this is precisely why i've largely* abandoned any tracking of stress. SBR are but three small variables swimming in a stew of other stresses that no PMC would be able to quantify.


*still track with pros and amateurs with minimal extraneous other/life stressors

---

36 kona qualifiers 2006-'23 - 3 Kona Podiums - 4 OA IM AG wins - 5 IM AG wins - 18 70.3 AG wins
I ka nana no a 'ike -- by observing, one learns | Kulia i ka nu'u -- strive for excellence
Garmin Glycogen Use App | Garmin Fat Use App

The medium hard/gray zone days are what I term triathlon specificity. Pretty much any race distance (non drafting) sprint to Ironman intensities are included in that span. It's stuff that has to be addressed. What the issue is is when that's the only type of training the subject ever finds themselves in. Sure spacing it out would give you more recovery time but oscillating between 1 and 3 provides a greater stimulus in the same time frame so why bother SST-ing it and then taking a lot of time off?

---

36 kona qualifiers 2006-'23 - 3 Kona Podiums - 4 OA IM AG wins - 5 IM AG wins - 18 70.3 AG wins
I ka nana no a 'ike -- by observing, one learns | Kulia i ka nu'u -- strive for excellence
Garmin Glycogen Use App | Garmin Fat Use App

really depends on the goal and time of year,

You could space it out and achieve high success or you could modulate load depending on how you measure and achieve success.

IE "better" or more spaced out recovery or a static, or spread out response to training load may or may not be the best option all the time.

Certain measurable outcomes are basically where the rubber hits the road in training, training stress, response and all the other things you do outside of training…. find those ;-)

Maurice

One nice thing about tracking RPE alone though is that you can do more things with it than if you just replace specific TSS values with TSS estimates based on RPE:
- you can then choose to (or not) multiply it by time. A 2/10 90min ride is not the same as a 2/10 4hr ride. There's an argument to be made both ways; that is, I think there's a reason to incorporate time into the RPE value and also to keep it separate. But if you want to spin out as being separate from time, this allows you to do that and to then make use of the fact that you are also recording time for the session.
- you can then see where RPE and TSS diverge and converge, which I think goes a long way towards being able to make better use of TSS.
- eventually, you could use RPE as a "correction" factor of sorts for TSS. Knowing both RPE and TSS also should help to improve the TSS model in the future...
- lastly, it's simpler. One of the best parts of RPE is, "quick, out of 10, how hard was that?" When you need to start doing multiplication, that makes it more complicated, which also makes it more likely that athletes will overthink the value they assign...

So, yes, you can currently use TSS as a proxy for RPE. But I think there are more reasons than not to keep RPE separate and to just allow you to enter it.

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

Dr Coggan, you have done some great work! With regard to the modeling of training load you are a trailblazer along with Bannister and Foster. Your introduction of the Performance Manager Chart is an innovation that athletes world wide have benefited from. Thank you.

You are correct, yes RPE based scores based on Foster's work can be used in the TMC. I am sure that as long as the methodolgy is held constant a valid apples to apples analysis will work. I'm sure that TRIMPS can work too.

Many curious athletes are reading articles, blogs and forum posts to learn the nuances of tracking training load. Of course caution should be used reagarding context and some authors are more authoritative than others. McGregor, Skiba, Sousa, Rapp, Ale Martinez, Friel among countless athletes and coaches have pointed out pros/cons and some real world perspective. Curious athletes who are utilizing this information typically see values expressed in the typical values of your TSS system. I haven't seen many recent articles apply Foster or Bannister's respective methods in the context of real world day to day applications.

The chart in the Friel blog that I linked earlier is in no way a trailblazing concept such as the collective work of you Foster and Bannister. It is a simple nuance in the real world application of this concept of training load. The benefit of this particular chart is that it allows for use of RPE and will give numerical values similar to thise expressed in these various sources of information giving real world examples to the applications of your principles. Athletes who have never seen that Friel article/chart intuitively do something similar. For example an athlete who may ride a steady aerobic 1 hour ride on a spin bike might theoretically say "that felt like 50 tss on my road bike" and then manually make an entry. The Friel chart just gives a little more structure to such a potential scenario.

There is no perfect application of subjective and objective information. Training will always remain a delicate balance of art and science. Thankfully there are trailblazers such as yourself who pioneer such innovative concepts. It's also fortunate that there are a bunch of people in the trenches willing to share their ideas and experiences which help advance understanding through this brainstorming process.

I am assuming these guys are training around 40 hours a week? What about age groupers who can only train 10-20 hours a week? The usual answer seems to be keep the same ratios. But why not drop 10-20 hours of easy training (Edit -Weekly) and do the same or less hard training but at a higher ratio relative to easy?

