Big difference between 'tested' FTP and computed CP/FTP - What does it mean?

[Peter Riegersperger]

This is my first season of training with a power meter, and I’m just getting the hang of it and try to understand all the data I’m collecting over the months. What puzzles me is that I have a big difference between my tested FTP and CP/FTP computed by the model.

My FTP apparently is 249W. This is the result of the „95% of 20 min time trial effort“ test suggested by Allen/Coggan, and also a real life performance (climbing Alpe d’Huez in 01:10:31 with an XPower of 249W).

 

I know that FTP and CP are conceptually different things, but what I’ve read so far suggests that they should be roughly the same value (or at least in the same ballpark).

When using the 2-Parameter-Model, GC puts my CP at 267W, which is quite a bit off (18W, or 7,2% higher than FTP). The Extended CP Model (with all parameters set to their default) calculates CP as 270W, and FTP appears to be overestimated with 262W. (Both computations use the whole seasons data)

This difference between what I observe in real life and what the model computes confuses me. I don’t think that I could have gone significantly harder when climbing Alpe d’Huez. Even assuming that I rode conservatively, adding adding 13W average seems to be a stretch.

I’m trying to understand where this gap comes from and what it means. Do some of you experience the same?

Should I just ignore it, or is this valuable information that I can incorporate into my training (Nathan, one of your posts in a different thread suggested it is)?

all the best,

peter

[Nathan Townsend]

On Friday, 24 July 2015 21:05:43 UTC+3, Peter Riegersperger wrote:

This is my first season of training with a power meter, and I’m just getting the hang of it and try to understand all the data I’m collecting over the months. What puzzles me is that I have a big difference between my tested FTP and CP/FTP computed by the model.

My FTP apparently is 249W. This is the result of the „95% of 20 min time trial effort“ test suggested by Allen/Coggan, and also a real life performance (climbing Alpe d’Huez in 01:10:31 with an XPower of 249W).

Can you break your Alpe d'Huez ride into 10min segments, calculate average power for each segment, and post it up here (or alternatively just post a screen shot of the power for the entire effort (smooth power to 30 s moving average for clarity).

Secondly what is the average power (not Xpower which is a manipulation of the raw data) for the entire effort?

 

 

I know that FTP and CP are conceptually different things, but what I’ve read so far suggests that they should be roughly the same value (or at least in the same ballpark).

When using the 2-Parameter-Model, GC puts my CP at 267W, which is quite a bit off (18W, or 7,2% higher than FTP). The Extended CP Model (with all parameters set to their default) calculates CP as 270W, and FTP appears to be overestimated with 262W. (Both computations use the whole seasons data)

This difference between what I observe in real life and what the model computes confuses me. I don’t think that I could have gone significantly harder when climbing Alpe d’Huez. Even assuming that I rode conservatively, adding adding 13W average seems to be a stretch.

I’m trying to understand where this gap comes from and what it means. Do some of you experience the same?

Should I just ignore it, or is this valuable information that I can incorporate into my training (Nathan, one of your posts in a different thread suggested it is)?

Andy Coggan conceptualises FTP as the highest sustainable aerobic power, and CP has almost the same definition but adds the phrase "without incurring a progressively increasinging non-aerobic energy contribution".  While this is sort of implied by Coggan's FTP, we can see that conceptually they're supposed to be the same thing ie: the highest sustainable aerobic power without incurring a progressively increasing contribution from non-aerobic sources.

 

The problem is the word "sustainable".  Coggan believes that humans can commence a 60min effort at this intensity and hold it there for 60min (which thus implies that no fatigue whatsoever occurs in this time), but there is evidence which suggests that most humans, even well trained cyclists, are incapable of holding this intensity for 60min without developing some fatigue which induces inefficiency (ie: less ATP per volume unit of oxygen consumed by the muscle), and thus having to either increase the VO2 slightly to maintain power which has its limits, or decrease the power slightly in order to continue at a steady VO2.  There is growing evidence that both central (nervous system) and peripheral (muscular) mechanisms contribute to this inability to maintain power at threshold for long periods of time eg: >20-30min for most people, and probably as long as 40-50min in those whom are highly endurance trained or having very high % type I fibers, such as pro cyclists and/or elite TT'ers.

