Rowing Power Profile – Curve After Following Test Protocol

[Lurk]

Hi, I’d like to ask about my Rowing Power Profile in Athletica. I followed the testing protocol exactly as described and completed all the recommended intervals over the course of a week. However, my profile (see attached image) is very “stepped” and doesn’t look smooth as I would expect. I’m not sure if this is correct, or if there might be an error in the calculation/model.

Has anyone else experienced this? Is it possible there’s a data processing issue, or is there something I should adjust in my settings or testing process? Thanks for any advice or experiences!

[Andrea]

Dear @Lurk thank you for your message.

I would say that this behaviour is quite common, especially if you targeted constant powers and all-out efforts.

What we suggest in the Test weeks is to perform these kind of sessions because they minimise the effort of testing. They allow us to detect exactly the “steps” or “deflection points” you mentioned.

Now, with time your profile will become smoother as new data will come in and it will help refine the estimations of the parameters. This is a pretty common procedure as we know that testing has a huge impact on the training preparation. We all digest testing in a different way, and completing a test might be quite invasive. So that is kept to a minimum.

What the AI coach suggest (after looking at your profile), is that the estimated maximal aerobic power is quite solid. However, the maximal power you can deliver for very short durations (<180 seconds) and prolonged duration (1h) might be improved. NOW: whether this is truly reflective of your abilities or it’s just lack of data is difficult for the AI coach to get. This is the best guess and, indeed, it’s asking: “did you perform max efforts between 3 and 5 mins?”. If you did, and you followed the test week to the letter then, in the future, I expect:

1. Your profile will be smoother
2. Peak power values will increase
3. Power delivered during prolonged duration will increase

To better understand whether the profile it’s reflective of your capacities or they are only indicating a lack of data, the AI coach will look at the ratios between MPP/MAP and CP/MAP.

I hope this can help you better understand how the AI coach interprets the profile, and how the profile is used to evaluate your preparation.

[Lurk]

Hi @Andrea
Thank you very much for your detailed reply. I completed the prescribed tests a week ago, and after seeing my profile, I was worried whether everything was calculated and imported correctly – especially whether my training zones are now more accurate.

I noticed that my 1’ ALL OUT test might not have transferred all data properly from the platform (see attached image). I’m unsure if I should repeat this test to ensure data completeness, or if the current data is sufficient. For the 2k and 6k tests, it seems like the data transferred correctly.

Should I repeat the 1’ test to fix the profile, or will additional training data over time resolve this? Thanks again for your guidance!

[Andrea]

Dear @Lurk thank you for your reply.
Unfortunately I do not see your power values for the 1k AO session on the 20th of May. Was this conducted on a different ergometer or with a different device? Anything to highlight?

If you click on “Data Analysis” you won’t see the Power (W) for this session, but you do see it for the other sessions. Did you change something in your equipment? If not, I need to dig deeper and find out why we do not have data and where the Power was lost on the way.

Please let me know!

[Lurk]

Hi @Andrea

It seems the issue occurred due to the app I used for the test. I tried to connect my Concept2 to the EXR app, but the workout didn’t upload properly. During the session, I also recorded the activity on my Garmin, but since it wasn’t connected to my rower, not all data was captured. I also tried transferring data directly from the PM5 monitor, but unfortunately, I couldn’t retrieve all the recorded metrics. This is likely why the data is incomplete.

For the other tests, I used my usual setup without experimenting—specifically, the ErgData app connected directly to my rower. I defined the workout in the app, completed it, and everything transferred correctly.

I’d like to repeat the 1k test using the same setup as for my 2k and 6k tests. However, I’d appreciate some clarification: You mentioned a 1k test, but in the test description, I see a 1-minute all-out effort. Should I set up a 1000m test instead of a 1-minute test?

Thank you very much for your help!

[Andrea]

Sorry @Lurk my bad: 1’ all out. As specified in the card.
I saw you repeated the test today. Now the data has been received correctly.

[Lurk]

Hi @Andrea

Thank you for the clarification. I thought it was probably the one-minute all-out (AO) test. I noticed that on that day I had a HIIT session planned, so I replaced it with this test to have complete data and moved the HIIT to another day. I also see that the AI coach’s message has changed, now suggesting a 30–60 second AO test and two maximal efforts lasting 2–5 minutes. So, I just need to continue training, and the profile will be updated as you mentioned, is that correct?

Additionally, I would like to ask what I should focus on during rowing training. Should I rely on watts or heart rate? Or should I monitor both and try to stay within the defined ranges for HR and watts?

