Forum rules
We spend our leisure time to make RUNALYZE as good as possible and we still have tons of great ideas, but time is scarce and after a hard training we first need some fruits or a smoothie. You can support the development of RUNALYZE with every financial contribution — We highly appreciate every grant.
https://blog.runalyze.com/donate/
aldipower
Neuling
Posts: 8
Joined: Wed 11. Apr 2018, 15:53

More calculations and values from powermeters

Thu 7. Jun 2018, 15:28

Hey!

First of all I would like to thank you for the Runalyze-shirt I've ordered.
It's the best shirt I have for the summertimes because of it's whiteness and and it doesn't rubs my nipples! :D

As power meters for running are upcoming and will be probably the new deal of the next years I would be able to still use my favorite analysis Software - Runalyze - for it. :)
Currently there is the power displayed in a graph together with average power and normalized power (you call it specific power), but the is a lot more you can do, for instances a form graph based on power.
Let me explain some values that could be useful in Runalyze for further analysis of my runs. These values are well explained in Jim Vance book "Run with power". Currently I need to go to trainingpeaks.com to analyze meaningful based on power.

rFTPw - Running Functional Threshold Power
This is the highest power average you could deliver over an hour.
Based on this value, each athlete has to determine by itself, you are able to calculated further intensity values.
This value is much more important then the HFmax value and also valuable to set your power zones already available in Runalyze.

NP - Normalized Power
This is the average power (AP) cleaned from unusual peaks or groves and anomalies. This treats intervals or hilly trainings much better.
To be honest I have no idea how to calculate this. :)

IF - Intensity Factor
Based on the rFTPw and the NP you can calculate an intensity factor.
IF = NP : rFTPw
A factor of 1 would be near your anaerobe power threshold.
This could also displayed as a percentage value in the data fields.
The IF could be valuable for the power zones.

VI - Variability Index
The VI shows how evenly and constant the power of a run was.
A run in a flat area has a very constant power distribution compared to that a run in the mountains varies a lot.
VI = NP : AP

TSS - Training Stress Score
The Training Stress Score is somehow comparable to the TRIMP but much more accurate.
It's calculated like this
TSS = ( (Duration of training in seconds * NP * IF) / (rFTPW * 3600s) ) * 100

kJ - Kilojoule

I can imagine this as a lot of work to do to integrated this in Runalyze. But I think it's really worth it, because it gains a lot of possibilities in running analysis and makes Runalyze future proof. Not to many analysis platforms have this for now, but they will.

Thanks again for this nice peace of software.

Beste Grüße aus Hamburg-Langenhorn! :)
Aldipower
1 x

User avatar
laufhannes
Core developer
Posts: 728
Joined: Mon 29. Jul 2013, 20:59

Re: More calculations and values from powermeters

Thu 7. Jun 2018, 15:40

In theory, that's on our list. We still need to clarify some things for that, especially because most of those metrics are registered trademarks of Peaksware, LLC. And our todo list is really long - so we can't announce any date when we'll dig deeper into power analysis.
0 x
User help -- short questions via Twitter, Facebook

aldipower
Neuling
Posts: 8
Joined: Wed 11. Apr 2018, 15:53

Re: More calculations and values from powermeters

Thu 7. Jun 2018, 15:53

Yeah I see. Kind of funny (or not) how simple formulas like IF = NF : rFTPw could be trademarked. :-O Maybe it's just the name that is trademarked?
For the beginning it just would be helpful to set my rFTPw - which isn't trademarked as I can see, it's a technique - somehow manually and then be able to display a percentage based on the average power in the data fields similar to the HFmax% ?
rFTPw% = AP / rFTPw * 100
This would be helpful to get an idea of my training stress.

