Energy Cost for cadence/power combo

Post a question about the software or find answers here.

Energy Cost for cadence/power combo

Postby the5krunner » Mon Mar 09, 2015 6:11 pm

Is the quadrant analysis tool (or any other in ST) able to tell me the HR cost of each cadence/power combo? ie to tell me what is the best cadence for each level of power at present.

I guess something similar to how Quadrant Analysis is now but the dot's colour would vary by cycling HR zone.

I've been trying to train my cadence up to the high 90s and I can operate at those cadences fine. Just that races are coming up now and I would like to know if I've been wasting my time and what is the best cadence(s) to use. I appreciate the argument is more nuanced if the prospect of a subsequent run comes into the equation
the5krunner
 
Posts: 458
Joined: Sun Feb 22, 2009 1:41 pm

Re: Energy Cost for cadence/power combo

Postby tndmbkr » Tue Mar 10, 2015 1:58 pm

I have been thinking about this subject too.
But I concluded I had to consider too many variables in my analysis.

Power = Pedalforce * Cadence
I have trained all my life with cadence=105.
My body had become habituated to the corresponding pedalforce.
When I started training with cadence=90. I hurt all over.
I could not sustain this higher pedal force over a satisfactory period of time.
My fitness disappeared.

You can only compare and Power and Cadence and HR in one chart, when you have equal fitness before and after the change in cadence.

But how does one establish ones fitness in this case?
TRIMP is established bij measuring your HR.
When I ride with cadence=105, my HR is substantially higher than with cadence=90
(for the same output of power)
So my fitness, measured in TRIMP, is lower.
Even if I could have sustained my efforts sufficiently long.

My power output did not change.
But I could not sustain the lower cadence sufficiently long.
So my fitness, measured in TSS, is lower.

======================
In the past few months I have gathered tons of bookmarks related to this subject.
Very often the research data could not be considered as reliable because the researchers demanded efforts from the athletes to which they were not accustomed to.
Maybe you are already familiair with this contribution ?
http://home.trainingpeaks.com/blog/arti ... t-analysis
tndmbkr
 
Posts: 233
Joined: Wed Jan 11, 2012 2:39 am
Location: Rotterdam

Re: Energy Cost for cadence/power combo

Postby the5krunner » Tue Mar 10, 2015 5:09 pm

hi

thank you for your considered and detailed reply. it's appreciated.

I was initially enquiring for a friend who trains at 95 (because I tell him to ) though he invariable slips to lower and substantially lower speeds.

what I really want to do IN ONCE SESSION OR RACE show what I have asked form above.

I appreciate there are many factors eg over a 90 minute race HR will tend upwards for a given power level.

Then I thought it might be nice for me to know as well :-)

I will read the TP link later. ty
the5krunner
 
Posts: 458
Joined: Sun Feb 22, 2009 1:41 pm

Re: Energy Cost for cadence/power combo

Postby tndmbkr » Sat Mar 14, 2015 4:20 pm

Apart from my theoritical considerations (see above), TrainingAnalysis can give us more insight in our power/cadence combo, than it does now.

TrainingAnalysis knows a tab "Critical Power" which shows us how much power we could produce for xx minutes.
The chart shows for every CPxx Power, NP, EF and Heart Rate.
Unfortunately it is not possible to define our own List Settings.
I would be very pleased if OMB could add a column for cadence.

I hope this chart can help us to get a better understanding of our personal cadence/power combo, when the right axis shows cadence instead of heartbeat.
Image

The average value of the cadence for every CPxx can already been looked up in the tab OtherTracks.
But it would be more convenient to have that information available in the tab CriticalPower.

When you click on a CPxx in the tab CriticalPower, and switch to the tab Correlation,
you can also see the average cadence with which you have obtained your CPxx (the correlating points will dash)
In my case the average value does not mean very much.
I do not see a clear-cut link between my pedal force and my cadence in the tab Correlation in the various phases of my activity.
But we are all different.


The ActivityDocumentation plugin makes it possible to compare ones CPxx between different activities.
Unfortunately the plugin does not make it possible to compare HR, and cadence for every CPxx.
tndmbkr
 
Posts: 233
Joined: Wed Jan 11, 2012 2:39 am
Location: Rotterdam

Re: Energy Cost for cadence/power combo

Postby old_man_biking » Wed Mar 25, 2015 2:38 am

The original post is in fact asking for a multivariant analysis of 3 parameters: power, cadence and heartrate.
Training Analysis plugin provides splatter charts to analyze the correlation of just 2 parameters, and gives you a linear regression like
Heart Rate(Power) = 128bpm + 0.06bmp/W

How can you use that to get a 3 dimensional regression?
The latest versions of my Custom Data Tracks plugin let you create your own calculated tracks.
The latest version of Training Analysis plugin let you use custom data tracks for correlation analysis.