So using your data given above, your 20min best effort was about 263W, which is not far away from the GC estimate of 267W.  So in reality there is about a 15W difference between the CP estimate and FTP estimate.  From a practical perspective, to be honest I don't really think this difference makes a big deal because most people set their interval intensities well beyond FTP + 15W and therefore they're working above CP and in the severe intensity domain anyway ie: an interval session where you set the power at FTP + 65W is basically the same workout as CP + 50W.  The important thing is the actual power you're pushing and how long you hold it for, not how you derived that number in the first place. 

However if you wanted to do long intervals just above threshold, and therefore in the severe intensity domain which will push VO2 up close to max upon nearing the end of the interval, then if you did these intervals around 255-260W, you just won't get there (ie: near VO2max).  However if you do them at about 270-275W then my bet is that this intensity will definitely not feel sustainable, and hence towards the end of a longer 12-15min interval perception of effort will resemble that which occurs during a much shorter but harder effort eg: 5min. However a 15min interval at 255-260w, despite the fact it is supposed to be above threshold according to Coggan, will "feel" as though it is sustainable and you can continue for longer. 

I think there is good value in doing these longer efforts in the severe domain, however again, from a practical training prescription perspective, if you belong to the Coggan FTP school of thought, a hard 15min interval which might only be 5-10W above CP would simply be equivalent to an FTP + 25W interval workout.  

The difference is ultimately academic, but IMO scientific accuracy/validity is still important for its own sake.  The evidence that inefficiency inducing mechanisms occur for all sustained intensity above the first lactate threshold (ie: an intensity well below FTP), is indisputable.   

[Peter Riegersperger]

On Saturday, July 25, 2015 at 9:17:59 AM UTC+2, Nathan Townsend wrote:

On Friday, 24 July 2015 21:05:43 UTC+3, Peter Riegersperger wrote:

This is my first season of training with a power meter, and I’m just getting the hang of it and try to understand all the data I’m collecting over the months. What puzzles me is that I have a big difference between my tested FTP and CP/FTP computed by the model.

My FTP apparently is 249W. This is the result of the „95% of 20 min time trial effort“ test suggested by Allen/Coggan, and also a real life performance (climbing Alpe d’Huez in 01:10:31 with an XPower of 249W).

Can you break your Alpe d'Huez ride into 10min segments, calculate average power for each segment, and post it up here (or alternatively just post a screen shot of the power for the entire effort (smooth power to 30 s moving average for clarity).

I don't know how to do that, so I asked GC to calculate intervals based on time:

The first interval is too low because the climb didn't start at exactly 10 min into the ride. Here's also the power plot for the complete segment, and a second version where I removed the final "sprint" for easier reading.

 

 

Secondly what is the average power (not Xpower which is a manipulation of the raw data) for the entire effort?

248W.

So using your data given above, your 20min best effort was about 263W, which is not far away from the GC estimate of 267W.  So in reality there is about a 15W difference between the CP estimate and FTP estimate.  From a practical perspective, to be honest I don't really think this difference makes a big deal because most people set their interval intensities well beyond FTP + 15W and therefore they're working above CP and in the severe intensity domain anyway ie: an interval session where you set the power at FTP + 65W is basically the same workout as CP + 50W.  The important thing is the actual power you're pushing and how long you hold it for, not how you derived that number in the first place. 

Thank you for the clarification and explanation. This makes sense to me and reflects what I experience when on the bike.

 

However if you wanted to do long intervals just above threshold, and therefore in the severe intensity domain which will push VO2 up close to max upon nearing the end of the interval, then if you did these intervals around 255-260W, you just won't get there (ie: near VO2max).  However if you do them at about 270-275W then my bet is that this intensity will definitely not feel sustainable, and hence towards the end of a longer 12-15min interval perception of effort will resemble that which occurs during a much shorter but harder effort eg: 5min. However a 15min interval at 255-260w, despite the fact it is supposed to be above threshold according to Coggan, will "feel" as though it is sustainable and you can continue for longer. 

That is exactly what I am experiencing and where the initial question stems from. Should I base my training and training zones on what the model says (and the 20 min test, if I don't substract the 5%), or on what I can sustain for 1 hour? As you say, the difference in intensity is quite big, and I'm unsure if I under- or overtrain when I'm using this or that value.

For example, if I'm doing VO2max-Intervals (as I did yesterday) at 110-115% of FTP (as prescribed by Allen/Coggan in their book) and use 249W as FTP, I work in the range 274-286W; But when I use the model calculation for CP, the work interval range is 294-307W (estimated FTP by model roughly the same). For me, this is a huge difference (and honestly I don't know if I can hold that power for the prescribed duration).

