Homebrewed "TSTWKT"

[acoggan]

Leipheimer lost 42s because he's a tiny dude and it was super windy + dumping rain when he went (unlike most of the competition.)
Plus, he's a bit of a sissy in the wet.

That could be it. Then again, didn't he lose about the same amount of time per kilometer as he gave up to people like Sebastian Lang in the Tour TTs?

 

[Tom Anhalt]

There are various types of stress, and they do not all have the same impact on a given performance nor the type of training you could do next (during that session, that block of days, that week...). And it will vary between individuals.

Exactly...and I'll repeat, "Hence, the use of NP and TSS. That's the whole point."

If you understand the definitions of NP and TSS, you'll understand that I'm not just trying to be cute here.

You also seem to be missing the point that the tool is not meant to be used to micromanage the content of the training, but merely to attempt to quantify the metabolic stress of the training...

 

[acoggan]

one thing I've realized about the name "Performance Manager" is that you can't just shorten it to PM since that will get confused with PM for Power Meter.

You could always refer to the Performance Management Chart, PMC.

 

[acoggan]

Hmmm...I'm going to guess that you guys coded up one of Busso's equations, right?

IIRC, Kraig W. tried that with less than stellar results, right Andy?

Actually, as I recall Kraig never would explain what he did...he just claimed to have "cracked the code" and the approach wasn't worthwhile.

 

[WarrenG]

Leipheimer lost 42s because he's a tiny dude and it was super windy + dumping rain when he went (unlike most of the competition.)
Plus, he's a bit of a sissy in the wet.

But yeah, I bet he'd love to have something to quantitatively manage his form, which seems to be all over the place this season.

Do you have any examples of how his form "has been all over the place this season"? What weeks or months was it lower than he wanted? What weeks or months was it higher than what he wanted? What weeks was it very, very good?


His form has been very good right when he wanted it to be. How well do you race day after day when you're sick? And I doubt you've had to race when you've had to deal with what he's had.

As for Tour of Germany... Who else is at the Tour of Germany who was near top 10 in the Tour this year? No Kloden, though he is a German on a German team. Different schedules this year. Different priorities and differerent motivations.

Regarding motivations, Levi's going to a new team, one that likes to provide support for their GC guys during major races that is far more useful than what he was getting at Gerolsteiner this year. His move to Discovery was not something considered only recently. I predicted his departure from Gerolsteiner before I knew about it.

 

[Tom Anhalt]

Actually, as I recall Kraig never would explain what he did...he just claimed to have "cracked the code" and the approach wasn't worthwhile.

Well, since that's all on his "NP Pay-per-view" board now, I guess all we have are memories

IIRC, you kept pressing him to explain how he calculated the coefficients and he just kept coming back with something like "I applied the approach exactly as given in the paper". Of course, at that time you guys were embroiled in your TSS vs. KJ "pi$$ing match" so not a lot of clarification was forthcoming...just a bunch of posturing IMHO

I also remember you throwing your KJ data into the PMC about that time to help demonstrate how TSS worked better as an input.

To be honest, that exchange between you two was a HUGE help to me in figuring out the PMC approach in that I realized that the Busso approach wasn't it

 

[WarrenG]

Exactly...and I'll repeat, "Hence, the use of NP and TSS. That's the whole point."

You also seem to be missing the point that the tool is not meant to be used to micromanage the content of the training, but merely to attempt to quantify the metabolic stress of the training...

You cannot (with excellent results) quantify or really evaluate the metabolic stress of the training when all training at a given power/duration is treated the same, as it is with TSS and NP. Simply, NP and TSS do not include enough information about the various types of training stress to provide excellent evaluations about the metabolic stresses of the various types of training one could/would do.

Do you need a long list of examples of training with similar TSS and NP but different types and amounts of stress, and that might vary among individuals, and could also depend on what was done previously?

 

[Tom Anhalt]

You cannot (with excellent results) quantify or really evaluate the metabolic stress of the training when all training at a given power/duration is treated the same, as it is with TSS and NP.

That statement right there demonstrates your lack of understanding of the concepts...

Do you need a long list of examples of training with similar TSS and NP but different types and amounts of stress, and that might vary among individuals, and could also depend on what was done previously?