I looked at the planned HIIT session where the 3-minute interval has a prescribed HR of 184–195, but I couldn’t reach that heart rate even during the test Should I always focus on watts for intervals then? Previously, I used to rely on pace, but that is missing here.

Thank you in advance for your response.

[Andrea]

Dear @Lurk sorry for my late reply.

Yes, you’re correct, the power profile should be keep updating, and it should reflect your current training intensities and your abilities.

@Lurk

Should I rely on watts or heart rate?

This is a quite intriguing question, and I go back with this at the HIIT Science book. As you know HR is an internal load metric and the speed/power are external load metrics. Now, how you respond internally to an external load is quite important.

For prescription purposes, if I had an external load metric I’d use that to guide training. I understand you perfectly when you say that your thresholds do not look or feel “aligned”. For us, it became very hard to trust ALL the HR data from every device. Some athletes wear a wrist sensor, which is playing havoc at high intensities.

Main reasons people prescribe training sessions with external load metrics is that:

1. They are the “truth” and materialistic aspect of the performance. As the old saying goes: “No medal has been awarded for a low HR” …
2. HR is delayed with respect to the external load variable (it reacts slowly to variations in power) so it is very difficult to align those from outdoor data
3. Beyond some powers/speed, HR plateaus so there is no point in reading a ratio between power:HR.

“Learning” the relationship and the correspondence between power and HR is also difficult in the lab. For sure you can conduct a test and you find your thresholds HR, but the tests (if done properly) involve ramps shorter than 9-12 minutes. We know that HR can drift with fatigue, heat, and caffeine, etc. This makes the prescription with HR even more problematic.

What I’m trying to say is that, with Athletica, I would trust more a prescription in terms of external metrics (speed/power) rather than HR. You can have a look at HR retrospectively and see if HR matched your expectations. Prescription with internal load metrics such as RPE and HR are immensely useful when:

1. Your purpose is to trigger or protect specifically the internal load, because you’re in altitude, or in the heat, or you know there are reasons to control your HR first.
2. when an external load metric is not available, e.g. trail running or hyrox or XC-ski.

For Athletica, that collects and process data collected in the wild, it becomes very difficult to learn the relationship between power and HR, and to determine a perfect correspondence between zone values. We are trying our best to taylor our algorithms so they can be highly individual, and in teh future we’ll see better aligned thresholds for sure. Until then, my advice is to follow the prescriptions of the external metrics.

Sorry for the long message, I hope this makes sense. Feel free to add whatever you think it’s necessary.

[Lurk]

@Andrea

Learning” the relationship and the correspondence between power and HR is also difficult in the lab. For sure you can conduct a test and you find your thresholds HR, but the tests (if done properly) involve ramps shorter than 9-12 minutes. We know that HR can drift with fatigue, heat, and caffeine, etc. This makes the prescription with HR even more problematic.

What I’m trying to say is that, with Athletica, I would trust more a prescription in terms of external metrics (speed/power) rather than HR. You can have a look at HR retrospectively and see if HR matched your expectations. Prescription with internal load metrics such as RPE and HR are immensely useful when:

1. Your purpose is to trigger or protect specifically the internal load, because you’re in altitude, or in the heat, or you know there are reasons to control your HR first.
2. when an external load metric is not available, e.g. trail running or hyrox or XC-ski.

For Athletica, that collects and process data collected in the wild, it becomes very difficult to learn the relationship between power and HR, and to determine a perfect correspondence between zone values. We are trying our best to taylor our algorithms so they can be highly individual, and in teh future we’ll see better aligned thresholds for sure. Until then, my advice is to follow the prescriptions of the external metrics.

Sorry for the long message, I hope this makes sense. Feel free to add whatever you think it’s necessary.

Hi @Andrea ,
Thank you for your detailed answer. I understand that for performance sessions and intervals, it’s better to use watts as the main guide. I’d like to clarify one more thing regarding aerobic workouts (for example, long endurance sessions in Zone 2):
So far, I’ve mainly used heart rate for these, measured with a Garmin PRO chest strap, and only monitored watts secondarily. According to your answer, though, Athletica still recommends using watts as the primary metric. Is this because even for aerobic training it’s better to rely on external metrics, or is there an exception for these sessions?
Here’s a concrete example:
In my plan, I have a HIIT session with a target HR of 184–195, but I can’t reach that range even during testing. If I follow the watt prescription, I can complete the intervals. On the other hand, during steady-state Zone 2 workouts, I try to keep my HR in the right zone, but my watts are often above or below the suggested range.
Would you recommend combining both metrics for aerobic training, or should I follow the system’s preference? How would you approach aerobic sessions in practice?
I’m attaching a screenshot from yesterday’s aerobic development workout on the rowing machine, where I decided to focus primarily on heart rate.
Thanks for clarifying!