[Nathan Townsend]

Hi Peter,

This is perfect. we can see straight away just from eyeballing the slight fatigue effect I'm referring to.  See comments below.....

On Saturday, 25 July 2015 14:43:07 UTC+3, Peter Riegersperger wrote:

The first interval is too low because the climb didn't start at exactly 10 min into the ride. Here's also the power plot for the complete segment, and a second version where I removed the final "sprint" for easier reading.

 

 

Secondly what is the average power (not Xpower which is a manipulation of the raw data) for the entire effort?

248W.

Just from eyeballing we can see that between approx 15min and 40min (=25min total) the average power is around 255W maybe even somewhere inbetween 255-260W (clearly you spend more time above 250 than below). I suspect that during this period your perception of effort was that you felt as though you were at threshold, not above?  Then you drop below 250 and average 240-245 or thereabouts.  So if you are supposed to be able to maintain threshold for about 60min, then you should be able to maintain this power.   

Nb: I notice from Mark's post about WKO4 the duration of FTP appears to have been downgraded to about 40min which thus shows us that without admitting it openly, Andy Coggan's new model has shifted towards what the scientific literature has been pointing us towards for a couple of decades... which is something I've been trying to tell him for several years.   FTP has also commonly been defined as 40km average power, but does anyone know anyone that can do 40km in 40min?

That is exactly what I am experiencing and where the initial question stems from. Should I base my training and training zones on what the model says (and the 20 min test, if I don't substract the 5%), or on what I can sustain for 1 hour? As you say, the difference in intensity is quite big, and I'm unsure if I under- or overtrain when I'm using this or that value.

For example, if I'm doing VO2max-Intervals (as I did yesterday) at 110-115% of FTP (as prescribed by Allen/Coggan in their book) and use 249W as FTP, I work in the range 274-286W; But when I use the model calculation for CP, the work interval range is 294-307W (estimated FTP by model roughly the same). For me, this is a huge difference (and honestly I don't know if I can hold that power for the prescribed duration).

The mere fact that Allen/Coggan recommend 110-115% FTP basically proves what the CP model predicts ie: that you need to exercise above CP in order to hit VO2max.   If FTP is really threshold, then why do they recommend at least 110% FTP?  You should be able to attain VO2max during an effort at 105% FTP because anything above threshold is non-steady state and this pushes VO2 upwards to max, but of course since FTP is below a true threshold, then at 105% FTP this won't happen.  That leaves a 5% intensity zone gap which Coggan claims is above threshold, yet does not induce physiological responses that resemble those which really are in the severe intensity domain.  

When using CP as your threshold anchor point, it is possible to conduct VO2max intervals anywhere between 105% to about 150% CP and you'll be able to hit VO2max.  In fact, it is entirely possible to do "VO2max" intervals using even higher power eg: all-out 10sec sprint efforts, as long as the recovery duration is short enough. 

From your data and everything you've said, it looks as though your threshold zone is 260-270W and therefore you can do "VO2max intervals" at ANY power output above this, and they will only cease to become VO2max intervals if the interval duration is not long enough, or recovery duration is too short.  

[Peter Riegersperger]

Hej Nathan!

Just from eyeballing we can see that between approx 15min and 40min (=25min total) the average power is around 255W maybe even somewhere inbetween 255-260W (clearly you spend more time above 250 than below). I suspect that during this period your perception of effort was that you felt as though you were at threshold, not above?  Then you drop below 250 and average 240-245 or thereabouts.  So if you are supposed to be able to maintain threshold for about 60min, then you should be able to maintain this power.

That is correct - I averaged 257W from 15min to 40 min. From there to the point before the last push, I averaged 241. And it is true - I started out with what I felt was a sustainable pace (considering that I had 1h of climbing in front of me) and then slowly faded. (I blamed it on the heat, it was 40° on that day)

The mere fact that Allen/Coggan recommend 110-115% FTP basically proves what the CP model predicts ie: that you need to exercise above CP in order to hit VO2max.   If FTP is really threshold, then why do they recommend at least 110% FTP?  You should be able to attain VO2max during an effort at 105% FTP because anything above threshold is non-steady state and this pushes VO2 upwards to max, but of course since FTP is below a true threshold, then at 105% FTP this won't happen.  That leaves a 5% intensity zone gap which Coggan claims is above threshold, yet does not induce physiological responses that resemble those which really are in the severe intensity domain.  