Those are XC skiers from Norway over the last ~40 years. They train 600-900hrs a year, much less than triathletes.

You have a very good point. Taking what someone does in one context and directly translating to another context is potentially a bad decision. Also somewhat pointless to post a graph with no context.

In a way Nordic sking is triathlons closest sport. Classic is good for running , skating good for cycling . And they use their arms too
It's not a suprise that many cyclist train Nordic skiing in the winter and I guess even less a suprise that many triathlete do Nordic sking camps in the winter

What this is showing is the distribution of intensity levels.
Youth and junior skiers also train like this and with big success.
Other athletes like Ingrid Kristiansen followed the same structure.

I have not seen any other posting actual training here. I would like to see the distribution for other athletes. But for some reason in other sports (and countries) coaches and athletes will not share their data.

In general, if you train 5+ times a week have two intervals day mostly in zone 4 (short rest) and the rest should be easy.

My own distributions for speed power. Versus a year ago.

Covered in here: http://blog.rappstar.com/...woodlands-redux.html

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

How does 2 interval days fit into a triathlon plan? All brick workouts? Only do a hard swim every other week?

Why 7 power zones the 1st year but 8 the 2nd year? That's going to skew the distributions, no?

wait, what? Is that a typo? according to you recovery = IF < 0.55 (and I would agree)

Active Recovery<55%<68<

---

____________________________________

Are you ready to do an Ultraman? | How I calculate Ironman race fueling | Strength Training for Athletes |

Lol, you could use that joke again from a few pages back ;-)

Actually 6 vs 8 bike. Maybe he was combining lower zones etc??

Maurice

Interesting... FWIW, the last released GoldenCheetah version allows to build a PMC based on Session RPE, the new development versions also allow to introduce the sport dependent "A" constants via a user defined metric which then could be used to build a PMC (just like any other metric like TRIMPs, TSS, etc.).

---

Ale Martinez
www.amtriathlon.com

The training levels are based on average power, and are expressed as a percentage.

Intensity factor is based on normalized power, and is expressed as a decimal.

Due to variability in power output, the two scales are not equivalent.

Level 1 is defined as an average power of <55% of FTP, and (due to power variability) typically results in an IF of <0.75.

When attempting to compare training described using my nomenclature to, e.g., research studies specifying intensity as a percentage of VO2max (which IIRC is the context here), it is better to think in terms of IF. IOW, since on average FTP occurs at about 80% of VO2max, recovery would be roughly equivalent to less than 80% x 0.75 = 60% of VO2max. You would not expect that intensity to provide much of a stimulus for adaptation in already well-trained subjects, unless perhaps it was performed for many hours per day.

I don't know. TP is giving me that.

I generated both graphs at the same time (yesterday). And this is what I have in settings:

If somehow I accidentally had Durata zones (8 zones) in there at this time last year, would TP hold onto that? If so, that would seem to be a bug in TP...

I have WKO4, but I am basically totally useless with it. I can't figure out how to make the same graphs in it...



EDIT: I figured out how to do it in WKO4. Updated the blog. Updated graph here:

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

Thanks for posting actual training. Not many are willing to do that.

And if you want to know what you are capable of, do a test.

i.e.
Testing is training.
Performance is the best predictor of performance

We need predictive models, not descriptive ones based upon a dubious metric.

Mark

I love how you quote me then turn right around and slam my contributions based on things that were thoroughly hashed over a decade ago. Not only are you late to the party, it seems that you're cut from the same cloth as good ol' Trev, i.e., you act like a spurned lover.

But anyway: while it is far from perfect, TSS is the only method of quantifying training load that has been validated based on measurement of glycogen utilization.

Wasn't aware of this, can you post a URL to the study ?

Mark

As discussed on the wattage list about 10 y ago, the data are drawn from the studies we did at UT-Austin while I was working on my PhD.

Shame, but helps to explain the very small sample size in the plot you posted.
As usual, your grand claims cannot be scrutinised.

Mark

That's a good explanation. There's also a tendency to forget it's just a relatively crude model of what's going on, and for a variety of reasons it can be a total lie.

E.g. a couple of times for 'A' races when my "PMC" told me I was at my fittest with a great "stress balance" I was utterly overtrained and worthless.

I no longer "TSS-chase." It's a tool. But not a panacea.

Even n=1 would be higher than, say, BikeScore, which you have endorsed/embraced by implementing in GoldenCheetahR. IOW, it seems that you're rather two-faced about what does/does not meet your standards.

In any case, the data are what they are, with the only thing one needs to know in addition to the graph and the information in the peer-reviewed papers is that I estimated FTP for each subject based on Coyle's definition of LT and his 1991 Med Sci Sports Exerc paper.

"In god we trust - everyone else must bring data." - W. Edwards Deming