W bal, what is it?

13,954 views
Skip to first unread message

Julian

unread,
Feb 16, 2014, 8:10:39 AM2/16/14
to golden-che...@googlegroups.com
Forgive me if I missed a conversation- but I have no idea what W' balance is and how to apply it- would someone mind filling me in or pointing me to some literature as searching the internet for "w' balance" is kind of useless.

Thanks in advance.

Mark Liversedge

unread,
Feb 16, 2014, 8:16:19 AM2/16/14
to golden-che...@googlegroups.com
Yeah, google doesn't like punctuation marks!!

This video is my favourite for explaining W', CP and W' bal.

Mark

Jon Lidster

unread,
Feb 16, 2014, 4:44:40 PM2/16/14
to golden-che...@googlegroups.com
Hi Mark

Just had a quick look at the Phil Skib vid you linked.

What recovery rate constant does GC apply to W'bal? Does it apply the same halflife at all powers below critical power, or does the halflife change as a fraction of power output below Critical power?

Interested to know if there is some literature on experimentally determined recovery rates as a function of work rate below critical power. Is it related to rates of blood lactate removal?

Just wondering how accurate the W' replenishment calculation is likely to be.

Regards

Jon 

Mark Liversedge

unread,
Feb 16, 2014, 5:10:05 PM2/16/14
to golden-che...@googlegroups.com
On Sunday, 16 February 2014 21:44:40 UTC, Jon Lidster wrote:
What recovery rate constant does GC apply to W'bal? Does it apply the same halflife at all powers below critical power, or does the halflife change as a fraction of power output below Critical power?


There is quite a lot of analysis about the recharge rate, because like you say, should it recharge more slowly is we are 'close' to CP but not over?

The data analysis in the paper 'suggesting a relationship between Tau and CP *irrespective* of sub-CP intensity domain' and then 'Above CP, Tau increased to non-physiological values, indicating no net recharge of the W' and merely a slightly lower rate of depletion during the recovery interval'.

So basically, there was no need to adjust based upon how far below CP the power drops. The paper goes on to present this formula (that we use) to determine tau;


If I have got this wrong, I am sure Dr Skiba will correct me :)

Mark

Dafydd Williams

unread,
Feb 17, 2014, 9:11:31 PM2/17/14
to golden-che...@googlegroups.com
There is something I am not understanding here.

Are you saying that the rate at which W'bal is replenished is not affected by the recovery power output at all, or just that the same formula applies regardless of how low below CP the recovery power goes?

As I understand it, the DCP in the formula you quoted above represents the difference between CP and the recovery power output, so it should have an effect on the rate of W'bal recovery, but on my GC ride graphs the rate of recovery appears to be unaffected by my power output during recovery periods, or the difference is so small it can't be seen. That seems to go against what is said in Dr. Skiba's video you linked above, and what I have read elsewhere - where I would expect to see W'bal increase more slowly as my recovery power increased and approached CP.

I don't have access to the full text of a lot of the papers, but I did find this http://www.humankinetics.com/acucustom/sitename/Documents/DocumentItem/03Noordhof%2020130282_475-482_ej.pdf which quotes almost the same formula for tau (see attached image) but it has -0.1DCP . I'm not sure which version is correct!
Untitled.jpg

Dr. Philip Skiba

unread,
Feb 17, 2014, 9:39:34 PM2/17/14
to golden-che...@googlegroups.com
Let me see if I can help.

To make the calculation easier with field data, a single time constant is used for the entire file. It is calculated by taking the average of all the samples in the file below CP. This probably works because average power is a reasonable proxy for the average oxidative state of the muscle.

The paper we just published in IJSPP demonstrated the statistical validity of the technique. This is not to say that the procedure could not be improved. I've got lots of ideas. But it works very well as is. I'll paste in my lay description of the study below.

The reference is here:

http://www.ncbi.nlm.nih.gov/pubmed/24509723

The purpose of the study was to see if the W'Bal model was valuable as a field measure. In other words, do people in real-life become exhausted at the time W'Bal = 0, or at some other point?

