. It isn’t going to happen, there are too many variables.
So anyone Pro to Cat 5 is still going to benefit just the same from a test that they use consitentily in their training.
That said, higher level folks might get some coaches who have insights and ideas on how to manipulate their training other than just their FTP value. Volume and types of intensity, volume and timing of training / racing.
In the end I hope we can stop having daily hand wringing threads on the subject of whether the Ramp test or any other test works. The Ramp CP8 and CP20 tests all work and all have variance associated with them, since they are all providing esitmates and not the ‘actual’ value.
For what it’s worth, the progressive ramp test has proven to be consistent with 2x8 and 1x20 testing protocols in my experience. I’ve been doing power based training since 2009, so that is my window. As with any performance test it’s a snapshot in time. Feel free to bump it up or down based on continual feedback observed on an ongoing basis. The tests do what they are supposed to do, which is establish baselines. It’s up to you to evaluate the data and apply it to your specific situation.
I’m still not quite sure what your goal is: one of TrainerRoad’s crown jewels is its database, based on which they can develop their training. It is correct that not publishing peer reviewed studies means there is no definitive publicly available evidence that the ramp test correlates well with other methods to measure your FTP or lactate threshold. Nevertheless, I have no doubt that TR’s approach is data driven, it is just that their data is proprietary. In my experience, the ramp test works quite well to my scale workouts appropriately — which is the ultimate purpose at the end.
Even if TrainerRoad did release its data (without compromising their users’s privacy), you would have to discuss the quality of the data. The ramp test is way easier to do for beginners than a 20-minute FTP test, which as far as I know is the old-school gold standard.
Moreover, even if you get an accurate measurement of your FTP, for a proper analysis of your abilities you’d need to assess other characteristic quantities such as VO2max power, sprint power, 5-minute power, etc.
So if your VO2max power is not 120 % of your FTP (which is the basis used in most workouts), but, say, 125 %, then you’d leave something on the table in workouts.
Put another way, TR is not as good as a pro-level coach who assesses you regularly and prescribes tailor-made workouts. TR’s purpose is to give you 90+ % for a fraction of the price. (Great coaches don’t scale.) Lastly, not even coaches always or solely rely on science, experience plays a huge role for them. One team mate of mine, who upgraded from the lowest to the highest amateur category within one season, has a coach. He doesn’t even seem to know his FTP, yet, races successfully. I’m not advocating for this approach, but there seems to be more to it than FTP tests.
I’d be more interested to hear the Standard deviation of say Ramp test FTP vs 40-60 minute power or the properly executed 5m+20m test method. Then what proportion of users fall more than 1, 2, 3 standard deviations above/below the mean.
Unfortunately as in a lot of sport-science, the magnitude of error that makes a material difference to an athlete’s performance/training is much smaller than would qualify as significant under statistical methods. Claims like “this works for most people” could be made whether a method fits for 51% of the population or 99%, and doesn’t give information about how wrong it could be if you are an outlier.
Just tossing this in for humour value: There is a lot of debate around the usefulness of FTP itself and its relation to MLSS (or not). So agonising over how best to estimate FTP probably isn’t that important. Find a test that works for you and stick with it.