Page 6 of 11

PostPosted: Sun Jan 11, 2009 1:48 pm
by texmurphy
There has been discussion elsewhere about charting weight with monthly distance. The TrainingLoad Plugin already graphs various weight metrics but does not have a chart for distance.

Seems to me that it would be easy to add a chart showing cummulitive distance, maybe with option to reset distance at each week, or month, or year break.

PostPosted: Sun Jan 11, 2009 2:26 pm
by HENNES
texmurphy wrote:There has been discussion elsewhere about charting weight with monthly distance. The TrainingLoad Plugin already graphs various weight metrics but does not have a chart for distance.

Seems to me that it would be easy to add a chart showing cummulitive distance, maybe with option to reset distance at each week, or month, or year break.


Brilliant idea. I like to build on that: in excel I used to have a "rolling window of sum of last 10d and 28d" also as a measure of "training load". So if the friendly plugin developer picks that idea up, I do suggest to make it configurable like

X days,
X weeks or
since DATE, or
LAST XX days

- that would be phantastic and sth i requested in other threads before.

PostPosted: Sun Jan 11, 2009 6:23 pm
by TX911
gerhard wrote:I have divided the HR zones so each zone is just a couple of beats to reduce margin effects. When studying similar activities (distance, route, TSB, fitness) but different effort (time, average HR), the TSS seem to relate more to the distance than the perceived effort. It therefore seems like I should have steeper factor. I have played around a little with the factors, but getting the exponential curve manually is a little hard.

Can the steepness of the curve be customized or is the factors scientific?


I agree the TRIMP seems to weight time more heavily than intensity. Maybe it's designed for runners who cover more distance per unit time than I do :P

I'm going to play around a bit with the factors myself.

PostPosted: Sun Jan 11, 2009 10:48 pm
by mechgt
gerhard wrote:Exception:

* Click table header - get exception below. If graph is selected instead, table is back


Glad you just posted this... I was just about to release.

I've fixed both the exception and the screwy table thing (I did something dumb and it apparently screwed up the table :oops: !)

TX911 wrote:I agree the TRIMP seems to weight time more heavily than intensity. Maybe it's designed for runners who cover more distance per unit time than I do

I'm going to play around a bit with the factors myself.

gerhard wrote:...
the TSS seem to relate more to the distance than the perceived effort. It therefore seems like I should have steeper factor. I have played around a little with the factors, but getting the exponential curve manually is a little hard.

Can the steepness of the curve be customized or is the factors scientific?

I get the same feeling sometimes... TRIMP wants volume... I don't have an answer for this. It's really hard to get a shorter intense workout to give you the same score as a longer less intense workout.

The calculation is scientific in the sense that it's backed by research papers. It's the only thing I've found from what I would consider to be a scientifically derived resource. If anyone else has information on factors, reply or shoot me a PM or something and I'll take a look at it.
________________________________

Just released:
VERSION 1.2.3
Changes:
- Bugfix: exception when sorting a column on the main page with no activities listed
- Bugfix: empty activity list sometimes when using the back button
- Added ToolTips and ToolTip resources entries for localization
- Allowed for future activities (enable forecasting)
- Added option for category to only filter activity list (not CTL/ATL/TSB charts)
- Added several languages
- Added Name column to ListSettings
- Added DynamicZones setting to calculate TRIMP based on current (at time of activity) HRrest & HRmax.
- Added tooltips and removed info box
- Re-arranged settings page a little


All of the translators out there:
I've added a couple new settings, and tooltips, so there's a few new items on the localization spreadsheet:

http://www.editgrid.com/user/mechgt/Tra ... calization

Fill it in, and use your fav. color to highlight any changes you make... I'll un-highlight them as I get them. Thanks!

PostPosted: Mon Jan 12, 2009 4:07 am
by HENNES
mechgt wrote:Allowed for future activities (enable forecasting)


I usually plan the next week, just with the distances spread across the week and the training explained in the notes section. Looks like:



Image



What does the TL-tool need to start calcs (or maybe what do I need to prevent from looking into the future)? Obviously we need a heartrate, which we dont have for future runs - or?

Should we type in an "average rate" we expect? Would that be sufficient. Or does it need real data minute by minute? Could I copy an old activity into the future with real data and see how the "indicators" perform?

Thanks!

All of the below!

PostPosted: Mon Jan 12, 2009 4:18 am
by Switch
HENNES wrote:What does the TL-tool need to start calcs (or maybe what do I need to prevent from looking into the future)? Obviously we need a heartrate, which we dont have for future runs - or?