To check this out, we collected power meter files from athletes who became prematurely exhausted during training or racing. We also collected files from athletes who had done hard training or racing, but had not "blown up". We first checked to see if there was a significant difference between the minimum W'Bal reported in each group (there was). On average, the athletes blew up at 0.5 ± 1.3 kJ (95% CI = 0 – 0.9 kJ). The SE was 1.0 kJ. We then used ROC analysis to determine how good the model was. This is more typically used in assessing medical tests, but fit our purpose nicely. The model fit the criteria for an excellent diagnostic test. On the basis of the present data, we can achieve 95% sensitivity w/ 24% ‘false positives’ if we say the athlete is at risk of blowing up if they go below 2.5 kJ. If the threshold is set at W′BAL = 1.5 kJ, 80% of athletes will be appropriately classified as exhausted and 88% appropriately classified as non-exhausted.

Anecdotally, athletes have reported feeling really bad if they let the W′BAL drop < 1.5 kJ. Thus, we feel it is both statistically and practically defensible to discourage athletes from proceeding below a W′BAL of 1.5 kJ (i.e. less than 10% of the W′ for most athletes) if they wish to avoid premature exhaustion.

So, there you have it. Hopefully, this makes you a bit more confident using the model.

I hope to have another manuscript for y'all in the next month or two...juggling writing papers with writing my thesis at the moment!

Cheers,

Phil

Lozy

unread,
Feb 18, 2014, 11:43:15 AM2/18/14
to golden-che...@googlegroups.com
Where does time below CP play into W'? By that I mean if I do a 5 hour ride, I am going to have less in the tank than I would after 2 hours. So surly my W' begins to decrease at some point. Or am I missing something, I am sure I am.

Dr. Philip Skiba

unread,
Feb 18, 2014, 12:00:59 PM2/18/14
to Lozy, golden-cheetah-users@googlegroups.com Mailing List
That’s a very good question. It doesn’t seem to make much difference during rides of a few hours in reasonably fit people, or longer in highly fit people. I can’t say much more than that until a few more publications are in press.

Phil

--

Dr. Philip Skiba
PhysFarm Training Systems, LLC


On Feb 18, 2014, at 10:43 AM, Lozy <lozon.j...@gmail.com> wrote:

> Where does time below CP play into W'? By that I mean if I do a 5 hour ride, I am going to have less in the tank than I would after 2 hours. So surly my W' begins to decrease at some point. Or am I missing something, I am sure I am.
>
> --
> --
> _______________________________________________
> 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 the Google Groups "golden-cheetah-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to golden-cheetah-u...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

Jon Lidster

unread,
Feb 18, 2014, 1:04:56 PM2/18/14
to golden-che...@googlegroups.com
I've heard it mentioned many times fatigue is multi-factorial. I'm guessing over longer workouts other fatigue mechanisms play more dominant roles like muscle fibre damage, neuromuscular fatigue, prolonged oxidative stress/ or hypoxic stress, prolonged cellular pH affects, phychological factors, hydration, depletion of other essential bodily chemistries and I also presume muscle and liver glycogen stores would drop steadily even when working below CP, I am simply assuming glycogen utilisation will be slightly quicker than replenishment even when constantly feeding. Can the body perform to CP on fat metabolism alone?

Anyway I speculate a lot but guess fatigue over shorter times relates more to glycogen depletion above CP plus to a minor extent other fatigue mechanisms, at long distances glycogen levels play less of a role and other mechanisms dominate.

Also would CP be overestimated slightly over times such as 5h?

Looking forward Dr Phils papers on this. You doing another PhD Dr Phil? Please slap me if my guessing is totally innaccurate. I'm not a sports scientist, biochemist or molecular biologist, just a recent enthusiast.


On Sunday, February 16, 2014 1:10:39 PM UTC, Julian wrote:

Dafydd Williams

unread,
Feb 18, 2014, 4:08:02 PM2/18/14
to golden-che...@googlegroups.com
Thanks for the explanation, I've got it now :)

Dr. Philip Skiba

unread,
Feb 19, 2014, 10:47:14 AM2/19/14
to golden-che...@googlegroups.com


On Tuesday, 18 February 2014 12:04:56 UTC-6, Jon Lidster wrote:
I've heard it mentioned many times fatigue is multi-factorial. I'm guessing over longer workouts other fatigue mechanisms play more dominant roles like muscle fibre damage, neuromuscular fatigue, prolonged oxidative stress/ or hypoxic stress, prolonged cellular pH affects, phychological factors, hydration, depletion of other essential bodily chemistries and I also presume muscle and liver glycogen stores would drop steadily even when working below CP, I am simply assuming glycogen utilisation will be slightly quicker than replenishment even when constantly feeding.


