Weekly ATL, CTL, and TSB calculations

[Mike McLoughlin]

does anyone know how GC does the weekly ATL, CTL, and TSB calculations?

looking to do an excel version but i can't figure it out.

Cheers

[Mark Liversedge]

They are averages for the 7 days.

Cheers,

Mark

[Mike McLoughlin]

ha ha ha, there i was thinking it would be a weighted progression or something complicated

Cheers

[Mark Liversedge]

We're not so clever :)

[Alexandre Prokoudine]

Speaking of which, I see something fishy in TSB calculations in GC.

According to http://help.trainingpeaks.com/hc/en-us/articles/204071764-Training-Stress-Balance, TSB = CTL - ATL.

However, here's what I see for some of my recent running workouts:

14 = 15 - 3
12 = 15 -4
11 = 14 - 5
9 = 14 - 6

There are more dramatic examples, e.g. 3 = 26 - 36, or 3 = 15 - 23

I understand that some rounding is invoilved, but surely -10 cannot be rounded to 3?

Alex

[Mark Liversedge]

On Wednesday, 11 November 2015 08:08:54 UTC, Alexandre Prokoudine wrote:

Speaking of which, I see something fishy in TSB calculations in GC.

According to http://help.trainingpeaks.com/hc/en-us/articles/204071764-Training-Stress-Balance, TSB = CTL - ATL.

However, here's what I see for some of my recent running workouts:

14 = 15 - 3
12 = 15 -4
11 = 14 - 5
9 = 14 - 6

There are more dramatic examples, e.g. 3 = 26 - 36, or 3 = 15 - 23

I understand that some rounding is invoilved, but surely -10 cannot be rounded to 3?

Alex

Are you grouping by week?

Its difficult to diagnose your misunderstanding without seeing data.

Mark 

[Ale Martinez]

El miércoles, 11 de noviembre de 2015, 5:08:54 (UTC-3), Alexandre Prokoudine escribió:

Speaking of which, I see something fishy in TSB calculations in GC.

According to http://help.trainingpeaks.com/hc/en-us/articles/204071764-Training-Stress-Balance, TSB = CTL - ATL.

However, here's what I see for some of my recent running workouts:

14 = 15 - 3
12 = 15 -4
11 = 14 - 5
9 = 14 - 6

There are more dramatic examples, e.g. 3 = 26 - 36, or 3 = 15 - 23

I understand that some rounding is invoilved, but surely -10 cannot be rounded to 3?

Alex

Are you aware TSBtoday = CTLyesterday - ATLyesterday unless you check the option "PMC Stress Balance Today" in preferences?

[RichA]

I have it working in sync with GC in a spreadsheet - I use a PMC in the spreadsheet so I can plan forward.

Row 4 = Start values

Row 5 = Formula

Col C = TSS

Col D = ATL =ROUND(D4+($C5-D4)*(1-EXP(-1/7)),3)

Col E = CTL =ROUND(E4+($C5-E4)*(1-EXP(-1/42)),3)

Col F = TSB =E4-D4

[Alexandre Prokoudine]

Nope, how would I know? :) It's not something I can see in a tooltip or even read in docs.

Mark, to answer your question, I guess you are asking if STS is set to 7 days in preferences? Yes, I kept the default 7 days for STS (and 42 for LTS).

Alex
 

[Karl Billeter]

On Wed, Nov 11, 2015 at 07:58:20PM -0800, Alexandre Prokoudine wrote:
>
>
> On Wednesday, November 11, 2015 at 5:10:23 PM UTC+3, Ale Martinez wrote:
> >
> > El miércoles, 11 de noviembre de 2015, 5:08:54 (UTC-3), Alexandre
> > Prokoudine escribió:
> >>
> >> Speaking of which, I see something fishy in TSB calculations in GC.
> >>
> >> According to
> >> http://help.trainingpeaks.com/hc/en-us/articles/204071764-Training-Stress-Balance,
> >> TSB = CTL - ATL.

Although from the same site

http://home.trainingpeaks.com/blog/article/the-performance-management-chart-past-present-and

Please note that we have not changed the behavior of TSB for Today on the
PMC. TSB for Today continues to be calculated, just as it always has, by
taking the difference between CTL and ATL from the previous day. As
always, we welcome and thank our users for your feedback and the
opportunity to adapt our product to even better serve your needs.

> >>
> >> However, here's what I see for some of my recent running workouts:
> >>
> >> 14 = 15 - 3
> >> 12 = 15 -4
> >> 11 = 14 - 5
> >> 9 = 14 - 6

Or
?
= 14, 15 - 3
= 12, 15 - 4
= 11, 14 - 5
= 9, 14 - 6

= 8 (presumably)

> > Are you aware TSBtoday = CTLyesterday - ATLyesterday unless you check the
> > option "PMC Stress Balance Today" in preferences?
> >
>
> Nope, how would I know? :) It's not something I can see in a tooltip or
> even read in docs.

Not sure about the docs :-), but yeah, I sort of put it together after seeing
the preference



Karl

[Mark Liversedge]

On Thursday, 12 November 2015 04:09:04 UTC, Karl Billeter wrote:

> Nope, how would I know? :) It's not something I can see in a tooltip or
> even read in docs.

Not sure about the docs :-), but yeah, I sort of put it together after seeing
the preference

I recommend getting up to speed on training with power, the key concepts and core science.

You can read about these things all over the net or ask questions here and in other forums dedicated to cycling and training methods; wattage, bikeradar, weight weenies etc etc etc.

The science section on the goldencheetah website is a good place to start it also has a link to Coggan's article on the PMC from October 2008.

There are many, many books on training with power that help those new to it, any of these would be a good book to get.

http://www.amazon.com/Training-Racing-Power-Meter-2nd/dp/1934030554

http://www.amazon.com/Training-Power-Meters-Louis-Passfield/dp/1847978975

http://www.amazon.com/Performance-Cycling-Science-James-Hopker/dp/1408146517/

Mark