You need a duration and an average heart rate. I guess you should be able to include that in your planning.

HENNES wrote:Should we type in an "average rate" we expect? Would that be sufficient. Or does it need real data minute by minute? Could I copy an old activity into the future with real data and see how the "indicators" perform?

See above. Copying the old activity should also work just fine.

Disclaimer: I have not yet installed the new version, but entering data like this already worked in the previous version (except for the "future" part).

PostPosted: Mon Jan 12, 2009 4:30 am
by Switch
HENNES wrote:Is that "in line" with your findings?

Truth is, my findings will only matter to myself and you need to find what is right for you. This plugin is a tremendous help in doing so, and so is the discussion going on around here.

Be changing the ATL- and CTL constants you can practically move your chart all over the place. I think one of the most important parts is finding out, which ATL and CTL settings are right for you and your sport. You are ahead of me there, as you kind of know from previous experience you need three weeks to taper. So your ATL constant should be at 21!
Does that result in the expected taper-date in the plugin?

Switch

PostPosted: Mon Jan 12, 2009 6:16 am
by thorwg
For your next week plan, can’t you just add the planned TRIMP value on the Notes field?

Re from the beginning of this thread: “1) Manual entry: Put the text 'TRIMP=nnnn' in the Notes field, where nnnn is your TRIMP value for that activity.”

I would guess you have a lot of similar runs in the log that could be use to get a good guess of TRIMP vs distance.

Thor ;)

Re: All of the below!

PostPosted: Mon Jan 12, 2009 6:31 am
by mechgt
HENNES wrote:What does the TL-tool need to start calcs (or maybe what do I need to prevent from looking into the future)?

Switch wrote:You need a duration and an average heart rate. I guess you should be able to include that in your planning.

[quote="thorwg"]
For your next week plan, can’t you just add the planned TRIMP value on the Notes field?

Re from the beginning of this thread: “1) Manual entry: Put the text 'TRIMP=nnnn' in the Notes field, where nnnn is your TRIMP value for that activity.â€

Re: All of the below!

PostPosted: Mon Jan 12, 2009 6:43 am
by clackerz
mechgt wrote:
HENNES wrote:What does the TL-tool need to start calcs (or maybe what do I need to prevent from looking into the future)?

Switch wrote:You need a duration and an average heart rate. I guess you should be able to include that in your planning.

[quote="thorwg"]
For your next week plan, can’t you just add the planned TRIMP value on the Notes field?

Re from the beginning of this thread: “1) Manual entry: Put the text 'TRIMP=nnnn' in the Notes field, where nnnn is your TRIMP value for that activity.â€

PostPosted: Mon Jan 12, 2009 6:52 am
by clackerz
Now I need to go away and study all of this info to understand and use this plugin!

I don't race (yet) etc, but I would like to structure some training and use this plugin to see and understand what my training is doing.

Currently I go out and ride - if I feel energetic I ride hard, if I wanna ride long I do, if I'm tired I'll ride recovery.

Having this will assist in developing some meaningful training - even if I'm not training for anything specific.

PostPosted: Mon Jan 12, 2009 8:22 am
by Switch
clackerz wrote:Having this will assist in developing some meaningful training - even if I'm not training for anything specific.

Exactly! I am not training for anything specific either, but I want to be in the best possible shape at the beginning of my next holiday, so I am training towards that date. This will hopefully also help me to figure out more about my own ATL- and CTL-constants.

Switch

PostPosted: Mon Jan 12, 2009 6:57 pm
by TX911
mechgt wrote:I get the same feeling sometimes... TRIMP wants volume... I don't have an answer for this. It's really hard to get a shorter intense workout to give you the same score as a longer less intense workout.

The calculation is scientific in the sense that it's backed by research papers. It's the only thing I've found from what I would consider to be a scientifically derived resource. If anyone else has information on factors, reply or shoot me a PM or something and I'll take a look at it.


I'm not sure how to post a picture, but the following seems to work for me:

%HRR :Factor
0-10: 0
10-20: 0
20-30: 0.3
30-40: 0.5
40-50: 0.8
50-60: 1.2
60-70: 1.6
70-80: 2.2
80-90: 3.5
90-100: 6.5

No science behind it, but the numbers seem to jive better with how I subjectively feel about the run, providing a steeper rise in TRIMP as my HR gets above 80% HRR.

PostPosted: Mon Jan 12, 2009 8:08 pm
by smaryka
Apologies for the long-winded post, but I thought I should chime in with what I've been reading about Intensity Factors for calcuting TSS.

