Login required to started new threads

Login required to post replies

Prev Next
Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it.
Quote | Reply
Another year of using WKO4 to crunch the numbers of cyclists power and other metrics has taught me that using short durations tests is a waste of time.

Majorly overestimate the FTP. Something Andy Coggan refers to as Vanity FTP.

Usual suspects are 95% of a 20min test or even worse a percentage of a shorter duration test.

Modelled FTP in WKO4 really highlights the folly of using just short term maximal efforts to determine threshold power, no matter what your definition of threshold is. A high FTP and a very short Time to Exhaustion (TTE).

This is well modelled in WKO4 for shorter durations as well. PMAX/FRC (Sprint Power), FRC (Track TT Power) and FRC/FTP (Pursuit Power).

Hamish Ferguson: Cycling Coach
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Kiwicoach] [ In reply to ]
Quote | Reply
how should the average cyclist estimate than? I have used 20 min TR FTP test at 95% power for a couple years. Sure I feel it's overestimating a bit but it at least gives me a baseline test to keep constant. If it's 3% off I am not sweating it.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [holograham] [ In reply to ]
Quote | Reply
holograham wrote:
how should the average cyclist estimate than? I have used 20 min TR FTP test at 95% power for a couple years. Sure I feel it's overestimating a bit but it at least gives me a baseline test to keep constant. If it's 3% off I am not sweating it.

Should at least do it as Hunter Allen prescribed with a 30s maximal effort and a 5min maximal effort preceding the 20min test.

Hamish Ferguson: Cycling Coach
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Kiwicoach] [ In reply to ]
Quote | Reply
Kiwicoach wrote:
holograham wrote:
how should the average cyclist estimate than? I have used 20 min TR FTP test at 95% power for a couple years. Sure I feel it's overestimating a bit but it at least gives me a baseline test to keep constant. If it's 3% off I am not sweating it.


Should at least do it as Hunter Allen prescribed with a 30s maximal effort and a 5min maximal effort preceding the 20min test.

I've been using the Spinervals Threshold test for many years. I has 3 x 30 secs, 3 min maximal, and then 3 x 1 min maximal, then the 20 min test. Regardless, it hasn't hurt me at all in execution even if Kiwicoach is all besides himself over it.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [holograham] [ In reply to ]
Quote | Reply
holograham wrote:
how should the average cyclist estimate than? I have used 20 min TR FTP test at 95% power for a couple years. Sure I feel it's overestimating a bit but it at least gives me a baseline test to keep constant. If it's 3% off I am not sweating it.


To me, it doesn't make a hill of beans how I estimate it. It matters only that the %-age I utilize for pacing - works for ME.

I couldn't care less what someone else thinks about the way I test. If it's possible, I care even less about how someone else does it.
Last edited by: nc452010: Dec 27, 17 12:01
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Kiwicoach] [ In reply to ]
Quote | Reply
Kiwicoach wrote:
Another year of using WKO4 to crunch the numbers of cyclists power and other metrics has taught me that using short durations tests is a waste of time.


However short duration tests are fantastic for testing short-duration power, like sprint power, track tt power, or pursuit power. Why go to WKO4 to calculate "PMax/FRC" to get sprint power when you can just go do some sprints!

Why bother with anything "modeled" at all when the real thing is so easy to test for these days with power meters on just about every bike and trainer? (at least in Slowtwitch land). Coggan himself says something like "the best test is the performance itself" (or something very similar to that) about 12x per day

I can see modelled FTP being useful because doing a "pure" FTP test regularly sucks for those who don't do pure FTP-test-like races. (FTP is super easy for me to track because I do lots of pure 40K time trials). Though I'm fast enough at 40K that you might claim it overestimates my FTP because it's too short. :)
Last edited by: trail: Dec 27, 17 12:08
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [holograham] [ In reply to ]
Quote | Reply
Quote:
how should the average cyclist estimate than?