Sure...and while your at it, can you provide details of exactly how the stress is measured and exactly how that stress measurement technique is superior? Of course, no "micro-interval" examples please.

 

[acoggan]

You cannot (with excellent results) quantify or really evaluate the metabolic stress of the training when all training at a given power/duration is treated the same, as it is with TSS and NP. Simply, NP and TSS do not include enough information about the various types of training stress to provide excellent evaluations about the metabolic stresses of the various types of training one could/would do.

Amber Neben's coach would disagree with you.

 

[WarrenG]

Originally Posted by WarrenG
"You cannot (with excellent results) quantify or really evaluate the metabolic stress of the training when all training at a given power/duration is treated the same, as it is with TSS and NP. "

That statement right there demonstrates your lack of understanding of the concepts...

I know what those numbers are in my Cyclingpeaks files and how they compare to reality. One can definitely do better than NP and TSS for evaluating training load, the various stresses, and their impact on an individual. There simply isn't enough information inputted (beyond watts and duration) to NP nor TSS to allow those metrics to be complete. IOW, incomplete information in, leads to incomplete information coming out.

This is why I said that RapDaddyo's approach could be better, because it included more information than just NP and TSS.


Originally Posted by WarrenG
"Do you need a long list of examples of training with similar TSS and NP but different types and amounts of stress, and that might vary among individuals, and could also depend on what was done previously? "

Sure...and while your at it, can you provide details of exactly how the stress is measured and exactly how that stress measurement technique is superior? Of course, no "micro-interval" examples please.

Here's a simple one, taken right from my own training plan (converted to % of FTP). All done in one session.

1x20' warmup, np= ~`70% of FTP
3x6' at 90% of FTP.
1x12' at 100% of FTP.
4x4' at 120% of FTP.
3x12" efforts at AP =400-500+% of FTP.

Think you know enough to tell me the relevant stress levels or whatever else you can calculate?

Also included in my plan are the terrain for these efforts, the cadences for each, the order that they are done, the rest between each effort, and whether they're done in the saddle or out of the saddle, or both. Does your calculation include these, or what I did the day before? Or the next day? Or what my natural and trained abilities are? How well I recover from each type of training included above? How about the temperature during the ride?

All of these affect the stress.

As for how the stresses are measured and evaluated... Sit down with a really good coach for a few hours and ask them about this topic. Here's the guy that has taught me about this...http://www.athleticamps.com/cycling/coaching/testa/

 

[WarrenG]

Amber Neben's coach would disagree with you.

So what?

 

[Tom Anhalt]

I know what those numbers are in my Cyclingpeaks files and how they compare to reality.

How is this comparison made? What is your criteria, or "gold standard" to say how they compare to "reality"? It's a simple question.

There simply isn't enough information inputted (beyond watts and duration) to NP nor TSS to allow those metrics to be complete. IOW, incomplete information in, leads to incomplete information coming out.

Nobody has ever said that the concepts of NP and TSS are absolute in their "completeness". The question is: "Are they complete enough to be useful in a performance management tool?"

Here's a simple one, taken right from my own training plan (converted to % of FTP). All done in one session.

1x20' warmup, np= ~`70% of FTP
3x6' at 90% of FTP.
1x12' at 100% of FTP.
4x4' at 120% of FTP.
3x12" efforts at AP =400-500+% of FTP.

Think you know enough to tell me the relevant stress levels or whatever else you can calculate?

No...because you left out the rest durations. I hate to say it, but once again, you're demonstrating your incomplete understanding of NP and TSS.

Also included in my plan are the terrain for these efforts, the cadences for each, the order that they are done, the rest between each effort, and whether they're done in the saddle or out of the saddle, or both. Does your calculation include these, or what I did the day before?

Of course the PMC takes into account what you did the day before...and the day before that...and the day before that too...and so on.

Or the next day?

Sure...you can put in "what if" scenarios and predict where you'll be.

Or what my natural and trained abilities are? How well I recover from each type of training included above?

This is accomodated to the level needed for the tool. Your trained ability is represented in your Chronic Training Load (CTL) and your Acute Training Load (ATL) calculation is governed by your recovery rate...and that's modifiable.