According to Coggan's book "Training and Racing with a Power Meter", the TSS value of 100 was derived as a baseline measure by using a typical cycling activity -- racing a 40km time trial -- done at threshold HR or watts. Generally speaking a good cyclist can do 40km in an hour, hence one hour at threshold = 100 TSS.

Coggan also lists some Intensity Factor (IF) values in his book to multiply against the base TSS for other workouts (either above or below threshold).

Level 1, Active Recovery = 0.75
Level 2, Endurance = 0.75-0-85
Level 3, Tempo = 0.85-0.95
Level 4, Lactate Threshold = 0.95-1.05
Level 5 and higher = 1.05-1.15 etc.

TSS = (s x W x IF)/(FTP x 3600)
where s is seconds, W is watts produced, IF is intensity factor, and FTP is your tested functional threshold power (you can assume that your threshold HR is a roughly equivalent measure to FTP). Multiply the fraction by 100 to get the actual TSS value.

Since mechgt's plugin for Training Load appears to be based on minutes (a factor of 60) rather than Coggan's baseline (a factor of 100) -- i.e., an hour at threshold would count for 60 TSS points in the plugin instead of 100 -- what I did was multiply each of the IF numbers by 1.67 to get my values for my HR chart.
Image

To make things even more complicated for runners, I was reading a bit about rTSS here, http://kriskarsten.blogspot.com/2008/02 ... ormat.html , and the idea that the cyclist's 40km TT is similar to the runner's 10k race. And that time spent at threshold as a runner is harder on your body than as a cyclist so that needs to be taken into account. Therefore, taking my 10km running time as 42 minutes (which is 70% of 60 minutes), I could then divide my calculated cycling intensities by 0.7 to get my running intensities for TSS.
Image

For example, my zone4 has an IF of 1.0 according to Coggan. Translated for the Training Load plugin, my cycling zone4 is 1.67 X 1.0 = 1.67. For running, it's (1.67 x 1.0)/0.7 = 2.39.

Hope this makes sense for people (and I hope I've done the math right! But I did check my results against some 5k running races and 1 hour cycling races and the calcuated TSS scores seem to make sense).

PostPosted: Tue Jan 13, 2009 3:05 pm
by draystex
I've been playing with the TL factors to match my WKO+ CTL/ATL/TSB results based on power. I've got things pretty close which allows me to use one as a proxy for the other when I don't have data (e.g. I only have HR when mountain biking). There's no magic formula, just trial and error and there are outliers but on average I'm getting close results. Two cases where I'm seeing outliers - both relate to the nature of power vs HR and the way TSS/TRIMP are calculated:

1. Short max effort (12x30 sec.) intervals on the trainer - power is easily recorded, while HR never gets a chance to rise given the short nature of the interval. TSS in WKO+ captures the intensity, TRIMP in TL does not. TRIMP < TSS by a significant amount

2. I had a week where my HR just did strange things - I cannot match the WKO+ and TL data at all without upsetting the rest of the data. I'm putting this down to the variability of HR which can be significantly influenced by environmental factors (fatigue, sickness, stress, heat, humidity etc.).

Interestingly, even with these outliers on specific workouts, the over result (as measured by current ATL/CTL/TSB values)is pretty close.

Current WKO+ CTL/ATL/TSB: 55.3/87.4/-32.1
Current TL CTL/ATL/TSB: 55/83/-28

Thought I'd share my results and see if anyone else is trying the same. The result is a stepped curve with inflection points rather than one that is perfectly smooth. Also note that given the time of year, I haven't been doing many workouts that take me above LTHR (161) so above the 160-165 zone has not been tested and compared yet (why those zones are missing).

Factors:

Zone HR Factor
1 0-100 0
2 100-105 0.25
3 105-110 0.3
4 110-115 0.35
5 115-120 0.55
6 120-125 0.7
7 125-130 0.9
8 130-135 1.1
9 135-140 1.3
10 140-145 1.4
11 145-150 1.6
12 150-155 2.1
13 155-160 2.2
14 160-165 2.4
15 165+ 3

Max HR (cycling) = 174, LTHR = 161 Resting HR = 50[/img]

PostPosted: Wed Jan 14, 2009 3:36 pm
by HENNES
So Im reading, reading, reading as probably many people here do now - just to throw in a practical question.

