## A Rant on Power-Based Training

A light bike doesn't replace good fitness.

Moderator: Moderator Team

basilic
Posts: 785
Joined: Wed Jan 05, 2011 8:05 am
Location: Geneva, Switzerland
this is not quite right.
The various formulas based on age are meant to predict the average maxHR for a given age group, not individual values. There is always variance between individuals of any age. Eg the link below examines measured maxHR of marathon runners, and reports a standard deviation of 10.2 mmHg (table 1) for maxHR, and 8.5 mmHg for the difference between measured maxHR and the value predicted by the Fox formula (220-age) (I am using this paper because participants are all trained athletes, the variance is even greater in the general population). So if the distribution was normal (or close enough), 95% of athletes would be within +-17 mmHg of the predicted value. Eg, if you are 40, predicted maxHR is 220-40=180 bpm, and most athletes aged 40 have measured maxHR between 163 and 197 bpm. Even if you were outside this range you should not be necessarily alarmed or suspect error, since 5% of people will be outside by definition.

https://www.ncbi.nlm.nih.gov/pmc/articl ... -00226.pdf

So if you are going to train based on HR zones, you better get the maxHR right. This means no formulas, do a maximal stepped stress test until your eye sockets bleed or you spill your guts, whichever comes first.

PS: in a post on previous page a iheartbianchi cites a paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4919019/
that concludes thus:
However, Bland-Altman revealed high limits of agreement (upper 25.31 and lower −24.67)
So this is even greater variability, the hypothetical 40 year olds would range from 155 to 205 bpm. The population is not limited to athletes.

iheartbianchi
Posts: 306
Joined: Sun Mar 24, 2019 9:17 am
basilic wrote:
Thu May 23, 2019 9:08 am
this is not quite right.
The various formulas based on age are meant to predict the average maxHR for a given age group, not individual values. There is always variance between individuals of any age. Eg the link below examines measured maxHR of marathon runners, and reports a standard deviation of 10.2 mmHg (table 1) for maxHR, and 8.5 mmHg for the difference between measured maxHR and the value predicted by the Fox formula (220-age) (I am using this paper because participants are all trained athletes, the variance is even greater in the general population). So if the distribution was normal (or close enough), 95% of athletes would be within +-17 mmHg of the predicted value. Eg, if you are 40, predicted maxHR is 220-40=180 bpm, and most athletes aged 40 have measured maxHR between 163 and 197 bpm. Even if you were outside this range you should not be necessarily alarmed or suspect error, since 5% of people will be outside by definition.

https://www.ncbi.nlm.nih.gov/pmc/articl ... -00226.pdf

So if you are going to train based on HR zones, you better get the maxHR right. This means no formulas, do a maximal stepped stress test until your eye sockets bleed or you spill your guts, whichever comes first.

PS: in a post on previous page a iheartbianchi cites a paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4919019/
that concludes thus:
However, Bland-Altman revealed high limits of agreement (upper 25.31 and lower −24.67)
So this is even greater variability, the hypothetical 40 year olds would range from 155 to 205 bpm. The population is not limited to athletes.
I thought we were all in agreement that formulas are off and should only be used as a very rough guide.

If the point is that there are significant differences in average maximum HR by age, we have to be very careful to really think about what this means in real life. If we look solely at the absolute range of maximum heart rates, yes you can come up with some wild gaps (you cited one example of 50...huge). But if you are presented with a 40 year old athlete who says their maximum heart rate is 155bpm, I surely hope your immediate response is not, "oh that's perfectly within the range and so you're fine," even though of course you would be right in saying though. What I hope you would say is, "that's nothing to be concerned about, but it's actually pretty far below the mean line, so let's do another test just to make sure." "Also let me know if you feel unwell or there's something else bothering you." This is I think the prudent and thorough thing to do. Because let's be intellectually honest here - it is much more likely that a 40 year old will have a Max HR of 185bpm as opposed to 155bpm. If you do another test, hopefully supervised this time, and you eliminate all other possible factors, and you still get 155bpm, well that's just where you are and that's that.