You are exactly right, fatigue is always multifactorial. It has been hypothesized that one of the differences between fatigue above CP (in the 'severe domain'), and fatigue in the heavy heavy domain (below CP but above LT) might have to do more with high energy phosphate depletion / metabolite accumulation in the former, and perhaps more glycogen depletion in the latter. IOW, above CP, you can see creatine phosphate and ATP decrease, and inorganic phostphate, hydrogen ion concentration, increase, to what might (or might not) be 'limiting values'. You have what may be construed as an energy crisis. In my (and others) opinion, you probably can't hang in the severe domain long enough such that you can deplete glycogen enough to actually force the cessation of exercise. This is by no means settled science, however.

Fatigue in the heavy domain is complicated, and there are various ways of trying to model it, but it will be far harder to apportion 'blame' for fatigue among all of the various things that are going on.

 
Can the body perform to CP on fat metabolism alone?

No, you really can't do much of anything on fat metabolism alone. You are always burning some mixture of fat and carbohydrate.
 


Also would CP be overestimated slightly over times such as 5h?

Not sure what you mean by that...could you rephrase the question?

 

Looking forward Dr Phils papers on this. You doing another PhD Dr Phil?


I'm actually a sports physician by training. This will be my first PhD. :-)  A few years ago, I realized that no one had the answers to the kinds of questions I was asking, so I took a sabbatical with Andy Jones and Anni Vanhatalo at the University of Exeter (UK) since they were the ones working on it and are really the world experts in the field.

 
Please slap me if my guessing is totally innaccurate. I'm not a sports scientist, biochemist or molecular biologist, just a recent enthusiast.
 

No slapping necessary. Not my style. And you mostly have the right idea in terms of the big picture!

Phil

Jon Lidster

unread,
Feb 20, 2014, 1:34:12 PM2/20/14
to golden-che...@googlegroups.com
Hi Dr Phil and thanks for the nice reply.

Sorry my poorly phrased question is perhaps down to my lack of total understanding of the definition of CP. I guess what I was trying to get at is that 5h power will be lower than CP as calculated by certain models and if approximating with CP60, since I notice the actual power profile continues to drop over prolonged periods rather than go horizontal. If that makes sense. 

Interesting to hear your explanation. Now you mention the phosphate groups I did notice an article title authored by yourself with 31P-MRI studies so I guess I need to have a read of that.

Hope you enjoyed England, though I'm not sure Exeter would be most peoples choice of destination ;) not much there other than excellent scientists :-)

Good luck with the thesis and viva voce, it took me 6 years to write mine (I got there in the end), but I can tell you're a much more motivated guy than me so that won't be an issue :)

All the very best

Jon

Dr. Philip Skiba

unread,
Feb 20, 2014, 4:27:54 PM2/20/14
to golden-che...@googlegroups.com
Hi Jon,

Please see below...


On Thursday, 20 February 2014 12:34:12 UTC-6, Jon Lidster wrote:
Hi Dr Phil and thanks for the nice reply.

Sorry my poorly phrased question is perhaps down to my lack of total understanding of the definition of CP. I guess what I was trying to get at is that 5h power will be lower than CP as calculated by certain models and if approximating with CP60, since I notice the actual power profile continues to drop over prolonged periods rather than go horizontal. If that makes sense. 


Let me apologize in advance if any of this is too basic. Not knowing you, it's hard to know where you are on the learning curve of physiology! (I'm curious, what did you do your PhD in?)

So first, let's mathematically define CP.  Basically, you have someone do 3 rides, as hard as they can. (You can also guess some power numbers, set a computrainer or ergometer, and see how long it takes them to drop their cadence more than 5 rpm despite vigorous encouragement).  You want the tests to be between about 2 and about 20 minutes duration. You make a graph of power vs time with these 3 rides, and you plot a line through them. Where it levels out = critical power. You end up with 2 parameters, if you are using the 2 parameter model...the CP, and the W', which was formerly known as the anaerobic work capacity.

