Me again. I’m partway through my test week, and I have some questions/concerns about the calculation of power and heart rate zones for rowing. For context, I am mostly a cyclist but every couple of years I spend the early winter on the rowing machine when I get a bit burnt out on the bike. I only have a few weeks of “rebuilding rowing form” rowing activities coming into this season.
When I first entered critical power, I used a pretty conservative number that I basically just guessed at (I debated for a bit and don’t remember exactly what I ended up using - maybe 110?), thinking Athletica would quickly update it as I did more sessions. Indeed, I quickly got notified that the smart coach wanted me to update my critical power, so I did, but it is very low(86). I have done some max efforts including a 2k test today, after which it suggested increasing my critical power, but still to a very low number (93).
Every workout I do has a much higher load than prescribed as a result of this very low critical power. So far, it hasn’t made a lot of difference to me as I simply row hard or easy as prescribed, but I was hoping for better guidance for interval targets.
Today I found an online critical power calculator for cycling that used 3 max efforts from the last week that gave a critical power number of 136. I have no idea if that is the same calculation as for rowing, but I believe my zones would be much more in line with my ability if my critical power was higher.
Another issue I am having is with heart rate zones. Age based calculations do not work with me as I am 58 years old with a max heart rate well into the 190s (today I saw 199 during my 2k test). I entered a threshold value of 180, but Athletica has changed it to 159, which results in very low values for me. I’m a little surprised that it hasn’t updated based on observed data during this test week.
I like the idea of an adaptive training plan, but I’m concerned that until the training zones are more realistic, there isn’t much more here than with a paper plan.
thank you for your message and it’s an honour to have an athlete like you on the platform. We are sorry to hear you’re not finding the Athletica experience satisfactory. But let me try to explain why you are seeing those numbers.
CP: we do not consider the <2’ efforts in the CP estimations. I’m not sure what on-line tool you adopted for the estimation, but in our experience 1’ efforts are adding too much noise to the profile, and this variability affects CP estimation badly. In your case, unfortunately, Athletica still misses important efforts longer than 10’ to establish your profile. I understand this might be somewhat disappointing, but it’s the most robust approach our experience can suggest. Please be aware that your 1’ effort is taken into account for in the calculation/estimation of the maximal peak power and the maximal aerobic power, which are considered in the estimation of the 6th and 7th zones. From your profile I can see max efforts clearly at 1’, at 5’, and 8’. The lower points for durations longer than 10’ are taking down the CP estimation. If I’m reading correctly, on Saturday 26th you will have a 6k all-out effort prescribed. Athletica will use that effort to complete your profile. We are quite confident that at that point your CP will settle down to the right place. Please get in touch if this does not happen.
HR: again, our robust/conservative approach has to be “blamed” here. I can see that you can hold your HR>180 for more than 10’, and this clearly indicates that threshold HR is up there where you would expect. Unfortunately, during that key session, HR signal was lost for a period just before the 2k all-out bout. In this case Athletica flags the session, and we do not consider the HR reliable. Again, sorry for this, but in our experience this is a robust approach that can give us the best trade-off between good/bad data inclusion/exclusion (there are pros and cons). I know how frustrating can be to loose data in such an important session, and I’m sure you can already take all the precautions by adjusting the chest belt before the start of the training, etc.
Please, let me know if I’m missing something. You might consider reading this blog post, where we discuss a little bit deeper how the thresholds are estimated.
Again, we are super excited to wrap up your test week and see if Athletica profiling can work for you. We are aware that including all the kind of efforts in CP or HR estimation might provide better reactivity, but we also know that we might end up by overestimating thresholds, which can be more harmful than underestimating them.
Thank you so much for looking into this. I do understand that once I have completed my 6k test tomorrow, my profile should be much better.