[Andrea]

Dear @Lurk , so sorry for the late reply. I thought about your questions these days and finally I got the time to sit down a bit and write about this.

I think that the way you’re framing the problem is compelling. I totally get what you mean here, and I like the idea of not having strict rules in applying P or HR as an indication. Your idea about following prescribed HR with Z2 - long aerobic sessions and P for high-intensity sessions make sense to me, to be honest.

I had a look at some of your recent sessions, and noticed how you could keep your HR within Z2 at 150 W on average. It looks like power zone Z3a/b allows you to keep your HR under control in Z2. I see a slight drift in HR in those 25’, so I would still take that into consideration (a true Z2 would not see HR drift in 20-25’ I guess). i could see on 6th June there were no primers (as prescribed) in WU, so HR took a little to adapt. Again, interval too short to judge completely if it was a gentle adaptation or a drift for the high W. But in terms of time in Z, we see that you had to make a compromise between following P or HR, and given that this was a Z2 training I agree with you that the HR is what you’re prioritising here.

When it comes at intervals, the HR adaptation/lagging after Z4-5 intervals is much more evident, and easy to understand why you cannot get to 184-195 right away. I would expect 184-195 to be an acceptable range at the end of the sets, but this did not happen, at least on Jun 4th, where intervals were prescribed at 172-176W.

You said “If I follow the watt prescription, I can complete the intervals.” Does this mean that with a higher W you wouldn’t be able to complete the session? It looks to me that you would be able to handle higher W in these intervals, is that possible?

[Lurk]

Hi @Andrea

Thank you for your reply. I’m still experimenting and trying to observe the patterns. Sometimes I try to keep both P and HR in the right zone, and other times I focus only on HR and don’t pay much attention to power. In today’s workout, I noticed the cardiac drift you mentioned—towards the end of the main set, my HR started to rise while my power was dropping. According to HR, I was at the upper edge of Z2, and power was still in Z3 but already approaching Z2. This usually happens after about 20 minutes, as my body gets warmer and the room heats up (I have my rower in a small room without a fan).

Regarding your question about intervals and power: I could definitely hit higher watts before reaching the prescribed HR zone. The more challenging part for me is staying exactly within the narrow power zone—there’s only a 4W range But I assume that for intervals, going above the zone isn’t a problem?

Thanks again and best regards

[Prof]

We also just published on this topic so could be relevant to you @Lurk

Pace vs Power vs Heart-Rate: Pick the Right Training Metric

[Andrea]

@Lurk

The more challenging part for me is staying exactly within the narrow power zone—there’s only a 4W range

Dear @Lurk that is why I am a bit worried your MAP isn’t correctly estimated, and you get too narrow zones. What I would recommend (and I exchanged with @Prof about this) is that you replace a 100m + 1k test (newly introduced in the global library - rowing - tests) with one of your key sessions this week. Perhaps the best chances come tomorrow after resting today (if you are) or Wednesday 25th when you have intervals (optimal choice). We amended the tests for rowing and we included an additional 1k test together with the 100m.

How do you feel about this? Possible? We feel like your rowing profile would benefit from those two tests added to your data records.

To do so: please go on the selected day → Add from library → Select Global Library → Filter for “Testing” Session Aim and “Row” modality → select one available and insert in the chosen day (e.g. Wednesday). Remove the existing intervals session.

Please let us know if you’re willing to follow this possibility.

[Lurk]

@Andrea

How do you feel about this? Possible? We feel like your rowing profile would benefit from those two tests added to your data records.

To do so: please go on the selected day → Add from library → Select Global Library → Filter for “Testing” Session Aim and “Row” modality → select one available and insert in the chosen day (e.g. Wednesday). Remove the existing intervals session.

Please let us know if you’re willing to follow this possibility.

Hi @Andrea ,thank you for your response. I’ll try the suggested test and see how it affects my profile. I’m not sure yet when exactly I’ll do it, but I’ll definitely replace one of my key sessions with it. This week I’ve decided to take things a bit easier—over the weekend I was at a cycling event and I have a bit of a sore throat, so I don’t plan to do any intense training for now, just to be safe.