Times change: see my reply to Francois.

The problem is that you haven't presented anything that could help someone train or monitor their workload. The method you 'discussed' (even though you refuse to give any specifics) has no ability to help anyone. The systems and calculations developed by Coggan do have the ability to help people.

I've trained with HR for nearly 20 years, used power for about 8-9 years. In my mind, there is no comparison, I would choose power over HR every time. I still use both, but more just to collect HR data since I like numbers. Power is easier to test, more consistent and seems more relevant to biking. I wish my other sports had objective measures. Running is getting closer but xc-skiing is still just a 'guess' with time in HR zones (hoping you did the zones correctly and that HR doesn't change that much).

I followed my PMC last year leading up to IMC and it seemed to follow my race progression very well. Adjusting to changes in power from heat, bikes, in/out was much easier than adjusting to the HR I was seeing or my RPE. I didn't do formal FTP testing, just monitored my workouts and adjusted zones as needed.

I don't think I'm expecting the TSS to be scientifically validated. I know it follows principles that I understand and it appears to fit in with my personal assessment of my workouts. It's not a training plan or guide, it's a tool to help monitor training so you or your coach can write a better plan. That's good enough for me and from the looks of it many other successful athletes and coaches as well.

Has anyone validated your power/heart rate metric?

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

I only realised there was a thread here after Andy Coggan made a comment on Facebook.

This was an abstract for presentation at the recent World Congress of Cycling Science in Leeds. It is the study for my Masters Degree and I am just working on the final draft of my Thesis. Hopefully there will be a paper or two in this.

With more crunching of the numbers the main thing we have found is that max mean powers in road cycling are highly variable. Even for TT accounting for distance or dropping the bottom half of the data out. Not earth shatteringly amazing stuff but coaches and athletes do obsess on max mean powers for certain durations. This data highlights the folly in that although I only had 5-7 riders with repeated time trial performances so the confidence limits are getting pretty wide.

At the conference in Leeds Dr Simon Jobson asked for my gut on the PMC. I will still use it as I see huge utility but will carry on researching this area and some of the cooler new metrics that are coming out around Golden Cheetah 3.1 and WKO+ 4.0.

---

Hamish Ferguson: Cycling Coach

Yes, fair comment indeed. The subjects were all National and International level cyclists and competition but MMP's from each event don't tell you if a rider was hiding in the pack or on team duties or if riding a TT in a stage race just saving energy for other stages. Or for some TT's if it was an out and back course vs a point to point.

---

Hamish Ferguson: Cycling Coach

One challenge not to forget, is that Golden Cheetah and WKO can't track training load form running and swimming. So if you need to monitor all 3 disciplines to stay on top of overall cardio system stress and associated risks of over training, then you need a system than can look at all 3. While not perfect for swimming and running, it's still a good way to get an overall picture. The challenge with running and swimming is that they are more dependent on technique and the overall efficiency at a given pace can vary more with conditions and equipment selection.

---

TrainingBible Coaching
http://www.trainingbible.com

You could estimate TSS for run and swim. TrainingPeaks online determines a TSShr score that I use for HR based clients to keep a track on the Fitness, Fatigue and Freshness. Power is nicer and I even then the meter is sometimes out of action and I have to make a manual entry.

---

Hamish Ferguson: Cycling Coach

Yes, I use it's pace based TSS estimate. The HR estimate doesn't work well for me. My Threshold HR is too variable. I think in training it's usually 165, but I've also run a 77 minute 1/2 marathon at an average of 178. So.... you can see why I'm not a big fan of HR. It would consistently underestimate my TSS for running I believe. I don't use a HRM for swimming. too much of a PITA. Moving pace in reference to a threshold pace is a lot easier.

For me it seem pretty valid, as in terms of overall energy levels, a 100TSS run seems similar to a 100TSS ride, though the leg fatigue itself is higher for the run of course.

---

TrainingBible Coaching
http://www.trainingbible.com

Thanks Andy. I didn't pick that particular study for quality purposes really. It was just the first that pop up on pubmed, and it was just to illustrate 'validity'.
Thanks for the links, I'll go over them.

As a side note, run TSS is pretty easy to estimate. For me at least it's about 8 per mile at a easy run pace. For Zone 2-3 its' 9 and Zone 4 it's about 10. Where the automatic calculation is nice, is in Trianingpeaks at least, it factors in the grade of the road and creates normalized pace. It also calculates "efficiency factor" by comparing HR ot pace as a ratio. I notice mine is all over the place and as a result, I can conclude that pace is a better measurement overall that HR for running.

One other parting note.... when I'm beat down and just not in the mood to train... I don't. It's not a big surprise that this feeling usually correlates to a extended period of low TSS. A large 3 week training block finally caught up with me and I was just crushed.

I will note that my HR is running low, and that itself is a indicator that I'm my fatigue level is pretty high. It funny, because for a long time I would have thought...wow, cool, my fitness must have jumped because I'm running the same pace at a lower HR. Now I realize, especially early in a run, that I'm just fatigued and my heart responds more slowly in that state.

---

TrainingBible Coaching
http://www.trainingbible.com

Yes, what I meant was within the stated framework where only 20 minute duration was looked at. I don't see how you can just blindly pick a duration and try to gauge level of performance from it.