Don't estimate it. Do a 40K TT and get the answer. It will only take about an hour and it's a great workout.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Supersquid] [ In reply to ]
Quote | Reply
Supersquid wrote:
Quote:
how should the average cyclist estimate than?


Don't estimate it. Do a 40K TT and get the answer. It will only take about an hour and it's a great workout.


Ditto - I do the full hour and hold parameters steady. If there was more there, I might wait 10 days and try again, but for me its remarkable how fine a line I can walk by "feel", where you know "I'm gonna blow up", vs "hurts, but is going OK".
Last edited by: Testament TN: Dec 27, 17 12:53
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [trail] [ In reply to ]
Quote | Reply
trail wrote:
Why bother with anything "modeled" at all when the real thing is so easy to test for these days with power meters on just about every bike and trainer?

Why indeed.

In this context, the purposes of modeling are 1) to extract accurate and precise parameter estimates reflective of different underlying physiological determinants of performance (without necessarily performing any formal tests), and 2) to smooth the measured mean maximal power data a bit, to provide a robust point of reference for additional calculations (e.g., you can imagine how much any adaptation score would bounce around if you based it on the raw mean maximal power data).
Last edited by: Andrew Coggan: Dec 27, 17 12:56
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Supersquid] [ In reply to ]
Quote | Reply
Supersquid wrote:
Quote:
how should the average cyclist estimate than?

Don't estimate it. Do a 40K TT and get the answer. It will only take about an hour and it's a great workout.

I did 4x15’ at around what I thought was my FTP with 1’ spin in between each. The resulting watts for an hour was two under my initial estimate

I also base FTP off what I can average over longer 2-4 hr rides. Looking at all ride data can also give a good idea. This is w/o WKO.

Blog: http://262toboylstonstreet.blogspot.com/
https://twitter.com/NateThomasTri
Coaching: https://bybtricoaching.com/ - accepting athletes for 2023
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:

In this context, the purposes of modeling are 1) to extract accurate and precise parameter estimates reflective of different underlying physiological determinants of performance (without necessarily performing any formal tests)


But not nearly as "accurate and precise" (whatever that means) as when performing formal tests (ideally performance itself). Calling something "accurate and precise" is only meaningful when you have some reference truth to measure precision and accuracy against. And that's performance. So if it's easy to go out and perform to acquire data for those 3-4 key parameters (e.g. sprint power, pursuit power, FTP, et al), might as well do that. I find it pretty easy to do, so I feed that information to the model so those key points have the highest quality data I can get. Then I can let it fill in the rest for me.

I do completely understand the benefits of WKO4 (and similar) in providing things like great longitudinal tracking of parameters and providing clear, easy-to-digest graphical representation of my current "signature" as a road cyclist.

I was being a bit facetious because I tire of the constant heartache on this forum about FTP testing accuracy.


Last edited by: trail: Dec 27, 17 13:27
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Kiwicoach] [ In reply to ]
Quote | Reply
I don't honestly see what all the fuss is with it in the first place. Who cares if you over or underestimate it a bit?

Do a workout; if you hit the numbers, great, if not, adjust them. Make a note for next time.

Seems pretty simple when you get right down to it.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [trail] [ In reply to ]
Quote | Reply
trail wrote:
I do completely understand the benefits of WKO4 (and similar) in providing things like great longitudinal tracking of parameters

Actually, I don't think you entirely do.

As an example: there is no test duration that would provide a 'pure' indication of FRC. It can only be separated from other determinants of performance via modeling.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [rubik] [ In reply to ]
Quote | Reply
rubik wrote:
I don't honestly see what all the fuss is with it in the first place. Who cares if you over or underestimate it a bit?

Do a workout; if you hit the numbers, great, if not, adjust them. Make a note for next time.

Seems pretty simple when you get right down to it.

There is a bit more to it than that, though.