I wonder if the way that Concept 2 records rest intervals has contributed to the system thinking it dropped the heart rate signal? I had a look at the fit file and noticed that during a rest interval, the cadence is recorded as the last cadence of the work section, no matter the actual cadence. I’m not sure this has any impact. I also noticed that it seems to record heart rate and power correctly if power is not zero, but if power is zero, it doesn’t record anything. On my last activity, I had a 2:30 rest interval programmed immediately prior to the 2000m all out portion. I rowed very lightly for the first 30 seconds, spent most of the rest period stretching and walking around, then started light rowing again a few seconds before the test started.
Tomorrow I will program the workout so that rest periods are just regular intervals to hopefully avoid setting off any flags.
I’ve just noticed something I assume is a bug - maybe for a new thread, but the distance based intervals like my 6k test tomorrow have a very short duration listed in the workout description. My 6000m has 55 seconds and the 2k was 14 seconds. I expect this throws off the prescribed load calculation for the workout.
Hi @Andrea – if you get a chance could you look at my CP profile. I’ve just gone through the process of manually uploading C2 fit files, replacing the Strava files which were not contributing (at least as noted by the system) to the PD curve.
However, once importing the system notified me of a significant drop in CP estimate (28w). The estimate prior to the import of the fit files was in-line with WKO’s estimate of FTP (with a TTE of 33 minutes). As I understand it, CP is associated with 30’ power so the previous CP made sense. The after-import estimate is more in line with 60’ power.
After inspecting your profile, with the data I see, I must admit that the CP estimation is solid (please notice, with the data I see).
I can see max efforts such as:
274 W at 3’50"
206 W at 12’
166 W at 33’
165 W at 1h
As you can see the number is settling down to 165 W so I might say that CP is around that figure.
As far as I understand, FTP is associated to the 1 hour effort, not 30’ (I might have misinterpreted though). A shortcut for the 1 hour test is the 20’ test and, in cycling, you can take 95% of that to estimate FTP. To me, CP is more activity-agnostic than FTP. In this thread an example of how different athletes report different %s for FTP estimation.
Besides the numbers, I would ask your feelings and sensations when cruising at 170/180 W and 130/140 W. If the profile indicates CP at 166 W of course it does not mean that the world ends beyond that point: I hope you can appreciate that there might still be a gray area where things are progressively changing somehow.
If you had to select the power for an hour record, I would keep that in mind and you will probably start oscillating a bit around the 166 W number, but in terms of training below (Z3) or above (Z4), that number will demarcate the separation between distinct adaptations AND sensations. For example, if we were sure that 166 W was reliable, you would feel and perceive quite differently at 145 W vs 180 W.
Thoughts? Happy to delve more into your data if needed.
Thanks for the reply! I would agree 166w is probably pretty close to one hour power. I just always associated CP with a shorter duration, more like 30’. This is why I asked. For example, Skiba states:
We take your best performances over 2 – 4 efforts that last between 2 and 20 min…CP is close to the best power you can hold for 20 – 40 min (for most of us).
High North also indicates CP is around 30 minutes. It sounds like Athletica is calculating one-hour power – which is fine – but I would associate this with maximal lactate steady state (MLSS), not CP.
@Fazel , a while back I read Skiba’s book “ Scientific Training for Endurance AthletesNew Book / E-Book! and your post (good post) caused me to go re-read his chapters on Anatomy & Physiology as it relates to CP. My recollection was he suggested caution on establishing an absolute time period behind the concept of Critical Power, unless it was so labeled (e.g. CP 30, CP 60).
On pg 46-47, he does discuss two-parameter CP model tests for cycling, running and swimming that involve both a 3-min and 20-min to be done on different days; on pg 48, he discusses the three-parameter model and points the reader to his Chapter 7.
The real meat of his discussion comes in Chapter 7, Engineering Training And Racing, where he really gets heavily into math and advanced application of it.
@BevOutside , I just noticed you remarked in your original post back in October about max HR. I’m a 69-yr old erg rower you doesn’t use the generic max HR calculations due to their general age bias.