When using CP as your threshold anchor point, it is possible to conduct VO2max intervals anywhere between 105% to about 150% CP and you'll be able to hit VO2max.  In fact, it is entirely possible to do "VO2max" intervals using even higher power eg: all-out 10sec sprint efforts, as long as the recovery duration is short enough. 

From your data and everything you've said, it looks as though your threshold zone is 260-270W and therefore you can do "VO2max intervals" at ANY power output above this, and they will only cease to become VO2max intervals if the interval duration is not long enough, or recovery duration is too short.  

That sounds more logical than adding a few watts to be above threshold „for good measure“ as the FTP-Model does.

Thank you for the explanation! 

all the best,

peter

pe...@lerouleurlent.net

~ Slow Since 2005 ~
http://lerouleurlent.net/

twitter: @lifeintheslow

instagram: lifeintheslow

[Nathan Townsend]

On Saturday, 25 July 2015 20:14:50 UTC+3, Peter Riegersperger wrote:

, or recovery duration is too short.  

Sorry I made a blunder here, this should say "recovery duration too LONG".

From here it should not be a great leap to comprehend the value of the work-balance model.  As long as you get CP + W' about right, then you can play around with an INFINITE number of permutations of interval power, duration, recovery power, recovery duration, sets, reps.  Low values of of W'bal (eg: <25%) will generally correspond with high values of VO2 (I don't know yet, but a guess would be >95% VO2max).  And therefore any interval workout that accumulates a lot of time at low values of W'bal could be considered a "VO2max workout".  So the idea that you have to stay within a narrow power band to be doing VO2max training is going to become outdated within the next 5-10yrs (it already is within the scientific literature).

Using the same concept, it should also be apparent that if you want to be doing a workout which specifically targets the anaerobic systems, then you need to adopt a strategy whereby fatigue occurs well before W'bal reaches a low value... eg: 50%, and then you need to allow enough recovery so that on subsequent efforts, it doesn't progressively get too low (eg: <25%).   

Future research should hopefully remove some of my guesswork here and replace those figures with objective data. 

[Peter Riegersperger]

So, if I understand you correctly, the difference between the Allen/Coggan FTP-Value and the Golden Cheetah FTP-Value can be ignored while training because even if the GC-value is higher, Allen/Coggan added some sort of „padding“, which results in the same (or nearly the same) training zones?

For example, A/C FTP is 249, and VO2Max zone would be 110%-115% -> 274 - 286W [1]

You suggest, on the other hand, that if i were to use the GC FTP at 262, VO2Max would be ~105%-110% -> 275 - 288W which is practically the same.

And another way to look at it would be simply going over CP, which is 270, and again, both computed zones fit into this nicely.

I find the concept of W’bal intriguing and intuitively plausible, but have to admit that I’m still struggling with the „basics“ (e.g. getting values as threshold right, and aligning the new power data with heartrate and perceived exhaustion and my general feeling on the bike)

p

[1] They use this range in their workout VO2-W2, other workouts use other values.

--
_______________________________________________
Golden-Cheetah-Users mailing list
golden-che...@googlegroups.com
http://groups.google.com/group/golden-cheetah-users?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "golden-cheetah-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/golden-cheetah-users/lmwEVcGrO1k/unsubscribe.
To unsubscribe from this group and all its topics, send an email to golden-cheetah-u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Nathan Townsend]

On Sunday, 26 July 2015 10:49:28 UTC+3, Peter Riegersperger wrote:

So, if I understand you correctly, the difference between the Allen/Coggan FTP-Value and the Golden Cheetah FTP-Value can be ignored while training because even if the GC-value is higher, Allen/Coggan added some sort of „padding“, which results in the same (or nearly the same) training zones?

For example, A/C FTP is 249, and VO2Max zone would be 110%-115% -> 274 - 286W [1]

You suggest, on the other hand, that if i were to use the GC FTP at 262, VO2Max would be ~105%-110% -> 275 - 288W which is practically the same.

Nail.On.The.Head

The lower intensity boundary zone for "VO2max training" should always be just above threshold eg: approx 10-15W or 5% of CP, whatever the lower value.  Therefore, the requirement of adding 10% is basically an open confession that FTP doesn't adequently represent the boundary between the moderate and severe intensity domains ie: it is a good 5% too low.