How about the temperature during the ride?

I agree that has an effect, but like most of the things you're referring to (cadence, temp, in saddle or out), you have to ask if the magnitude of the affect is enough to invalidate the tool. The answer is no.

As for how the stresses are measured and evaluated... Sit down with a really good coach for a few hours and ask them about this topic. Here's the guy that has taught me about this...http://www.athleticamps.com/cycling/coaching/testa/

I'll take that as an "I don't know" or "I don't have one" answer. Thanks.

BTW, it's a common saying that "If you don't know a subject well enough to be able to teach it to someone else, you really don't know it". So, if you can't teach me how these stresses are measured and evaluated...'nuff said.

 

[RapDaddyo]

Here's a simple one, taken right from my own training plan (converted to % of FTP). All done in one session.

1x20' warmup, np= ~`70% of FTP
3x6' at 90% of FTP.
1x12' at 100% of FTP.
4x4' at 120% of FTP.
3x12" efforts at AP =400-500+% of FTP.

Think you know enough to tell me the relevant stress levels or whatever else you can calculate?

Why bother computing TSS? It's only 29mins of L4-L7.

 

[WarrenG]

How is this comparison made? What is your criteria, or "gold standard" to say how they compare to "reality"? It's a simple question.

If the TSS says I should feel x, but I don't, or I am able to handle more of one type of training than TSS would predict, or less of some kind of training than TSS would indicate, or the accumulated TSS is x but I feel different than TSS indicates, etc.

The simple fact that TSS doesn't concern itself with the various types of training stress but lumps them all together is a deal breaker for anyone who understands that not all types of training stress are the same.

It also does an especially poor job for neuromuscular training. For many racers this training is important and its stress can not be minimized the way it is with TSS/NP.

And then, TSS doesn't care what order different training is performed, within a session, nor day to day, and this can be very important. I used to look at TSS but it was of so little value compared to what I already knew by other means that I haven't looked at it lately.

Nobody has ever said that the concepts of NP and TSS are absolute in their "completeness". The question is: "Are they complete enough to be useful in a performance management tool?"

If good enough is good enough for you, then yes. For me, that tool is not good enough nor as accurate as what I can do without it. I guess your standard of what is useful is lower than mine, or that you lack the understanding about training stress that would enable you to know more than you can with only NP/TSS.

No...because you left out the rest durations. I hate to say it, but once again, you're demonstrating your incomplete understanding of NP and TSS.

Fine. Use 4' at 60% of FTP. Now, answer my questions. With specifics. So again...

Here's a simple one, taken right from my own training plan (converted to % of FTP). All done in one session.

1x20' warmup, np= ~`70% of FTP
3x6' at 90% of FTP.
1x12' at 100% of FTP.
4x4' at 120% of FTP.
3x12" efforts at AP =400-500+% of FTP.

Think you know enough to tell me the relevant stress levels or whatever else you can calculate?

Also included in my plan are the terrain for these efforts, the cadences for each, the order that they are done, the rest between each effort, and whether they're done in the saddle or out of the saddle, or both. Does your calculation include these, or what I did the day before? Or the next day? Or what my natural and trained abilities are? How well I recover from each type of training included above? How about the temperature during the ride?

I agree that has an effect, but like most of the things you're referring to (cadence, temp, in saddle or out), you have to ask if the magnitude of the affect is enough to invalidate the tool. The answer is no.

No, for you, but for many others that tool isn't better than what they already have. Also, the tool could be made more useful if it allowed for these things. Incomplete information in, incomplete information out. We(?) can do better than that.

Of course the PMC takes into account what you did the day before...and the day before that...and the day before that too...and so on.

Does it take into account what I did for each type of training I did. For example, what is the accumulated stress of doing x type of training 2-3 days in a row? An easy example for you, how about if it's at 120% of FTP, 4x4', 4' rests at 60% of FTP, and you do this 3 days in a row? or 2 days in 5, or 3 days in 7? Or on a hill sometimes and on the flat other times? At 70 rpm's vs. 100 rpm's? Let's hear your answers as predicted by your software.