Best
0 x

User avatar
laufhannes
Core developer
Posts: 728
Joined: Mon 29. Jul 2013, 20:59

Re: More calculations and values from powermeters

Thu 7. Jun 2018, 16:31

Yes, mainly the names are trademarked. But the formulas are kind of useless if we can't name what they express. SportTracks just uses different names for those metrics (but clearly states the original metrics and their trademarks in their glossary) - I don't know if that's a good way or if they have an agreement with Peaksware. I guess we go instead for Dr. Skibas metrics (that's what GoldenCheetah does, but we'll probably contact him first).

Concerning (r)FTP: As far as I get, there are a lot of different models for estimating this and probably most of those models are not published?
0 x
User help -- short questions via Twitter, Facebook

aldipower
Neuling
Posts: 8
Joined: Wed 11. Apr 2018, 15:53

Re: More calculations and values from powermeters

Thu 7. Jun 2018, 16:59

Yes, fair enough, finding relevant names should be doable. :) The GoldenCheetah metrics are doing the trick too of course.

Estimating the rFTPw is fairly simple, but not easy, hehe.
There exists multiple methods.

1. 3/9-Test
15min warm-up
3 minute interval with maximum intensity
30min recovery mixed with walking and slow jogging
9 minute interval with maximum intensity
Cool-down

rFTPw = (3min power + 9min power) / 2

2. 30 minutes at max
Warm-up
30min at max intensity
Cool-down

rFTPw = Average power of last 20min

3. Race
Take a race on flat terrain that lasts around one hour and try to guess the rFTPw on your average power of the race.

Theses methods are also used by the Stryd Powercenter and are described in "Running with power" too.
0 x

User avatar
laufhannes
Core developer
Posts: 728
Joined: Mon 29. Jul 2013, 20:59

Re: More calculations and values from powermeters

Sat 9. Jun 2018, 15:14

So that are all manual methods and the athlete does these tests himself and just enters his rFTPw the same way he enters his HRmax?
0 x
User help -- short questions via Twitter, Facebook

aldipower
Neuling
Posts: 8
Joined: Wed 11. Apr 2018, 15:53

Re: More calculations and values from powermeters

Mon 11. Jun 2018, 14:04

Yes, exactly, I would suggest a manual input field for the rFTPw.
I guess there will never be _the_ method to determine it, it is always a little bit blurry.

Possibly in the future it maybe would be nice to have something like a conversation table from VDot <> Power <> Pace.
At the Stryd Powercenter for instance I am able to type in a 10k race result pace.
42:32min results in a rFTPw of 265 Watt. But not sure how accurate this is..
0 x

aldipower
Neuling
Posts: 8
Joined: Wed 11. Apr 2018, 15:53

Re: More calculations and values from powermeters

Fri 22. Jun 2018, 11:15

BTW, normalized power is calculated like this and should be easy to integrate:

NP = fourthRoot(sum30secPowerSamples(30secSample^4) / samplesCount)

https://www.mathworks.com/matlabcentral ... ized-power

So if you compare NP to avarage power you can get easily an idea of the training variability.
1 x

User avatar
worstje
Neuling
Posts: 8
Joined: Wed 29. Aug 2018, 09:09
Location: Berlin/Germany

Re: More calculations and values from powermeters

Thu 30. Aug 2018, 16:11

Dear Runalyze Crew,

I am also highly interested in having NP and the related power metrics calculated in Runalyze. The simple equation provided by aldipower should do the job for NP. All other equations are straightforward to calculate (cf. https://medium.com/critical-powers/form ... 295c661b46).

Further, having the aforementioned metrics available in the weekly and monthly views (and trend assessments) would allow anyone to avoid time-consuming spreadsheet calculations or individual coding to analyse and understand the long-term performance.

Instead of TSS, I recommend to consider the STRYD RSS for running activities. I can provide the link to the related white paper if required. Consequently, one could alternatively use RSS to calculate CTL, ATL and TSB to become independent of the HR-related uncertainties.

The trademarked metrics may be simply renamed as done in Goldencheetah (e.g. xPower for NP, etc.).



Thanks and best regards, worstje
1 x

Return to “Ideas”

Who is online

Users browsing this forum: No registered users and 1 guest