The upper intensity boundary for VO2max training is there is no upper bound. It is essentially your peak 3 sec power.  We know that all out sprint interval training is a powerful stimulus to increase VO2max if done appropriately as stated ie: do enough reps and keep the recovery duration short.  

Effects of sprint interval training on VO2max and aerobic exercise performance: A systematic review and meta-analysis

http://www.ncbi.nlm.nih.gov/pubmed/23889316

[Armando Mastracci]

Does failure occur at intensities just below CP at the ~20-30min mark?  I.e. is there a possibility that the athlete can continue along without failure at a lower work-rate?

One can speculate that a "steady-state" work-rate decreases (oxymoron aside for the time being) with decreasing CHO availability with longer workout durations and the lower availability gradient ultimately decreases metabolite turnover in the blood.  Perhaps the maximum work-rate for 20 or 40 or 60 minutes is simply a matter of how long a constant, quasi steady-state work-rate can be sustained before we see an unsustainable increase in metabolites?  Maybe the length of time one sees a true steady-state is a function of starting CHO in the system together with the training status of the athlete?

In other words, is the percentage of CP/FTP to use for training dependent on how long and intensely into the workout one has been working and how well one has been at replenishing CHO?

[Nathan Townsend]

On Monday, 27 July 2015 00:54:43 UTC+3, Armando wrote:

Does failure occur at intensities just below CP at the ~20-30min mark?  I.e. is there a possibility that the athlete can continue along without failure at a lower work-rate?

For some people, surely it does. In the de Lucas paper the mean TTe @ CP was 22.9min +/- 7.5min.  So 20-30min is actually a little bit liberal according to that study, since by definition 68% of cases fall within 1 SD, so that means that 32/2 = 16% of cases fell below 15.5min!

 

One can speculate that a "steady-state" work-rate decreases (oxymoron aside for the time being) with decreasing CHO availability with longer workout durations and the lower availability gradient ultimately decreases metabolite turnover in the blood.  Perhaps the maximum work-rate for 20 or 40 or 60 minutes is simply a matter of how long a constant, quasi steady-state work-rate can be sustained before we see an unsustainable increase in metabolites?  Maybe the length of time one sees a true steady-state is a function of starting CHO in the system together with the training status of the athlete?

In other words, is the percentage of CP/FTP to use for training dependent on how long and intensely into the workout one has been working and how well one has been at replenishing CHO?

Yes, but fatigue is multifactorial so CHO availability would be just one of a broader set of mechanisms that might reduce the duration one can sustain at CP. 

Food for thought....

Phosphocreatine degradation in type I and type II muscle fibres during submaximal exercise in man: effect of carbohydrate ingestion

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2278942/

 

[Armando Mastracci]

On 2015-07-27, at 1:40 AM, Nathan Townsend <nathant...@gmail.com> wrote:

On Monday, 27 July 2015 00:54:43 UTC+3, Armando wrote:

Does failure occur at intensities just below CP at the ~20-30min mark?  I.e. is there a possibility that the athlete can continue along without failure at a lower work-rate?

For some people, surely it does. In the de Lucas paper the mean TTe @ CP was 22.9min +/- 7.5min.  So 20-30min is actually a little bit liberal according to that study, since by definition 68% of cases fall within 1 SD, so that means that 32/2 = 16% of cases fell below 15.5min!

Perhaps I didn't explain the point correctly.  If we decrease power over time, that would be sustainable far longer than 20-30 min.  So rather than maintain a fixed power over the duration, introduce a gradual decrease over time and determine the minimal rate of decrease that would be sustainable ad infinitum, virtually.   What I'm suggesting is that maximum steady-state *decreases* with fatigue (again, apologies for the oxymoron).  In fact, it has to, as anyone who has ever bonked can attest.  CP in that state is considerably lower than CP with full glycogen availability. 

 

One can speculate that a "steady-state" work-rate decreases (oxymoron aside for the time being) with decreasing CHO availability with longer workout durations and the lower availability gradient ultimately decreases metabolite turnover in the blood.  Perhaps the maximum work-rate for 20 or 40 or 60 minutes is simply a matter of how long a constant, quasi steady-state work-rate can be sustained before we see an unsustainable increase in metabolites?  Maybe the length of time one sees a true steady-state is a function of starting CHO in the system together with the training status of the athlete?