For example, successful use of various pacing guidelines requires having a reasonably accurate estimate of your FTP. If instead you go by your 'vanity FTP', you are likely to overshoot the mark, and pay the consequences.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [natethomas] [ In reply to ]
Quote | Reply
natethomas wrote:
I did 4x15’ at around what I thought was my FTP with 1’ spin in between each. The resulting watts for an hour was two under my initial estimate

Deadly sin #6 is my favorite as well:

http://lists.topica.com/...e.html?mid=910290920
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [rubik] [ In reply to ]
Quote | Reply
Agreed, wholeheartedly. It seems like so many feel they have to retest to justify moving FTP up. Or if the test shows a certain number that they are somehow prohibited from using a different FTP number for workouts. I “manually” adjust mine up (and possibly down) without testing regularly when I notice previous targets are too easy/hard.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:
trail wrote:

I do completely understand the benefits of WKO4 (and similar) in providing things like great longitudinal tracking of parameters


Actually, I don't think you entirely do.

As an example: there is no test duration that would provide a 'pure' indication of FRC. It can only be separated from other determinants of performance via modeling.


Yeah, I just, personally, haven't found FRC all that useful. Not because it isn't inherently useful, just that it's not part of my current toolbox. Disclosure: I don't regularly WKO4 (though I bought a license and tried to use it). Not because of it's science/math underpinnings - other reasons.

Edit: But what I don't understand is the OP's implying he uses FRC as some sort of proxy for track TT power. When hopefully his athletes would do actual track TTs if that's what they're training for. That's sort of what I was needling him about - being uppity about modeled FTP accuracy then using a non-directly-measurable parameter like FRC as a proxy for something very easily directly measurable. Getting a solid track TT measurement for someone with access to a track takes 3-5 minutes after their regular track warmup!
Last edited by: trail: Dec 27, 17 14:38
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [HuffNPuff] [ In reply to ]
Quote | Reply
HuffNPuff wrote:
Regardless, it hasn't hurt me at all in execution even if Kiwicoach is all besides himself over it.

Yeah, keeps me up at nights.

Hamish Ferguson: Cycling Coach
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [nc452010] [ In reply to ]
Quote | Reply
nc452010 wrote:
holograham wrote:
how should the average cyclist estimate than? I have used 20 min TR FTP test at 95% power for a couple years. Sure I feel it's overestimating a bit but it at least gives me a baseline test to keep constant. If it's 3% off I am not sweating it.


To me, it doesn't make a hill of beans how I estimate it. It matters only that the %-age I utilize for pacing - works for ME.

I couldn't care less what someone else thinks about the way I test. If it's possible, I care even less about how someone else does it.

That's nice and sometimes that's all that matters.

But what if you turn up on race day and find your current threshold doesn't cut it in competition. Sure alls you can do is alls you can do but even when it is 100% me, how do you know that is your true 100%.

Hamish Ferguson: Cycling Coach
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Kiwicoach] [ In reply to ]
Quote | Reply
When did shilling become allowed on this forum?
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [trail] [ In reply to ]
Quote | Reply
trail wrote:
But what I don't understand is the OP's implying he uses FRC as some sort of proxy for track TT power. When hopefully his athletes would do actual track TTs if that's what they're training for. That's sort of what I was needling him about - being uppity about modeled FTP accuracy then using a non-directly-measurable parameter like FRC as a proxy for something very easily directly measurable. Getting a solid track TT measurement for someone with access to a track takes 3-5 minutes after their regular track warmup!

Not to speak for Hamish, but suppose you were coaching a pursuiter who was trying to meet the time standard for automatic selection to the national team (and hence access to financial and 'in kind' support, invitations to team training camps, etc.). You could measure how much power they could produce for the requisite duration, and with some additional measurements/knowledge estimate how close they were to meeting their goal. What that would not tell you, though, is where the greatest remaining gains are likely to be found. OTOH, if you knew that their mFTP was at an all-time high but that their FRC was lacking, you would have an answer to that question, and would have a better idea of how to proceed.
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [aravilare] [ In reply to ]
Quote | Reply
aravilare wrote:
When did shilling become allowed on this forum?

??