You can read a little article I wrote on CP here: http://physfarm.com/new/?page_id=511

...and on some aspects of my work with W' here:  http://physfarm.com/new/?page_id=563

A good way to think about critical power is in terms of VO2. If you exercise below the lactate threshold, what you see is that oxygen use rises and stabilizes. If you exercise between the LT and the CP (in the "heavy domain"), you get an interesting phenomenon. Oxygen use rises, kind of levels out, and then you see a second "hump" where oxygen use rises again, before stabilizing as much as 20 minutes later. This second part is called the "slow component of VO2".

If you go above CP, that second "hump" (the slow component) never stabilizes. That is, VO2 rises continuously until you reach VO2max, and then you punk out soon afterwards. So, the question is, how do we make the division between an effort where you are doing something hard but sustainable, to something that puts you on an inevitable path towards rapid fatigue and VO2max. The CP model is useful in this regard.

(If  the field of VO2 kinetics per se is of interest to you, a selection of authors you want to look up are Brian Whipp, Susan Ward, David Poole, Andy Jones, Tom Barstow, Bruno Grassi, to just name a few. There are a heck of a lot more. You can follow their lineage backwards if you want more historical perspective, and forward as well. David and Andy put out a review a year or two ago that is really nice. The textbook they co-edited is also very good, and is not hard to read.)

The modern understanding of CP goes like this: We aren't really saying that you could do something "for a long time", or "forever", or 5 hours. When calculated correctly, we are defining that infinitesimally thin line between sustainable and time limited exercise, and the boundary between the heavy and severe domains, between states that are physiologically compensable or not. This is how it is understood by the present-day physiologists who study these phenomena in major academic centers and constantly publish on them. It's a bit like gravity: Newton understood it one way and described it nicely. Einstein came along and described how gravity is better understood as the result of the warping of space. Same physical phenomenon, different understanding.

Joe Cyclist, in decent shape, can hold something approximating CP for 30-40 minutes. A really great athlete may hold it as long as an hour. However, aiming to ride at 'CP' is in itself a dangerous enterprise. For example, if you calculate CP at 250W, and then you try to ride just at 250W, you are (statistically speaking) just as likely to be riding above it as you are below it. You might be riding at 252W, or 248W. Or, you may be a watt off in your calculation. It makes a big difference.

It is important to realize that there is no particular reason to assign special significance to 1 hour power, often referred to as CP60 (as you just did above...to be totally pedantic you shouldn't do that. CP is CP...adding numbers is a bastardization of the scientific term. But I know what you are driving at). Why should "CP60" be more important than "CP50" or "CP70"? We pin special significance to CP because modern exercise physiology (and in particular, VO2 kinetics) indicates that it is modeling a fundamental physiological change that results in rapid fatigue, and we want to know about that so that we can avoid it. The CP calculation pins it down quite nicely. My small contribution to the field has been in defining the extent to which we can go above it / below it, i.e. a mathematical framework for understanding discharge and recharge of the W'. This new model works very, very well. We can predict athlete exhaustion to within just a handful of seconds / within a kJ or so. And not just in the lab, but in the field with real athletes in real races / training situations. See http://www.ncbi.nlm.nih.gov/pubmed/24509723 for details on the stats.



Interesting to hear your explanation. Now you mention the phosphate groups I did notice an article title authored by yourself with 31P-MRI studies so I guess I need to have a read of that.
 

The MRI stuff is very cool to do because you can actually watch all the stuff go wrong in real time, inside the muscle. You can actually see a bit of our rig in this video:

https://www.youtube.com/watch?v=CcdeimzYWZM&feature=c4-overview&list=UUKKhgfDW8KpINAd4W1QzRUA



Hope you enjoyed England, though I'm not sure Exeter would be most peoples choice of destination ;) not much there other than excellent scientists :-)


To be honest, I loved County Devon. Nice cider and seafood, beautiful beaches. Good surfing. Nice people. I actually had the opportunity to stay, and I would have happily lived and died in Exeter if it weren't for my USA sized medical student loan debt!