In other words, is the percentage of CP/FTP to use for training dependent on how long and intensely into the workout one has been working and how well one has been at replenishing CHO?

Yes, but fatigue is multifactorial so CHO availability would be just one of a broader set of mechanisms that might reduce the duration one can sustain at CP. 

For sure, but it plays a starring role in fatigue as durations increase.  

Food for thought....

Phosphocreatine degradation in type I and type II muscle fibres during submaximal exercise in man: effect of carbohydrate ingestion

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2278942/

The effect of glycogen depletion on the curvature constant parameter of the power-duration curve for cycle ergometry.

http://www.ncbi.nlm.nih.gov/pubmed/10661696 

Although they claim a non statistically-significant reduction in CP, the reduction seen warrants further research.

[mike veloclinic]

I agree with Armando in the value of considering CP as a threshold phenomena that is dynamic and may change as a function of substrate availability and fatigue.

http://veloclinic.tumblr.com/post/125144792393/example-of-modelling-dynamic-critical-power

as an example use

[Mark Liversedge]

I do hope that dynamical models start to appear. CP and W' also fluctuate with training load (up/down stress/recover/adapt) too. We are looking at the future here people :)

CP in a dynamic model starting at MLSS power and falling to ~40km TT power after 50 mins.

Ha!

Mark 

[Armando Mastracci]

As a dynamic model, one could look at the change in CP as a function of initial glycogen, work performed above LT1 and CHO ingestion/absorption. It's not really equivalent duration-wise to W' as far as capacity and expenditure/reconstitution so a bit trickier to model.  We don't know where we're starting at and need a way to track how much and when we're eating as inputs to the model.

[Peter Riegersperger]

Sorry to come back to this - what is the practical recommendation for using both the Allen/Coggan training book and W’bal?

Setting CP to the Golden Cheetah model value (2 parameter model) for W’bal, and using an Excel sheet for A/C compatible training zones to be able to use their workout recommendations?

At the moment I find it hard to consolidate both views on my workouts. Setting everything according to A/C makes the power zones fit, but W’bal gets negative on practically every VO2max work interval, and when using the model value for CP all the power zones are skewed and the workouts appear to be too easy, which impacts TSS computation.

p

[Mark Liversedge]

On Sunday, 2 August 2015 10:35:08 UTC+1, Peter Riegersperger wrote:

Sorry to come back to this - what is the practical recommendation for using both the Allen/Coggan training book and W’bal?

Setting CP to the Golden Cheetah model value (2 parameter model) for W’bal, and using an Excel sheet for A/C compatible training zones to be able to use their workout recommendations?

At the moment I find it hard to consolidate both views on my workouts. Setting everything according to A/C makes the power zones fit, but W’bal gets negative on practically every VO2max work interval, and when using the model value for CP all the power zones are skewed and the workouts appear to be too easy, which impacts TSS computation.

That's why you can change the zone default percentages on the Default tab in Power Zones :)

We should add a drop down for CP vs FTP.

Mark 

[Peter Riegersperger]

On 02.08.2015, at 11:44, Mark Liversedge <liver...@gmail.com> wrote:

Setting everything according to A/C makes the power zones fit, but W’bal gets negative on practically every VO2max work interval, and when using the model value for CP all the power zones are skewed and the workouts appear to be too easy, which impacts TSS computation.

That's why you can change the zone default percentages on the Default tab in Power Zones :)

We should add a drop down for CP vs FTP.

Got it, thanks! (Actually, I modified the Zones on the „Critical Power“ tab, but this should have the same effect I assume)

[Peter Riegersperger]

But doing this will still lead to an underestimation of training stress, no?

With TSS = (sec x NP x IF)/(FTP x 3600) x 100 , and setting FTP to CP, TSS will be underreported.

Or am I missing something?

peter

On 02.08.2015, at 11:44, Mark Liversedge <liver...@gmail.com> wrote:

--
_______________________________________________
Golden-Cheetah-Users mailing list
golden-che...@googlegroups.com
http://groups.google.com/group/golden-cheetah-users?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "golden-cheetah-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/golden-cheetah-users/lmwEVcGrO1k/unsubscribe.
To unsubscribe from this group and all its topics, send an email to golden-cheetah-u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Neil Pugh]

From my understanding you can't set your CP to your 20 min max average power and then hope to implement the WKO way of scoring things cos as you say it just underestimates the Coggan TSS value.

