Are you not arguing in good faith or is there something else going on?
You quote “…who don’t test” when the actual quote was “who don’t test or who use other testing methods such as the 8- or 20-minute test.”
You quote “…on TrainerRoad” as if that’s disqualifying my claim when my original statement was “most successful predictor of being able to complete their workouts at a given ftp than any of the other tests”
Who the heck did you think I meant by “their workouts” if not TrainerRoad?
You can pick things apart to be stubborn and to resist the unpleasant feelings of cognitive dissonance but my original claim was so closely matched to the quotes I found on their blog it’s phenomenal I did it from memory.
Mine:
access to all the workout data
ramp test is the most successful predictor of being able to complete their workouts
Theirs:
7000 tests completed during beta…hundreds of thousands of tests since released
data shows riders who take ramp test fail workouts at a lower rate that those who … use other testing methods such as 8 or 20 minute
Perhaps we should be discussing the clock face test rather than which ftp test works best
“the information gleaned from the ramp protocol is more accurate”
“than an FTP estimate based on a poorly paced 8- or 20-minute test”
I’ll not bore you with my anecdata, and simply say that for me the statement above is not true. In my opinion, looking at my data, even slightly over or under paced 20-minute tests give better information on my threshold than the ramp.
And I believe the ramp best 1-min power does a great job at defining a ceiling on 4 or 5 minute all-out efforts.
There was a thread started by Brennus a while ago, discussing how long people have been able to hold their ramp 1-min power. It was an interesting thread, although largely inconclusive because a lot of people don’t go out and do all-out 4-6 minute efforts.
Fair enough, you’re right. TR has written training plans based on execution of the ramp test. Thus executing the ramp test makes it more likely that you’re going to complete their training plans.
I’m sorry, I was arguing another point which was not what you were discussing - I recognize that now - and that is that the ramp test is not predictive of performance, nor is it a particularly accurate way to estimate FTP for a substantial portion of the population.
That said, I don’t appreciate the accusation that I’m not discussing things in good faith, and at this point will end my participation in any further discussion here. Have a nice day!
I like this question…as much as it pains the ego, I am personally in the camp that it is much better to train with an FTP that is too low.
Not only does training at an FTP that is too high cause everything that you mentioned, but psychologically I think it can be devastating, especially if you are constantly failing workouts.
Also, I think for most of us, if we are training at an FTP that is too low and the workout ends up feeling too easy, there is a high likelihood that the person increases the intensity manually anyways!
The bad faith part is intentionally clipping off “or use other testing methods” and claim it only says the ramp test is better than nothing.
They used the expression “poorly paced” once while then saying it outperforms “those who use the 8 or 12 minute test” over 100s of thousands of tests worth of data.
The relevance of “poorly paced” does not mean they handpicked the riders who paced poorly. It’s an attempt to explain that based on the mountain of data, the ramp test gets users to the right FTP more often. By “right” I (and they) mean the one that allows them to successfully complete their workouts.
Which is 100% in line with my claim “ramp test is the most successful predictor of being able to complete their workouts”
What I said (and what the blog says) can both be paraphrased as follows:
If you use the ramp test, you are more likely to complete your TR workouts than if you do one of the other tests.
This would be a fun discussion over a couple of beers, as there are other explanations and interpretations of the simple top-level stats that TR provides. Throw this into the discussion
Maybe those users weren’t serious? Just kicking the tires? Or simply using it when they can’t ride outside? What if, like some local heavy hitters, they don’t upload their outside riding to TR and simply use it for trainer control instead of getting on Zwift when they can’t ride outside? What conclusion can TR draw if it has no visibility into outside testing or all the workouts?
I say we sit down after a ride, quaff some fine beverages, and have a fun debate
Do you have a source for that? Just curious as I haven’t seen anything other than 75%.
72-77% 78% was MAP (Ric Stern). I think 93-97% was for the 20 minute TT (if one completed the 5 min VO2 Max test beforehand, Dr. Coggan & Hunter Allen). I don’t believe CTS specified a range for the 2x8 minute test, 90% best power average of the two.
Edit: Thought I seen 78% in more recent post but I was wrong
1. How does TrainerRoad calculate my FTP from the Ramp Test?
We take 75% of your best one-minute power. If you are above target for the final minute, the app will reduce your result by a small amount. It’s important to follow the target power closely as you progress through your test.