Cheers,

Phil


 

Menko Johnson

unread,
Feb 20, 2014, 7:41:44 PM2/20/14
to Dr. Philip Skiba, golden-cheetah-users
Hi everyone,

This is a a really interesting point that has always confused me on CP.  I think because of connections to FTP and holding it for 60 minutes, CP (ala CP60) has always been confusing to me.  CP has always been much higher than FTP for me, and my curve shows that.  Dr. Phil's statement that CP is what an "average" cyclist can do at CP is maybe 30-40 minutes.  I'm much below average in that respect, as the CP and my actual power data only intersect between 5m-20m, and anything longer my actual power dips significantly from CP and anything shorter than 5 mins I fall off again (see attached)Inline image 2


So I've never really found much utility in using CP for helping me focus on performance, as racing seemed to really fall into a few categories: 5-20s efforts (accelerating out of corners, moving up, etc), 1-2 minute efforts (creating gaps or bridging them), and then longer efforts over 20 minutes or the cumulative effort of an entire race (W' is of particular interest here to me).

FTP in my case (sort of cut off in this picture) is about 235w in my particular case, so now I'm more interested in what the training implications are for thinking of extending my actual performance to better match the CP curve at times beyond 20 minutes.  While I've spent lots of time working on SST (2x20s @ 90-95% of FTP/235w) and trying to push that CP power out past 20 minutes, I'm curious about how it can be a more predictive tool for helping highlight or target areas for my own training.

Of course this has all be hashed over and thoroughly confusing on the wattage list recently, but I'm more interested in the practical application of this data vs. the physiological underpinings or what's actually being measured ;) 

So my question is, how can I use this CP data more effectively (or in conjunction with FTP) to help direct my own progress and improvement.

thanks!

MJ





--
--
_______________________________________________
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 the Google Groups "golden-cheetah-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to golden-cheetah-u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.



--
Menko Johnson

Jon Lidster

unread,
Feb 21, 2014, 5:24:52 AM2/21/14
to golden-che...@googlegroups.com
Hi Dr Phil

Thanks again for the excellent description of CP and the links and references I will be sure to read them. No need to appologize I am a chemist not a physician and not a very good mathematician either. But to answer your question and try to avoid hijacking Menkos excellent question.

My PhD was cancer targeted medicinal chemistry, which didn't realy make any grounbreaking discoveries in terms of advanced cancer drugs but there was some solid chemistry done  http://bradscholars.brad.ac.uk/handle/10454/5679

I had a look at your video of the MRI which is cool. I didn't realise you actually obtain a 31P spectrum, It's like a big NMR spectrometer and the human is the NMR tube, I thought the MRI was mainly an imaging tool. How do you determine the cellular proton concentration using 31P MRI, is it the chemical shift change of the peaks you observe?

You are also right, Devon and Cornwall are some great counties and you can't beat the pasties down there.

Many thanks

Jon

MrDwebborn

unread,
Feb 21, 2014, 5:27:01 AM2/21/14
to golden-che...@googlegroups.com
The CP graph will only ever be as good as the data you put into it. When I do 5x5 or 2x20 for example, unless I go for broke on the first set I'm never realistically going to get 5m or 20m PBs. Similarly for a race, with all the repeated efforts, I'm unlikely to be producing powers for a duration that I couldn't better if I was doing it from rested as a single effort.

To meaningfully track CP you'll need do the max effort tests dr. phil describes to give it proper parameters.

ps. how are you calculating your ftp, when it's clear from the graph that you've never done a max effort beyond c.18m?




Menko Johnson

unread,
Feb 21, 2014, 9:42:26 AM2/21/14
to MrDwebborn, golden-che...@googlegroups.com
Thanks for the reply.  I've done FTP testing in multiple ways: 1 hr TT, 20 min x.95, are the two primary. The data you see is accurate and does include efforts greater than 18'. Two confounds are: I'm light at 128 lbs so while I can put out 270w for 20' on a climb, on the flats it's more like 245w. The second is there are few climbs over 20 minutes here. 