In other words, you can't be using GC as if it was WKO, you just have to implement one way of looking at things and stick with it.

On Monday, August 3, 2015 at 1:02:07 PM UTC+1, Peter Riegersperger wrote:

But doing this will still lead to an underestimation of training stress, no?

With TSS = (sec x NP x IF)/(FTP x 3600) x 100 , and setting FTP to CP, TSS will be underreported.

Or am I missing something?

peter

On 02.08.2015, at 11:44, Mark Liversedge <liver...@gmail.com> wrote:

On Sunday, 2 August 2015 10:35:08 UTC+1, Peter Riegersperger wrote:

Sorry to come back to this - what is the practical recommendation for using both the Allen/Coggan training book and W’bal?

Setting CP to the Golden Cheetah model value (2 parameter model) for W’bal, and using an Excel sheet for A/C compatible training zones to be able to use their workout recommendations?

At the moment I find it hard to consolidate both views on my workouts. Setting everything according to A/C makes the power zones fit, but W’bal gets negative on practically every VO2max work interval, and when using the model value for CP all the power zones are skewed and the workouts appear to be too easy, which impacts TSS computation.

That's why you can change the zone default percentages on the Default tab in Power Zones :)

We should add a drop down for CP vs FTP.

Mark 

--
_______________________________________________
Golden-Cheetah-Users mailing list
golden-che...@googlegroups.com
http://groups.google.com/group/golden-cheetah-users?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "golden-cheetah-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/golden-cheetah-users/lmwEVcGrO1k/unsubscribe.

To unsubscribe from this group and all its topics, send an email to golden-cheetah-users+unsub...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Nathan Townsend]

On Monday, 3 August 2015 15:02:07 UTC+3, Peter Riegersperger wrote:

But doing this will still lead to an underestimation of training stress, no?

With TSS = (sec x NP x IF)/(FTP x 3600) x 100 , and setting FTP to CP, TSS will be underreported.

Or am I missing something?

peter

Hi Peter, the TSS has arbitrary units so the values obtained cannot over or underreport training load.  What the formula does is assign 100 units to a training "dose" which is equivalent to a 60 min max effort ie: 60MMP. The purpose of the FTP in the formula is to create an "anchor" point which denotes your individual fitness level.  Therefore all of your training is expressed relative to that anchor point (this is what IF does in the formula).  If you alter any constants in the formula (including FTP) then you just re-assign a different value to a particular "dose" of training. Since the units are arbitrary this doesn't matter, there is no particular reason why we need to use a value of 100 units as the anchor point.  What is important is the pattern of loading/unloading over a period of time.  

I don't think there is anything wrong with this formula. It has some limitations in terms of describing training intensity distribution, but where the fundamental issue arises is in how people "test" their FTP.  Most people do not actually conduct a 60min TT.  They do a 20min TT and then apply a correction factor to estimate FTP.  What this means is that 60MMP must always be 95% of 20MMP within and individual.  If FTP increases but 20MMP does not, or visa versa, then you will introduce a greater level of uncertainty in your FTP estimate.  Therefore if you use this estimate in your TSS formula, then you introduce greater uncertainty into your TSS calculations.

As a sport scientist with >15yrs experience conducting performance testing on athletes and research participants (I was also on a sport science laboratory standards committee for several years in Australia), it just seems totally illogical to take a perfectly good raw value take from a real performance test ie: your 20MMP, but rather than using this value as your performance level "anchor point" you estimate some other point on the power-duration curve, and use that other point instead of the raw data.  The AIS has been using 30min TTs for example going back nearly 20yrs on elite cyclists.  You do not NEED a 60min TT for the purpose of establishing a performance index. 20MMP will do just fine. If you want to assign a value of 100 to a 20MMP, then you change the constant 3600 (seconds in 60min) to 1200.  If you really wanted, you could even make a 20MMP equal to say 666 arbitrary units, in that case you would also change the constant 100 to 666.   Since these values are constants, then they make no impact on the pattern of loading/unloading over time. 

If 20MMP = 320 and FTP = 300 then a 60MMP effort becomes 88 arbitrary units instead of 100.   Americans use 1609 m as a standard distance unit whereas the rest of the world uses 1000 m, so when I ride with my friend from the US, I travel 100 distance units but he only travels 62 units on the same ride! da fuq??  Doesn't that mean Americans underestimate training volume? 