If you don't know the answers maybe you can learn them from a good coach. I have. Ask a specific question, provide your specific answer from your software as it applies to the examples I have provided, and I'll provide my specific answer. Can you do that?

 

[WarrenG]

Originally Posted by WarrenG
Here's a simple one, taken right from my own training plan (converted to % of FTP). All done in one session.

1x20' warmup, np= ~`70% of FTP
3x6' at 90% of FTP.
1x12' at 100% of FTP.
4x4' at 120% of FTP.
3x12" efforts at AP =400-500+% of FTP.

Think you know enough to tell me the relevant stress levels or whatever else you can calculate?

Why bother computing TSS? It's only 29mins of L4-L7.

If that's all that TSS would tell you about the stress of that training session you've missed a lot of useful information.

 

[frenchyge]

Do you need a long list of examples of training with similar TSS and NP but different types and amounts of stress, and that might vary among individuals, and could also depend on what was done previously?

Actually, a model which used your proposed methodology and provided better results than the Performance Manager (....ugh ) would prove your point most elegantly.

You can rail all you want about the quality of this new tool, but how can you say it's inferior without anything to compare it against?

 

[WarrenG]

Actually, a model which used your proposed methodology and provided better results than the Performance Manager (....ugh ) would prove your point most elegantly.

You can rail all you want about the quality of this new tool, but how can you say it's inferior without anything to compare it against?

I see it every week-objective and subjective evaluation of training stress done skillfully with really good results, with consideration for all the factors I've mentioned. Many, many other people do likewise. To teach yourself, ask why these things matter. Talk with someone who can help you understand why they matter.

If you ever manage to quantify at least most of those factors you'll have something that might be better, but a "tool" that does not even consider so many relevant factors will not be as good as it could be.

Entertaining oneself with mediocre tools is a distraction. Spend the time instead learning what can be more useful, and maybe spend some time developing tools that include at least most of the relevant factors.

 

[Tom Anhalt]

If the TSS says I should feel x, but I don't, or I am able to handle more of one type of training than TSS would predict, or less of some kind of training than TSS would indicate, or the accumulated TSS is x but I feel different than TSS indicates, etc.

The simple fact that TSS doesn't concern itself with the various types of training stress but lumps them all together is a deal breaker for anyone who understands that not all types of training stress are the same.

Well...all I can say is that your experience with NP and TSS is at odds with my experience and the experience of a large (and growing) number of people. In fact, I find it so valuable and accurate that I can actually plan workouts of vastly different power levels and even know ahead of time if the workout is "doable". But, like any tool, if used incorrectly it won't work very well...perhaps that's your problem.

If good enough is good enough for you, then yes. For me, that tool is not good enough nor as accurate as what I can do without it.

You keep saying this...yet you never back it up with real data. Hmmmm....

Fine. Use 4' at 60% of FTP. Now, answer my questions. With specifics. So again...

After failing to give specific answers to my questions, you now are asking for specifics? The nerve of some people...

BTW, you can calculate the TSS for your described workout very easily with a TSS planning spreadsheet that can be easily Googled up. It's not that hard.

Does it take into account what I did for each type of training I did. For example, what is the accumulated stress of doing x type of training 2-3 days in a row? An easy example for you, how about if it's at 120% of FTP, 4x4', 4' rests at 60% of FTP, and you do this 3 days in a row? or 2 days in 5, or 3 days in 7? Or on a hill sometimes and on the flat other times? At 70 rpm's vs. 100 rpm's? Let's hear your answers as predicted by your software.

If you understand anything about an impulse-response training model you'll realize that the accumulated metabolic stress is well accounted for...both on a long term (CTL) and short term (ATL) basis. However, with the time constants for both of those responses being much longer than that you'll need a lot more data than just 1 weeks worth to get anything meaningful out of the model.

So...let's turn this around. Based on your "techniques", what is the accumulated stress? Specifics please

If you don't know the answers maybe you can learn them from a good coach. I have. Ask a specific question, provide your specific answer from your software as it applies to the examples I have provided, and I'll provide my specific answer. Can you do that?