It's a similar analysis when medical professionals see a patient presenting with a somehwat elevated heart rate of maybe 75bpm while sitting. This is within the "normal" range, but it would behoove the doctor to suspect an infection or depending on any other symtpoms, some other underlying cause.

As mentioned earlier, Figure 1 in the study I cited shows a very dense number of results close to the mean line, hence the survey results indicate an average deviation from the mean line of 15bpm. This is about what I would normally expect, and anyone who is too far above, or too far below, I would be immediately suspicious. I met a 50year old recreational cyclist who told me his max HR was 210. He showed me his Strava to prove it. I had him change the battery on his HR sensor, checked his meter positioning, had him turn off his phone/bluetooth, other bike sensors etc. to make sure there was no signal interference, made sure he worked up a good sweat to ensure good contact and then had him do the test again. He couldn't get his HR over 175. Yeah, he was bummed out, but at least he no longer was training at HR zones that were wholly inappropriate for him.

Also, as stated in that study, "Leg fatigue was the primary test termination criteria in 82% of the exercise tests" so if you are doing a cycling test, make sure leg fatigue is not coming too early.
Bianchi Oltre XR4
Celeste Matte
Campy SR 11spd mechanical
Bora Ultra 50 tubs
Viseon 5D / stock bits and parts

Bianchi Specialissima Pantani Edition
Campy SR / Chorus 11spd mechanical
Fulcrum Racing Speed 35 tubs
FSA / Deda bits and parts

basilic
Posts: 785
Joined: Wed Jan 05, 2011 8:05 am
Location: Geneva, Switzerland
I agree with much of what you wrote, except
I thought we were all in agreement that formulas are off and should only be used as a very rough guide.
I diagree with the end. Lets not use that at all for any HR based training. Say the 40yo cyclist aims for 60% of maxHR. It could be 98 (from 163), 108 (from 180) or 118 (from 197). Telling everyone to aim for 108 is not good enough. Why bother with an HRM at all.

It's a similar analysis when medical professionals see a patient presenting with a somehwat elevated heart rate of maybe 75bpm while sitting. This is within the "normal" range, but it would behoove the doctor to suspect an infection or depending on any other symtpoms, some other underlying cause.
I hope you just made that up. Normal resting HR (in medical terms) is 60-100. A doc investigating 75 bpm (with no other evidence of a problem), I don't think so.

I met a 50year old recreational cyclist who told me his max HR was 210.
I'm glad it all ended well, but I would not let that that guy near his bike. Send him see a cardiologist. He could be having bouts of VT or other arrhythmia.
Note that the age-based range for this guy would be 153 to 187, so at 210 he is ways above.

iheartbianchi
Posts: 306
Joined: Sun Mar 24, 2019 9:17 am
basilic wrote:
Thu May 23, 2019 1:35 pm

I hope you just made that up. Normal resting HR (in medical terms) is 60-100. A doc investigating 75 bpm (with no other evidence of a problem), I don't think so.
This was a real-life anectode unfortunately. Guy in his 20s went in for a routine physical for his job, very fit guy but had a relatively high resting HR (high 70s). Was suggested to do some blood work and discovered anemia. Did some more tests and was diagnosed with leukemia. This was 20 some years ago. Guy is still alive and kicking today, thank goodness that doctor got suspicious. That guy was my brother. They say your resting heart rate, and heart rate during exercise are incredibly useful indicators of underlying health issues.

Funny story that's not really relevant but just came to my mind - so I became very paranoid about elevated heart rates following the above story. One day a couple of years back my heart was pounding pretty hard (85bpm or so) just sitting down. I got really nervous and went to the ER. Doctor ran a bunch of heart tests, etc., told me I was in perfect health, my pulse was normal, don't waste our time spiel that you typically get in the ER. I put up a fight, saying my resting HR is normally in the 40s/50s and something is not right with me. They reluctantly did a few more tests and I was diagnosed with a respiratory infection.

Regarding the point on the rather wide range of each Max HR zone, I do agree it's not a perfect tool. But as amateurs we are limited by the tools at our disposal. Zone 2 (5 Zone) would be 60-70%, and even the science can't pinpoint the exact "optimal" BPM per zone, if there is any such thing. It's a gradual increase and I think it's close enough for everyday use.