Here's data from an all out effort Tuesday where I did a 20 min hill climb which almost exactly is my CP number (lap 2)

image.jpeg
I didn't snap the photo of a particular ride that was to do CP testing I just took the most recent to get a picture of the curve. 

So my question is still the same. If I'm doing a 2-3' and 20' effort to determine CP the data I have is accurate but as an athlete the big differential between that number and 1hr sustainable power by which all my zones are set makes me wonder how to use that information more effectively.  Even with using coggan's zone chart on w/kg I have the same pattern: untrained couch potato in efforts <30s, stronger 5m and then drop off at 60m. 

What's interesting or confusing is the "what to do about it" piece as if you say I want to raise FTP and do 2x20s @93-95% FTP then I'm hammering out at 220 on the trainer which is a hard effort for the flats. However with the CP number so strongly influenced by the climbing durations and not coming from a specific field test it would say i should be up in the 245+ range which is not possible on the trainer and maybe maximal on the flats at 20 mins. 

So how should I be making the best use of CP numbers?  Should I ignore the auto curve?  



MJ
-------------------
Mastering 1 finger typing

Mark Liversedge

unread,
Feb 21, 2014, 9:43:26 AM2/21/14
to golden-che...@googlegroups.com
Jon,

If you have good maths I have an interesting "linear algebra" problem you can help with in GC ... 

Let me know,
Mark

MrDwebborn

unread,
Feb 21, 2014, 10:16:08 AM2/21/14
to golden-che...@googlegroups.com, MrDwebborn
I referenced 18' as there is a big drop off after that on your chart, which suggests you've not ridden any longer durations at the same effort - assume it comes from a hill climb. Differences in what you can produce hill v flat, road v tt bike, outdoors v indoors are not uncommon. 

GC has an excellent filter, so I would suggest that you tag your rides with sufficient comments to enable you to filter out relevant comparisons. 


Jon Lidster

unread,
Feb 21, 2014, 11:00:38 AM2/21/14
to golden-che...@googlegroups.com
Mark I said I am NOT a good mathematician, infact very incorrect statement I am not a mathematician at all!! I should have said I am NOT very good at maths, haha

Sorry if I have disappointed you. I wish I was good and could help in some way.

Kind regards

Jon

Mark Liversedge

unread,
Feb 21, 2014, 12:21:45 PM2/21/14
to golden-che...@googlegroups.com
Oops - I missed that. But my post flushed out a volunteer so here's hoping!
Mark

vingrirr

unread,
Feb 21, 2014, 6:40:14 PM2/21/14
to golden-che...@googlegroups.com
Mark - I'm traveling currently but might be able to take a swing at it or at least serve as a sounding board.  If it involves setting W'bal = 0 and the Simplex Method, I think I know where you're going!

Christopher Smith

unread,
Jul 29, 2014, 5:28:18 PM7/29/14
to golden-che...@googlegroups.com
Dr Skiba indicated 3 rides of 2 to 20 minute duration for testing. Are there specific durations that yield more accurate results, or does the power/time function provide accurate modeling regardless of the chosen duration, so long as the efforts are truly maximal. In other words, is it safe to say the CP and W' values **should** be the same if one does maximal efforts for 3/9/15 minutes vs, say, 5/12/20 minute durations?

Forgive me if I've not articulated this well, or seem ignorant. I'm trying to get a good grasp on the concept of W' balance and it's practical real world use.

Chris.

Mark Liversedge

unread,
Jul 29, 2014, 5:34:31 PM7/29/14
to golden-che...@googlegroups.com
Hi Chris,

You're spot on. So long as the efforts are in the circa 2 and 20 min range and all out they will be perfect.
Noone can nail an all out effort to the second (or minute!) so so long as they are truly maximal they will be good ...

.. but they will hurt ;)

There is also a 3 minute all out test that might be of interest to you and also a wingate. I'm going to add testing to the train view in 3.11.

http://www.ncbi.nlm.nih.gov/m/pubmed/24888425/
http://ep.physoc.org/content/93/3/383.full

Mark

Andy Coggan

unread,
Jul 29, 2014, 7:41:03 PM7/29/14
to golden-che...@googlegroups.com
On Tuesday, July 29, 2014 4:28:18 PM UTC-5, Christopher Smith wrote:
 
does the power/time function provide accurate modeling regardless of the chosen duration, so long as the efforts are truly maximal.