Even better...I'll provide you with my last 1 year's worth of power data and you can then try to predict for me during which time frame I set personal bests for 5 min. AP, 20 min NP, and 60 min. NP. Of course, you could always just search the data for those bests, but that would be unfair. Maybe, I'll edit out my races and tests where those PBs occurred.

BUT...I'll only do that after you go to the track with an SRM as offered by "trackiejay" on FGF

 

[frenchyge]

If you ever manage to quantify at least most of those factors you'll have something that might be better, but a "tool" that does not even consider so many relevant factors will not be as good as it could be.

Entertaining oneself with mediocre tools is a distraction. Spend the time instead learning what can be more useful, and maybe spend some time developing tools that include at least most of the relevant factors.

But the thing about a "tool" is that you can make copies of it, give them to other people, and those copies will do something for them, too. Even if it isn't the best that it can be, if it does something useful for one person, it'll do the same for *everyone* that uses it that way. If it's not perfect, then the next guy can help build on it to improve it.

For all your experience and access to top-notch coaching, people are never going to fully appreciate your contributions here because it's easy to think that your experiences are specific to you. OTOH, Andy has created a tool which many people are now using, and so whenever he says something there are 10 other people that chime in with "yep, thats how it worked for me" and even "here's how I found a new use for the tool." There's so much more power in dealing with principles than in experiences because principles can be universally applied.

 

[WarrenG]

Well...all I can say is that your experience with NP and TSS is at odds with my experience and the experience of a large (and growing) number of people. In fact, I find it so valuable and accurate that I can actually plan workouts of vastly different power levels and even know ahead of time if the workout is "doable". But, like any tool, if used incorrectly it won't work very well...perhaps that's your problem.

Like the tens of thousands of people who race their bikes better than you, and don't use TSS at all? And what about the growing number of people who have tried TSS and then abandoned it for something better?

After failing to give specific answers to my questions, you now are asking for specifics? The nerve of some people...

C'mon smart guy. Answer the question using your software, and then I'll tell you specific answers to the question based on real world experience. Stop dodging the question. Answer with specifics.

BTW, you can calculate the TSS for your described workout very easily with a TSS planning spreadsheet that can be easily Googled up. It's not that hard.

So what? TSS is not accurate enough to evaluate the effects of the workout because it doesn't consider enough relevant factors, which I've listed. There is better information available and I'll provide you with specifc answers once you stop dodging my question based on a specifc example.

Even better...I'll provide you with my last 1 year's worth of power data and you can then try to predict for me during which time frame I set personal bests for 5 min. AP, 20 min NP, and 60 min. NP. Of course, you could always just search the data for those bests, but that would be unfair. Maybe, I'll edit out my races and tests where those PBs occurred.

BUT...I'll only do that after you go to the track with an SRM as offered by "trackiejay" on FGF

Two more weak attempts to dodge the question.

First, I have no interest whatsoever in looking at a year of your data and getting clear answers to all of the relevant questions. I have only asked you to look at one, simple training session. Secondly, that offer from Jay was because at that time I didn't have a PM capable of showing my 5spower and the decline in power in 1s increments as Andy Coggan predicted. He said it would be only 25-50% of peak power near the end of a 200mTT or sprint. I have since been using a PT and I've seen dozens of examples in training with the PT where his prediction is wrong, by significant amounts, just as I've been saying on that topic for years now.

So, stop dodging the simple, specific example and questions I posed about it. Tell us all that you think your software can say about that example and then I'll tell you all the other relevant information I have, that you apparently don't.

Can you manage that, without dodging the question again?


The question for Tom, for the third time...

Use 4' at 60% of FTP for the rest periods.
All done in one session.

1x20' warmup, np= ~`70% of FTP
3x6' at 90% of FTP.
1x12' at 100% of FTP.
4x4' at 120% of FTP.
3x12" efforts at AP =400-500+% of FTP.

Tell me the relevant stress levels or whatever else you can calculate?

Also included in my plan are the terrain for these efforts, the cadences for each, the order that they are done, the rest between each effort, and whether they're done in the saddle or out of the saddle, or both. Does your calculation include these, or what I did the day before? Or the next day? Or what my natural and trained abilities are? How well I recover from each type of training included above? How about the temperature during the ride?

And, how, and why could the stress be different for a different person?