As long as you understand the meaning, then it makes no difference ;-)

Lastly, it should be apparent that using a TT of any duration is better than using CP, because there is a greater level of uncertainty in CP estimates generally than there is for TT tests.

[Peter Riegersperger]

Nathan,

I understand that the formula (as most others I have encountered in this short season with a power meter) generate arbitrary numbers that make sense either relatively or when put in context.

The problem I as a novice have is that there is a ton of literature and articles out there that is working with FTP and not CP, and basing recommendations on this.

(For example - and this article actually raised my question about FTP/CP and the PMC of Golden Cheetah - http://home.trainingpeaks.com/blog/article/managing-your-training-stress-balance )

I don’t have the educational background to judge these articles, so I tend to start working with their recommendations (as I do with yours, by the way ;).

I know that i can translate between TSS derived from FTP and CP, and I can adapt the recommendations to fit my own model, but it would be a huge timesaver to not have to do that.

Lastly, it should be apparent that using a TT of any duration is better than using CP, because there is a greater level of uncertainty in CP estimates generally than there is for TT tests.

So, what would be a sensible value to use in your experience? 30min best?

Thank you for your insights,

peter

--
_______________________________________________
Golden-Cheetah-Users mailing list
golden-che...@googlegroups.com
http://groups.google.com/group/golden-cheetah-users?hl=en
---
You received this message because you are subscribed to a topic in the Google Groups "golden-cheetah-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/golden-cheetah-users/lmwEVcGrO1k/unsubscribe.
To unsubscribe from this group and all its topics, send an email to golden-cheetah-u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Nathan Townsend]

On Tuesday, 4 August 2015 11:36:23 UTC+3, Peter Riegersperger wrote:

Nathan,

I understand that the formula (as most others I have encountered in this short season with a power meter) generate arbitrary numbers that make sense either relatively or when put in context.

The problem I as a novice have is that there is a ton of literature and articles out there that is working with FTP and not CP, and basing recommendations on this.

(For example - and this article actually raised my question about FTP/CP and the PMC of Golden Cheetah - http://home.trainingpeaks.com/blog/article/managing-your-training-stress-balance )

I don’t have the educational background to judge these articles, so I tend to start working with their recommendations (as I do with yours, by the way ;).

I know that i can translate between TSS derived from FTP and CP, and I can adapt the recommendations to fit my own model, but it would be a huge timesaver to not have to do that.

Lastly, it should be apparent that using a TT of any duration is better than using CP, because there is a greater level of uncertainty in CP estimates generally than there is for TT tests.

So, what would be a sensible value to use in your experience? 30min best?

Thank you for your insights,

peter

The principles in that article regarding CTL are generally sound, but you can basically ignore the absolute values quoted for TSB and learn what works best for you. I think a 20MMP is a great choice of performance test because it is practical to conduct and very reliable.  One of the reasons that I suspect the training peaks philosophy has this FTP centric view of the universe is because 40km is a popular distance in the USA, but in the UK the most popular distance is a 10 mile TT which is close to 20-25min for a lot of cyclists.

 If you simply replace a 60MMP with a 20MMP in the TSS calculation then it will make little difference overall to TSB values, whereas absolute values for ATL and CTL will be slightly lower than had you used 60MMP in the formula.  Note that in the article it talks in general terms anyway... eg: CTL should rise to a peak, then decline etc etc.  these recommendations stay the same regardless of whether 60MMP or 20MMP are used. 

I know a few coaches who uses TP in this manner (ie: just enters the actual 20MMP test result instead of estimating FTP).  I designed an altitude training program for one of these coaches and his cyclist in the lead up to the Aus masters track championships. Results are below.....  

World

Men 35-39

3000 metres

Michael Gallager

AUS

3.20.940

14/03/2014

http://www.cycling.org.au/Track/Masters-Best-Times

Disclaimer: No FTP estimates were (ab)used in the breaking of this masters all comers world record.

[Ale Martinez]

El martes, 4 de agosto de 2015, 11:24:58 (UTC-3), Nathan Townsend escribió:

 If you simply replace a 60MMP with a 20MMP in the TSS calculation then it will make little difference overall to TSB values, whereas absolute values for ATL and CTL will be slightly lower than had you used 60MMP in the formula.

If FTP/20MMP~0.95 TSS/CTL/ATL/TSB based on 20MMP will be ~10% lower than those based on FTP, due to the quadratic relationship between TSS and IF.