When I was last on-the-water (18 mos. ago ), the members at my local rowing club suggested using the following erg rowing-based test protocol from Rowing Australia to help estimate my rowing max HR, sans the use of O2 uptake and lactate testing procedures mention in the document.
A side benefit is it is one heck of a multi-interval workout…….
If my max heart rate is higher than what I’ve already observed, I don’t want to know it! As it is, I just use 190 (but I’ve seen 199 this year) in calculators, because it actually makes very little practical difference in zone calculations anyway. I’ve pretty much given up on Athletica modelling as it seems determined that I am rather a generic dishrag version of myself.
It’s an interesting book! I wondered…its your quoted text that caused me to go back and re-read the three chapters…I needed to wrap my head around it all in full context.
I agree….at 69, I just wanted to get close to a target of when to back-off to give my better half some assurance I would back off my compulsive tendencies when I workout.
Hey @BevOutside … sorry to hear this. Could you kindly elaborate for us so we can consider how we might improve in the future? For reference, our current zone models are here:
I think Athletica’s zones 5-7 are really helpful, but the middle zones are confusing – specifically the apparent conflation of critical power and one-hour power.
I am unaware of any critical power model that takes into account efforts beyond 30’ (please correct me if I am mistaken); the vast majority suggest two-to-four efforts between 3’ and 20’.
The benefit of using CP is that it is based on academic work, as it demarcates the boundary between the heavy and severe domains (preaching to the choir here). Threshold is somewhat more nebulous or ill-defined, which may be why WKO has added time-to-exhaustion to its mFTP.
Athletica presents its calculation as CP, but appears to calculate and apply one-hour power instead (CP60/MLSS?).
Intervals.icu row CP: 190w (3 parameter)
WKO row modeled FTP: 191w with a time to exhaustion of 33’
Athletica row CP: 165w
Finally, Athletica’s zone four is labeled threshold but placed above CP (in the severe domain). This is confusing for those who have the understanding I’ve described here.
Dear @Fazel thank you so much for your insightful posts.
I’m afraid I do not quite understand. The definition of CP in the CP model is of an asymptote, not a value for the power. There is a difference between:
The data you use to build (fit) the CP model
The power you consider as a proxy for CP (the physiological threshold)
I’m afraid there is no workaround here, the CP is defined when the power profile becomes flat within that hour period. Then, if we want to debate wether a CP30 or CP60 is better associated with that heavy-severe intensity domain it’s fine to me, but from a physio/math standpoint, the CP is an asymptote. This consideration is based on the very definition of CP, not about further adaptations implemented by Skiba and others.
We can also ague that (to build the profile, Point 1) some durations are better than others for different modalities and fitness levels. Also debatable and subjective.
When I inspected your profile I saw 166 W at 30’, so how could it be CP30=190 W with other platforms? I’m afraid I do not understand. Either Athletica is not seeing some of your sessions, or I am missing the point entirely.
Follow up: I can now see that your values have been changing so I do not quite know if this post might be useful for you.
Hi @Andrea and thank you for the reply. I’m not trying to be difficult and I am certainly not an expert. My CP has been changing but not because of legitimate efforts, but as a result of funky data provided by C2 (not Athletica’s fault).
Then, if we want to debate wether a CP30 or CP60 is better associated with that heavy-severe intensity domain it’s fine to me, but from a physio/math standpoint, the CP is an asymptote.
This is fair, but I don’t want to argue . I actually find CP60 better to be honest, I just am used to software using durations shorter than 20’ and calling it a day.
I’m going to do a 30’ effort and see what happens.
I held 194w for 30’, deleted the row with bad data and Athletica now has CP at 178w. This is much better and I think probably spot on for around 50-60’ max effort. Thank you!
.
), the members at my local rowing club suggested using the following erg rowing-based test protocol from Rowing Australia to help estimate my rowing max HR, sans the use of O2 uptake and lactate testing procedures mention in the document.
I wondered…its your quoted text that caused me to go back and re-read the three chapters…I needed to wrap my head around it all in full context.