That's just silly. There needs to be a continuous function that is scaled to different durations. You can learn quite a lot by comparing a 20min and 60min TT, but your best 20min within the 60min TT is not important.

I don't know enough about physiology (ie ~zero) to offer an alternative, but it seems like the accumulated fatigue from repeated efforts that approach CP levels in crits or road races are what wear you down... so any semi-accurate measure of stress or performance needs to model the whole spectrum of CPs. For instance if myself and another rider have the same CP20, but I have a higher CP1 and CP5, then the short high intensity efforts within the race will take less of a toll on me.

That's not likely true. I'd actually expect the opposite to be true. The guy with a higher CP20, has a bigger overall engine and can ride at a higher CP60, CP90. The stress would be best measured by normalized power. The guys with a bigger engine can attack more often and will be a better TT rider and best in a break away.. The guy with a better CP1 and CP5, will be a better sprinter. Just different abilities. It also depends on how you train.

remember too, that it's not always the watts you generate attacking the first 1-3 minutes, it's what you can then sustain after the initial attack for the next 10-15 minutes. I would actually think that those with more fasttwitch muscles would recover slower from repeated efforts.

---

TrainingBible Coaching
http://www.trainingbible.com

Thanks for the citation. Refresh my memory. Isn't the impulse-response model used in this abstract the very same impulse-response model you have harangued yours truly (and others) about for years, saying that it isn't useful, that you need to do too many performance tests, that this is why you put forth the performance manager, yadda yadda yadda...

You are going to need to make a decision here. Either the Banister IR model works well as written, using a reasonable number of performance tests, and was appropriately used here to support the validity of TSS, or the bolded results in the above cited abstract are completely meaningless because of all the criticisms you have made of the IR model over the past few years. You really can't have it both ways.

Cheers,

Phil

--

---

Dr. Philip Skiba
Scientific Training for Endurance Athletes now available on Amazon!

Wrong. The IR model has been used to model 5 km time trial performance for running (~17-20 min duration).

Sorry, I assumed that you had access to an adequate number of appropriate performance tests to avoid the model being overparameterized - are you saying that wasn't the case?

Anyway, you're right, the PMC was developed as a more practical alternative to Banister's impulse-response model in part because the latter requires that you do many performance tests of appropriate duration in a short period of time in order to get truly statistically-reliable results. (Even then, it really only tells you when to train, not how/how much to train.) Nonetheless, regardless of any limitations to the modeling approach used, your abstract demonstrates that TSS seems to work reasonably well as an input function.

Citation? That study seems to have slipped my mind.

http://www.ncbi.nlm.nih.gov/pubmed/10029340
http://www.ncbi.nlm.nih.gov/pubmed/2246166

Thanks, I'd forgotten those. Still, it would be interesting to see how well the impulse-response model works with much longer duration events (e.g., marathon, IM triathlon).

Personally, I think the data was just fine. But you tell me, since you gave me the data. To refresh your memory, it was a female track cyclist whom I believe you know very well. ;-)


--

---

Dr. Philip Skiba
Scientific Training for Endurance Athletes now available on Amazon!

I had forgotten about that. So how many performances did you use to fit the model? (Also, why did you never publish this study? I know the attempt to combine it w/ Lindsay's and Simon's results sort of fell apart when they moved on, but it seems you could have followed up independently on this abstract.)

I do a few things with power and heart rate, so do many people who use Dr Coggan's ideas.

I'm more interested in training horses. My daughter is looking at Dr Coggan's ideas with regard to using them for training horses.

I don't see anyone improving on Dr Coggan's ideas with regard to using power in isolation. I just think that more effort should be put into looking at heart rate alongside power.

I do think there are many problems with using power alone to measure training stress. I won't list them all here, but the difference in sustainable power on the flat or track compared to climbing is an example.

Just because I criticise TSS etc does not mean I don't think it is useful or that there is a better way of using power to measure training stress.

But I also think that heart rate is perfectly adequate for measuring training stress, in fact heart rate has some advantages particularly for tri athletes and people who do other sports as well as cycling.

I think it is a mistake to ignore heart rate. If you combine power with heart rate you have more useful data. I don't see it as good science to ignore reliable data.


I was very impressed by Dr Coggan's WKO4 webinars. I'm surprised by how little Dr Coggan's ideas have been picked up on in other sports? Rowing in particular. Also horse racing where pace can now be measured in real time via GPS etc.


Has Dr Coggan looked at horse racing data?

E.g.
http://cyclingtips.com.au/...utputs-are-affected/

"When doing the research, they found found that a rider could ride a TT and a climb with the same power, and same cadence, but there could be a 10bpm variation in heartrate, as well in (sic) differences in oxygen consumption and lactates. It comes down to an athlete’s predisposition of muscle fibre make-up."

So what best describes the stress on the rider in those two scenarios? Power, which is the same in both cases, or HR which I suspect reflects the differences in oxygen consumption and lactates?

The fixation with heart rate is pretty interesting. The body probably does, like, 10,000 things in response to effort...and we just hone in on heart rate so much. I guess because it is easy to measure.

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

The current illogical emotional fashion to ignore heart rate is more interesting.