1. create a track named "HR per power ratio" with the calculation formula "{DataPoint.HeartRate}/(128.0 + 0.06*{DataPoint.Power})".
Of course you will take the values for displacement (128.0) and slope (0.06) from representative activities done by YOU.
What does this track mean? The denominator is the expected heart rate, according to the linear regression heart rate vs. power.
So whenever your actual heart rate equals this expected heart rate, the value of your track is 1.0. When your heart rate is less (i.e. your "engine" is under less stress than you would expect according to your power output) the value is below 1.0, ...
2. go to the "Training Analysis" page of an activity, tab "Correlation" and select "Cadence" for x axis, your calculated track for y axis.
That'll tell you if cadence has an influence on your heart rate per power ratio.

If I were one of the "big players" in the market, I'd give this approach a fancy scientific name and get a patent for it :)

Try it out.

Cheers,
OMB
old_man_biking
Donated!
Donated!
 
Posts: 6513
Joined: Mon May 28, 2007 8:07 am
Location: Munich, Germany

Re: Energy Cost for cadence/power combo

Postby texmurphy » Wed Mar 25, 2015 9:14 am

old_man_biking wrote:...If I were one of the "big players" in the market, I'd give this approach a fancy scientific name and get a patent for it :)


You are a big player...
google search: old man biking wrote:About 3,860,000 results (0.35 seconds)
Search Results
Old Man Biking's SportTracks Plugins for SportTracks ...
texmurphy
Donated!
Donated!
 
Posts: 2126
Joined: Wed Jul 05, 2006 7:38 pm
Location: Maryland, USA

Re: Energy Cost for cadence/power combo

Postby the5krunner » Wed Mar 25, 2015 11:40 am

old_man_biking wrote:
If I were one of the "big players" in the market, I'd give this approach a fancy scientific name and get a patent for it :)



although if I were their lawyer I would point out that the method is already public domain and therefore unpatentable.

anyway, I digress. I will try it out. :lol:
the5krunner
 
Posts: 458
Joined: Sun Feb 22, 2009 1:41 pm

Re: Energy Cost for cadence/power combo

Postby tndmbkr » Fri Apr 17, 2015 2:31 pm

Because I have got to deal with a left/right balance problem, I have created this chart in CustomDataTracks.
In this chart I can see whether a high cadence influences the power generated by the left leg in a positive or a negative way.
Image
The purple line is the pedalforce left leg.
The red line is the power generated by the right leg.
When the red line is over the purple line, the proportion between pedalforce and cadence is average.
When the red line is above the purple line, I have put more than average emphasis on a high cadence.
power=force*cadence.

Note: In order to keep the chart readable, I did not include the cadence in the chart.
But of course you can do that for your own activitytracks.
================================
In order to have the purple and red line close to each other, requires some zooming.
It is possible to zoom in on the purple line, and zoom out on the red line, independently to each other.
It is also possible to move the red line up, and move the purple line down, independently to each other.
viewtopic.php?p=32504#p32504
viewtopic.php?f=37&t=15785

================================
The formulas for leftpedalforce and leftpower can be found here:
viewtopic.php?f=24&t=16186#p85011
Last edited by tndmbkr on Fri Jun 12, 2015 12:22 am, edited 5 times in total.
tndmbkr
 
Posts: 233
Joined: Wed Jan 11, 2012 2:39 am
Location: Rotterdam

Re: Energy Cost for cadence/power combo

Postby admin » Fri Apr 17, 2015 3:30 pm

amazing. 8)
Get the latest info: Fan us on facebook or follow us on twitter
admin
Site Admin
Site Admin
 
Posts: 3013
Joined: Tue Apr 05, 2005 8:52 pm
Location: USA

Re: Energy Cost for cadence/power combo

Postby the5krunner » Fri Apr 17, 2015 6:13 pm

tndmbkr wrote:Because I have got to deal with a left/right balance problem, I have created this chart in CustomDataTracks.
In this chart I can see whether a high cadence influences the power generated by the left leg in a positive or a negative way.
Image
The purple line is the pedalforce left leg.
The red line is the power generated by the left leg
When the red line is above the purple line, the bikespeed benefits from a high cadence, because power=force*cadence.

Note: In order to keep the chart readable, I did not include the cadence in the chart.
But of course you can do that for your own activitytracks.

The formulas for leftpedalforce and leftpower can be found here:
viewtopic.php?f=24&t=16186#p85011


(edit) did it tell you your 'optimum' cadence - different to what you thought?
Last edited by the5krunner on Sat Apr 18, 2015 5:56 am, edited 1 time in total.
the5krunner
 
Posts: 458
Joined: Sun Feb 22, 2009 1:41 pm

Re: Energy Cost for cadence/power combo

Postby tndmbkr » Sat Apr 18, 2015 2:46 am

the5krunner wrote:.....did it tell you your 'optimum' cadence?
what's on the X-axis...time?

I do not think there is an optimum cadence for a ride as a whole.
A ride is an addition of many short efforts, because of windgusts, bridges, etc
During short efforts you can put more emphasis on high pedal force, for a high poweroutput.
In between these short efforts you have got to "relax" in a high cadence.