This are numbers from 2 Marathons from me:

Image

Can somebody who is familar with this indicators tell anything from the plain numbers of the outcome of the 2, i.e. which is faster/slower. Or would one need more data before the event to judge?

thanks

PostPosted: Wed Jan 14, 2009 4:01 pm
by smaryka
Well Hennes, you left the min/km part uncovered so I could see which one was faster. :wink:

But I would say that the one with the higher TSB value should have been your better race, just based on your being less fatigued and more well-rested. The one that you ran with a -15 TSB means you probably weren't rested enough -- hence the higher HR yet slower time.

PostPosted: Wed Jan 14, 2009 4:04 pm
by Switch
HENNES wrote:Can somebody who is familar with this indicators tell anything from the plain numbers of the outcome of the 2, i.e. which is faster/slower. Or would one need more data before the event to judge?

My guess would be, that you were faster in the second one. CTL is higher (better trained) and ATL is higher as well (better rested / recovered from your training), resulting in a higher TSB (before).

Oh, and your average speed is higher the second time as well, and the distance was less :lol:

But... Your TRIMP score, your average HR and maximum HR in the second one are lower. So here is a question back at you: how do you explain that?

Switch

PostPosted: Wed Jan 14, 2009 4:58 pm
by Gunnar
I would agree with Switch for the same reasons. And would go on to say that the heart rate and TrImp score was lower because his body was better conditioned, for the same percieved effort.

But, more info prior to the event would give a better indicator.

PostPosted: Wed Jan 14, 2009 11:20 pm
by mechgt
HENNES wrote:So Im reading, reading, reading as probably many people here do now - just to throw in a practical question.

This are numbers from 2 Marathons from me:

Image

Can somebody who is familar with this indicators tell anything from the plain numbers of the outcome of the 2, i.e. which is faster/slower. Or would one need more data before the event to judge?

thanks


I would guess the 2nd:
- CTL higher
- TSB positive (higher)
- ATL being higher here is not relavent I don't think, I might be wrong though; and note higher = MORE fatigued, but I think it's relative to your CTL, thus TSB is the real indicator.

Looks like you had to work harder for a similar result because TRIMP is quite a bit higher (I think fatigue may be the culprit here).

PostPosted: Thu Jan 15, 2009 12:10 am
by racerfern
I too would say the 2nd event was the faster one. If this is a major event for you, then your TSB needs to be higher than -15 at the start. That indicates fatigue (not well rested).

PostPosted: Thu Jan 15, 2009 3:07 pm
by ildibad
problem report :

- while creating future events with TRIMP=xxxx , TSB doesn't accrue. (TSB before = 0)

- dynamic zone doesn't work : it is always using "norm" .

PostPosted: Fri Jan 16, 2009 2:02 am
by HENNES
Before I come back to the marathon puzzle - i just ran into a different question:

See this chart:



Image



I wonder why TSS does not go into the upper part of the scale. Ok, its says HIGH at the "1" and at the "2", but I think those were nearly ultimate session as

1. was a HM ran at PB with averaging 87% HFmax
(see for details in the table underneath the picture)

2. was 35km at 70% HFmax and lastet 3:08

Interesting to learn that both are judged for about the same level - which probably is right, but what shall I do to reach the upper part of the scale if thats not enough?


@mechgt:
BTW: as those numbers in the table and the grey square in bottom right are highly dependend on the user settings, which we are all experimenting with (and changing around all the time) I think it would be a good idea to list the actual settings (ATL CTL) in that grey square, so other people do notice and one can remember self, please - there seems to be enough space. Maybe aditional "data" from settings page, can be displayed there.

And another question...

PostPosted: Fri Jan 16, 2009 3:07 am
by Switch
First things first: respect! Half marathon under 1:30 8)
What strikes me as odd, is the small changes in TSB. Example: 29.11. where TSB only changed 5 points. If I remember correctly (can't check right now) my TSB changes are much higher.
@Hennes: what zone settings are you using?

Switch

Re: And another question...

PostPosted: Fri Jan 16, 2009 4:16 am
by HENNES
Switch wrote:First things first: respect! Half marathon under 1:30 8)
What strikes me as odd, is the small changes in TSB. Example: 29.11. where TSB only changed 5 points. If I remember correctly (can't check right now) my TSB changes are much higher.
@Hennes: what zone settings are you using?


Thanks for the flowers :lol: quite ok for an old elephant!

I read and changed a lot around last days - and as said before - the settings should be in the "picture" as otherwise nobody can analyze the data.

As per "the boss" I set a special HR-zone and tunded that as fine as possible by the automatic split - the factors are the standard as provided by TL.



Image



as you mentioned TSB - here is the chart and more table data.




Image



...going for a 30km run now and already looking forward for that spike :D