Regarding your final point on the guy with a claimed 210 BPM - if he had continued to show such a high heart rate even after adjusting his HR strap, etc., I would have told him to stop exercise immediately and go see a doctor. But as soon as those adjustments were made, his HR went straight down to reasonable levels. It was a bit strange - I"ve seen HR monitors miss heart beats but I didn't expect them to add heart beats.
Bianchi Oltre XR4
Celeste Matte
Campy SR 11spd mechanical
Bora Ultra 50 tubs
Viseon 5D / stock bits and parts

Bianchi Specialissima Pantani Edition
Campy SR / Chorus 11spd mechanical
Fulcrum Racing Speed 35 tubs
FSA / Deda bits and parts

AeroObsessive
Posts: 115
Joined: Wed Dec 12, 2018 7:42 am
Wow, so much variablitliy and external factors with heart rate. If only there was some device that could consitently measure physical output instead of the response to physical output...

iheartbianchi
Posts: 306
Joined: Sun Mar 24, 2019 9:17 am
Deleted.
Bianchi Oltre XR4
Celeste Matte
Campy SR 11spd mechanical
Bora Ultra 50 tubs
Viseon 5D / stock bits and parts

Bianchi Specialissima Pantani Edition
Campy SR / Chorus 11spd mechanical
Fulcrum Racing Speed 35 tubs
FSA / Deda bits and parts

basilic
Posts: 785
Joined: Wed Jan 05, 2011 8:05 am
Location: Geneva, Switzerland
Wow, so much variablitliy and external factors with heart rate.
The variability doesn't go away if you don't measure it. All physiological parameters that reflect the work being done - HR, lactate, expiratory quotient, etc - will vary, with the intensity of the work but also with other factors. This is not measurement error, it's the real thing.
...measure physical output instead of the response to physical output...
I think the causation here is backwards. The physiological parameters that reflect work being done are not a response to physical output, they are its cause.

Actually I also agree with your (maybe unnecessarily fierce) advocacy for also measuring power output if you want to understand what is going on (not that I have a PM, but i don't train)

@iheartbianchi: thanks for the additional details. This is why sharing a drink on a terrace beats the internet every time.

iheartbianchi
Posts: 306
Joined: Sun Mar 24, 2019 9:17 am
basilic wrote:
Fri May 24, 2019 9:42 am

@iheartbianchi: thanks for the additional details. This is why sharing a drink on a terrace beats the internet every time.
I didn't realize it was such a tough crowd here I have been careful trying to be as precise as possible, but I'm talking on an internet forum with a diverse audience so I've had to either simplify things or convert lessons that are really designed for elite athletes into a practical application for recreational athletes...for example the point of using a Max HR derived from running and using it for cycling is an absolute no-no for elite cyclists. But it works for amateurs as we don't require that level of precision, and the most important thing for us is to go out and ride within reason. There are several other limitations that prevent us from getting an accurate Max HR, FTP or LTHR measurement for a recreational cyclist so we have to creatively look towards other methods. Not to mention, our lactate thresholds and Vo2Max tend to be far below that of elite athletes which further hampers our testing methods. In the end, we're all guessing (even the elites), but our guesswork has become pretty-darn accurate thanks to generations of athletes who have tried and tested these methods!

I didn't think I'd have to treat my posts as research papers But I'm game!
Bianchi Oltre XR4
Celeste Matte
Campy SR 11spd mechanical
Bora Ultra 50 tubs
Viseon 5D / stock bits and parts

Bianchi Specialissima Pantani Edition
Campy SR / Chorus 11spd mechanical
Fulcrum Racing Speed 35 tubs
FSA / Deda bits and parts

onemanpeloton
Posts: 291
Joined: Wed Jul 12, 2017 11:30 am
Location: Edinburgh, UK
Keep the conversation going, I say. Some people seem incapable of rational debate which is a shame. However those same people offer nothing to this conversation other than unnecessary and unsupported nay-saying so they're easy to filter out.
2017 Giant TCR Disc
2015 Giant TCX
2016 Cube Stereo 140

AeroObsessive
Posts: 115
Joined: Wed Dec 12, 2018 7:42 am
basilic wrote:
Fri May 24, 2019 9:42 am
...measure physical output instead of the response to physical output...
I think the causation here is backwards. The physiological parameters that reflect work being done are not a response to physical output, they are its cause.
But that's just it, HR for example, can vary whilst all the while the power to pedals hasn't changed in the slightest. The reasons for this have been mentioned prior. I have seen HRs into 160 and beyond with no physical output at all. So *some* of the measured HR *might* be attributable to power output, except for when it isn't. Of course with training history, multiple data points from a variety of sources can help dial this in and variation can then be more closely attributed.