The chart above makes it possible to look for stretches where I preferred a high cadence, and stretches where I preferred a high pedalforce. How long are these stretches? Are these stretches well balanced?
The chart above also made me clear there is a transition from high to low cadence during an interval. Where, when and why did this transition occur?

For a ride as a whole, I have never noticed a clear-cut link between my poweroutput and my cadence.
My correlationschart is always very scattered.
With this chart I can split my activity into comparable parts, and get a better understanding of the correlationschart.

OMB has explained us how to correlate power, cadence, and HR.
I select a part of the chart above in CustomDataTracks.
I make a note of this selection with StickyNotes (TrackColoring plugin).
The streetmap helps to remember me what happened on different moments.
I switch to the Correlationtab of TrainingAnalysis, and I see the relevant correlations blinking for the selected part of the activity.



=================
The X-axis in the chart is elapsed time.
It shows an effort for 5 minutes
tndmbkr
 
Posts: 233
Joined: Wed Jan 11, 2012 2:39 am
Location: Rotterdam

Re: Energy Cost for cadence/power combo

Postby tndmbkr » Wed Apr 22, 2015 11:56 am

the5krunner wrote:...............(edit) did it tell you your 'optimum' cadence - different to what you thought?

I am sorry I do not have one answer.
I am still trying to get a better understanding of my Power/Cadence/HR combo for short efforts and long efforts.

Have you ever considered using the MeanMax plugin by Mechgt?
http://www.zonefivesoftware.com/sporttr ... p=mean-max

At this very moment I am looking for (parts of) activities that can be of use for this analysis.
But I have not discovered my "favorite" combo yet.

Below is a typical chart for a low effort/long distance activity.
But I have not found higher efforts near lactate treshold with an obvious chart, yet.
Image
Last edited by tndmbkr on Thu Apr 23, 2015 1:27 am, edited 1 time in total.
tndmbkr
 
Posts: 233
Joined: Wed Jan 11, 2012 2:39 am
Location: Rotterdam

Re: Energy Cost for cadence/power combo

Postby admin » Wed Apr 22, 2015 12:26 pm

Does mean/max plugin do plots for segments, or just the entire workout?
Get the latest info: Fan us on facebook or follow us on twitter
admin
Site Admin
Site Admin
 
Posts: 3013
Joined: Tue Apr 05, 2005 8:52 pm
Location: USA

Re: Energy Cost for cadence/power combo

Postby texmurphy » Wed Apr 22, 2015 1:22 pm

admin wrote:Does mean/max plugin do plots for segments, or just the entire workout?

Entire activity, although you could split the activity into separate segments.

Mean-Max lets you add Heartrate and Grade to the plot as well as Cadence.
Note the abscissa is a log scale.
texmurphy
Donated!
Donated!
 
Posts: 2126
Joined: Wed Jul 05, 2006 7:38 pm
Location: Maryland, USA

Re: Energy Cost for cadence/power combo

Postby admin » Wed Apr 22, 2015 1:56 pm

texmurphy wrote:Entire activity, although you could split the activity into separate segments.


Yeah... that's what I thought. It's a bit of a limitation for peak analysis on roads, given grade changes. Lots of apples and oranges and bananas tossed in the same bowl to compare.

For anyone using the SportTracks web app I'll also mention that it includes segmented critical power plots, with cadence (and HR) overlays, and also a correctly labeled (eg. "useful" :) ) log scale on the x-axis. The plots are interactive, so you can drag across segments for a specific hill, or high-intensity interval on a trainer ride.

Post about the feature from last year:

Power has never been so critical for cyclists

I'm not up to speed on every software app out there, but I think it's fairly unique. I'm sure "the usuals" do it (WKO, GC, CA).

Cheers!
Get the latest info: Fan us on facebook or follow us on twitter
admin
Site Admin
Site Admin
 
Posts: 3013
Joined: Tue Apr 05, 2005 8:52 pm
Location: USA

Re: Energy Cost for cadence/power combo

Postby tndmbkr » Thu Jun 11, 2015 3:27 pm

old_man_biking wrote:The original post is in fact asking for a multivariant analysis of 3 parameters: power, cadence and heartrate.....
How can you use that to get a 3 dimensional regression?....
1. create a track named "HR per power ratio" .....
2. go to the "Training Analysis" page of an activity, tab "Correlation" and select "Cadence" for x axis, your calculated track for y axis.
That'll tell you if cadence has an influence on your heart rate per power ratio......

I have analysed some similar activities (interval 90% on a flat road).
And found the correlation below, more than once:
Image
In the chart the range of my cadence was between 95 and 105 rpm
I could not sustain my power for a longer period of time, when I tried to ride with 90 or 110 rpm.
Because I had to "survive" wind gusts, short ramps, etc
Did I have to try harder to get accustomed to other cadence?

I wondered whether other readers of this post have managed to change their cadence/power combo on the basis of this analysis
tndmbkr
 
Posts: 233
Joined: Wed Jan 11, 2012 2:39 am
Location: Rotterdam


Return to Questions

Who is online

Users browsing this forum: No registered users and 2 guests

cron