You have to go back to when there was a Facebook TR group (pre forum) with Ramp Test X (for experimental) and we would take it and wait for support to give us the secret calculation that evolved over many weeks.
I was a huge fan of it back then because I hated long indoor intervals and didn’t have enough mental toughness to execute them, including the long tests.
But SSB was hell for all the threshold workouts. Eventually I got into SSB high volume outside at a lower ftp and that was much more game changing for me to learn how to push myself. These days, I would never use a ramp test because I also know I get a heavily biased result due to high relative anaerobic capacity, and I regularly do 3 or 4 x 20 min type of rides. But early cycling me probably still wouldn’t listen to older and wiser me so ramp test may still have its place
I missed the context above with my other reply, but that is true. TR used something around 78% for my very first test as part of the beta group testing. It was discussed a bit on Strava, but largely in the FB TR Beta group what was separate from the existing regular FB TR group.
TR support would email results on my first ramp tests in early 2018. My fuzzy recollection was they started with 77% but it could have been 78%. Emails deleted and although easy to do, I’m not going to pull up the first tests and figure out the %.
That may be what happened but honestly I don’t know why they would push the ramp test for any reason other than helping their users complete the workouts
We know that users are more likely to keep coming back (and keep paying for TR) if the workouts are productive AND if they can complete them.
They saw that some users skipped the tests and those users did poorly at completing the workouts
They then introduced the ramp test and found that users who completed the ramp test were more likely to complete the workouts than those that used one of the other ftp tests.
Maybe I drank the koolaid but I truly believe if 12 minute test users did better on the workouts, that’s the one they’d be recommending to everyone.
Caveat: I’m talking about on the aggregate. I know there are outliers who nail the 12 min test perfectly and it’s better for them. But the ramp test results in the right number for more users so it’s more likely (statistically) to get you the right number.
Are you a highly trained stud that does better on a 12 min test? Okay np. Not arguing that happens sometimes.
But what started this whole thing… I do not think AT is a bandaid for a broken ftp system. The ramp test is the bandaid bc many of their users were testing inaccurately before the ramp test which lead to unfinished workouts. The ramp test fixed that for most of their users. AT came along to fix that one number doesn’t cover every single zone and users don’t improve in every zone at the same rate
Bit of historical perspective - a quote from he who shall not be named on the wattage forum in a “7 deadly sins” discussion in 2004:
BTW, another method that could be added to this list would be to do an
incremental exercise test to determine ‘MAP’, then estimate functional
threshold power as being ~75% (range 72-77%, using Ric Stern’s
guidelines) of this value. You could then use this estimate as is, or if
necessary/desired, further refine it using one of the methods described
above (e.g., by doing a TT).
BTW, the reason this approach works is because in trained cyclists, LT
falls within a fairly narrow range as a percentage of VO2max, and there
is tendency for those with the highest LTs to have the lowest anaerobic
capacities (and therefore a slightly lower MAP relative to power at
VO2max), and vice-versa. In any case, at the very least knowing your MAP
will help ‘bracket’ what could be considered a reasonable range into
which you expect your threshold power to fall.
Late to the party and just listened to the podcast starting at the 56:00 mark.
The discussion came across cocky and arrogant with a disingenuous characterization of the Ramp Test, TR and AT. There are limitations to the ramp test but KM’s description and understanding of TR’s utilization of it is a joke .
I would just like to say though I do deeply appreciate how @Nate_Pearson, @Jonathan and the team do not engage in trash talking people or other companies. The TR team seems more apt to engage in intellectual debate rather than O my WhO duZ ramp Test?!? So uneducated. Tsk tsk.”
There might’ve been an episode discussing “Revolver” from the SufferFest and that it was almost an unnecessary workout but I might be misattributing.
That’s not what happened though. They rushed out polarized plans for sure, after slating from some people. AT has been in the works for a lot longer than that.
Ramp test is a poor ftp test for the reason I already mentioned.
However, I think it’s smart to be using it to set training zones because you can never fail the workout since it is a progression to failure by design.
Contrast this to a steady ride of any duration and you get the pacing argument as to why it isn’t good (went too hard and blew up, didn’t go as hard as you could have, and everything in between).
So from a practical application stand point, I would favor a ramp to estimate training zones over a test that has user- error (pacing issues)
.