Who do you think is shilling?
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [trail] [ In reply to ]
Quote | Reply
trail wrote:
Kiwicoach wrote:
Another year of using WKO4 to crunch the numbers of cyclists power and other metrics has taught me that using short durations tests is a waste of time.


However short duration tests are fantastic for testing short-duration power, like sprint power, track tt power, or pursuit power. Why go to WKO4 to calculate "PMax/FRC" to get sprint power when you can just go do some sprints!

Why bother with anything "modeled" at all when the real thing is so easy to test for these days with power meters on just about every bike and trainer? (at least in Slowtwitch land). Coggan himself says something like "the best test is the performance itself" (or something very similar to that) about 12x per day

I can see modelled FTP being useful because doing a "pure" FTP test regularly sucks for those who don't do pure FTP-test-like races. (FTP is super easy for me to track because I do lots of pure 40K time trials). Though I'm fast enough at 40K that you might claim it overestimates my FTP because it's too short. :)

Well yes, I had a Sprint Cyclist in Rio, nice little cartwheel in the Keirin. She doesn't do any FTP testing.

WKO4 tells you not only your power at PMAX/FRC but the duration you can sustain this. So while a PMAX/FRC of 2400 watts sounds impressive the ability to hold it for only 3-4 means a rider who will start fast and die just as fast. Vanity PMAX/FRC.

As Andy mentioned we test the model to normalise the data. To know whether our good is good enough. To compare with others and to compare our progress. A rider turned down for a Development U23 team in U19 because they couldn't see any further progress in the rider or a rider in our coaching group signing for a World Tour team purely off his power test data (clearly a solid background on track and conti pro leading into that).

Andy does often remind us that "the best predictor of performance is performance itself" but then no two races are ever the same so is what you are testing actually relevant to your test. In cycling the FTP is strongly related to performance in everything from a pursuit to a long road race.

I would argue a well developed threshold with good time to exhaustion is the best way to start the specific preparation for a goal performance.

Hamish Ferguson: Cycling Coach
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [trail] [ In reply to ]
Quote | Reply
trail wrote:
Andrew Coggan wrote:

In this context, the purposes of modeling are 1) to extract accurate and precise parameter estimates reflective of different underlying physiological determinants of performance (without necessarily performing any formal tests)


But not nearly as "accurate and precise" (whatever that means) as when performing formal tests (ideally performance itself). Calling something "accurate and precise" is only meaningful when you have some reference truth to measure precision and accuracy against. And that's performance. So if it's easy to go out and perform to acquire data for those 3-4 key parameters (e.g. sprint power, pursuit power, FTP, et al), might as well do that. I find it pretty easy to do, so I feed that information to the model so those key points have the highest quality data I can get. Then I can let it fill in the rest for me.

I do completely understand the benefits of WKO4 (and similar) in providing things like great longitudinal tracking of parameters and providing clear, easy-to-digest graphical representation of my current "signature" as a road cyclist.

I was being a bit facetious because I tire of the constant heartache on this forum about FTP testing accuracy.


Similar? Nothing similar to WKO4.

I say it again, no two performances will ever be the same. Am told in swimming the saying is "you never dive into the same water twice".

Hamish Ferguson: Cycling Coach
Quote Reply
Re: Feels like a knife to the heart every time someone misuses the concept of FTP and how you test it. [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:

Not to speak for Hamish, but suppose you were coaching a pursuiter who was trying to meet the time standard for automatic selection to the national team (and hence access to financial and 'in kind' support, invitations to team training camps, etc.). You could measure how much power they could produce for the requisite duration, and with some additional measurements/knowledge estimate how close they were to meeting their goal. What that would not tell you, though, is where the greatest remaining gains are likely to be found. OTOH, if you knew that their mFTP was at an all-time high but that their FRC was lacking, you would have an answer to that question, and would have a better idea of how to proceed.

I could see how, in principle, that would be a useful tool in the toolbox of a coach. (I say "in principle" only because I, as noted, haven't put much effort into adding that tool to be my toolbox to speak directly about it with any authority).
Quote Reply

Prev Next