Christopher Smith

unread,
Jul 29, 2014, 8:12:15 PM7/29/14
to golden-che...@googlegroups.com
Thank you for your response, Dr Coggan. From the article, "Repeated measures ANOVA revealed that CPI123 (201.0+/-37.9W) > CPI135 (176.1+/-27.6W) > CPI345 (164.0+/-22.8W) (P<0.05)." 

These are substantial differences! "Anaerobic inertia" is also mentioned as a plausible explanation, and that is precisely what led me to ask my original question about the accuracy of modeling being independent of duration. Without much real thought, I would have guessed the anaerobic impact would introduce significant differences in the CP values relative to the combination of durations used to model CP. Values that range from 164 to 201...a 37 watt difference. I would guess the disparity increases as CP values increase? My next question becomes, how does this disparity in modeling impact the reliability and predictability of W' balance? Am I completely missing something here? With such differences in CP modeling, it seems like it might almost nullify the reliability of W' balance across the spectrum of any given athlete, no?

Again, forgive my response if I sound ignorant. It's partially because because I am. Feel free to suggest I revisit a few text books before taking part in this conversation.

Chris

Mark Liversedge

unread,
Jul 30, 2014, 3:11:00 AM7/30/14
to golden-che...@googlegroups.com
That's why approximately 2-20 intervals are recommended not 1-10.

Andy Coggan

unread,
Jul 30, 2014, 6:42:24 AM7/30/14
to golden-che...@googlegroups.com
As Mark indicates, lengthening the duration of the efforts used to calculate CP and W' helps, which is why I originally made that suggestion in the chapter on power-based training I wrote for USA Cycling ca. 2000. However, it doesn't solve the problem completely, so it is best to either always use tests of the same duration, or use a better model (note that even the 3-parameter version of the CP model is sensitive to test duration)>

Xplora213

unread,
Aug 8, 2014, 9:43:32 AM8/8/14
to golden-che...@googlegroups.com
On this topic, if my typical experience of hard riding seems to be that, assuming a CP of 250w, I struggle to maintain 300w for 5 minutes, but find I crack for a minute and then can return to that strong 300W riding again (much faster than my peers), would this indicate an above average W' ?? Anaerobic recovery seems to be good, but the baseline aerobic isn't quite as good?

Have had a lot of trouble understanding how this applies to me, despite reading Skiba's articles etc.

Pete from AUS

unread,
Aug 15, 2014, 11:45:18 PM8/15/14
to golden-che...@googlegroups.com
Is W'Bal calculated from ride time or time-of-day?
For example if we re-group at the top of a hill my PM will 'pause' recording.
But obviously I am recovering whilst we wait a minute or two.
Looking at the ride plot, W'Bal doesn't recover until we roll off.
time.JPG
plot.JPG

Mark Liversedge

unread,
Aug 16, 2014, 2:47:49 AM8/16/14
to golden-che...@googlegroups.com
The pause in recording should be switched off or use the fix gaps tool to ensure you have 1 second recording.

Pete from AUS

unread,
Aug 16, 2014, 3:12:43 AM8/16/14
to golden-che...@googlegroups.com
Could it be that the iBike to GC mapping uses the "moving time" from iBike CSV files because in GC edit there is no gap in time-of-day to fill. I'll look into whether my unit can have different record settings
cp edit.JPG

Mark Liversedge

unread,
Aug 16, 2014, 3:44:08 AM8/16/14
to golden-che...@googlegroups.com
Sorry Pete I'm reading on a phone and didn't notice that. It makes sense because the w'bal code smooths out gaps anyway.

Is the real time stamp available in the files, coz we could use that instead?

Mark

Pete from AUS

unread,
Aug 16, 2014, 3:46:41 AM8/16/14
to golden-che...@googlegroups.com
Yep column O: "Timestamp"

Pete from AUS

unread,
Aug 16, 2014, 4:11:29 AM8/16/14
to golden-che...@googlegroups.com
Timestamp
2014-08-15T19:43:54Z
Reply all
Reply to author
Forward
0 new messages