So HR *might* be a reflection of output, but not always. Whereas watts can only be watts.

spartacus
Posts: 136
Joined: Mon Apr 04, 2011 6:53 pm
AeroObsessive wrote:
Fri May 24, 2019 2:01 pm

So HR *might* be a reflection of output, but not always. Whereas watts can only be watts.
I’m pretty sure that’s exactly what he’s been saying the whole time dude, the point is that your heart rate is a more consistent indication of how hard you are pushing your bodies training stimulus than watts. If your heart rate is at 160 you are going to get more tired than if your heart rate is 120 for the same amount of watts. I guess you’ll never understand at this point though if you haven’t already, lol... sigh..

spartacus
Posts: 136
Joined: Mon Apr 04, 2011 6:53 pm
Accidental post
Last edited by spartacus on Fri May 24, 2019 6:21 pm, edited 1 time in total.

spartacus
Posts: 136
Joined: Mon Apr 04, 2011 6:53 pm
Accidental post

AeroObsessive
Posts: 115
Joined: Wed Dec 12, 2018 7:42 am
spartacus wrote:
Fri May 24, 2019 6:19 pm
AeroObsessive wrote:
Fri May 24, 2019 2:01 pm

So HR *might* be a reflection of output, but not always. Whereas watts can only be watts.
I’m pretty sure that’s exactly what he’s been saying the whole time dude, the point is that your heart rate is a more consistent indication of how hard you are pushing your bodies training stimulus than watts. If your heart rate is at 160 you are going to get more tired than if your heart rate is 120 for the same amount of watts. I guess you’ll never understand at this point though if you haven’t already, lol... sigh..
Oh I understand all too well. The problem is if you aren't monitoring watts then don't know what you don't know. If only monitoring HR you wouldn't perform the above ride because it most likely would be out of the zone. Then one has to questions the effectiveness of training in your above mentioned example.

The easy solution is to run both of course. The more data the better. Hell, I'd run a real time lactate monitor if it were practical.

spartacus
Posts: 136
Joined: Mon Apr 04, 2011 6:53 pm
AeroObsessive wrote:
Fri May 24, 2019 10:59 pm
spartacus wrote:
Fri May 24, 2019 6:19 pm
AeroObsessive wrote:
Fri May 24, 2019 2:01 pm

So HR *might* be a reflection of output, but not always. Whereas watts can only be watts.
I’m pretty sure that’s exactly what he’s been saying the whole time dude, the point is that your heart rate is a more consistent indication of how hard you are pushing your bodies training stimulus than watts. If your heart rate is at 160 you are going to get more tired than if your heart rate is 120 for the same amount of watts. I guess you’ll never understand at this point though if you haven’t already, lol... sigh..
Oh I understand all too well. The problem is if you aren't monitoring watts then don't know what you don't know. If only monitoring HR you wouldn't perform the above ride because it most likely would be out of the zone. Then one has to questions the effectiveness of training in your above mentioned example.

The easy solution is to run both of course. The more data the better. Hell, I'd run a real time lactate monitor if it were practical.
I think you should monitor your wattage for sure, but I think from a day to day training perspective it doesn’t necessarily matter how many watts you’re putting out if the limiting factor is the % of heart rate you’re willing to allow at the time. On the other hand it might be hard to see how fatigued you are without the correlation between hr and power, but it seems like bianchidude’s point is just that HR is more effective for purely training than just a power meter since with just power you’re kind of guessing how hard you’re pushing. Just from my perspective, seems like he’s making a good point and people are piling on him for things he isn’t necessarily claiming.