TSS spreadsheets with ATL, CTL, Form


#1

Anyone have any tss excel documents with calculating atl, ctl, form pre-made? I’m trying to see if I can manage this on my own and update regularly via the cloud and predict performance. Anyone have anything they’d be willing to share? Triathlete friendly would be a bonus! Thanks.


Tracking form using TR?
#2

I have one that I can send you if you PM your email address.

Mike


#3

That’s an open offer to anyone who’d like a copy of my PMC spreadsheet.

Mike


#4

#5

This link should work:

https://1drv.ms/x/s!AloUeW1MOGCihP1DaEY6Wsaq4rG6ig

Mike


Other than FTP, What are some subtle ways to see if you're gaining fitness?
#6

this is great, thanks.


#7

The link doesn’t work anymore. Any chance you could update? Thanks.


#8

Updated with a new link.

Hopefully this one won’t expire.

Mike


#9

I’m interested to know where you got the CTL and ATL equations from. You are using the versions I first found when I was researching the spreadsheet:

  • CTL = TSS x (1-e^(-1/42)) + Yesterdays CTL x e^(-1/42)

  • ATL = TSS x (1-e^(-1/7)) + Yesterdays ATL x e^(-1/7)

I couldn’t get those equations to match what is in Trainingpeaks or WKO4 so I asked on the Facebook group.

WKO4 uses the following equations:

  • CTL = yesterday’s CTL + (today’s TSS - yesterday’s CTL)/42

  • ATL = yesterday’s ATL + (today’s TSS - yesterday’s ATL)/7

Mike


#10

…can’t tell you. I’m not the creator of the spreadsheet. There are many other/different spreadsheets out there.




[Disclaimer: For me the whole TSS/ATL/CTL/TSB/PMC thing doesn’t work at all – my body isn’t a machine; my ability to recover and/or handle training stress varies form day to day (and from week to week, month to month and and and)]


#11

Ah, cool.

The constants (42 and 7) have been determined from a population study and will vary from person to person. Might be worth you playing around with them to see if you can get something that matches your historical data.

Mike


#12

That’s the root problem: Me (body & soul) isn’t a constant… temperature levels, humidity level, sleep time and quality, sudden change in weather, seasons, daily stress, hydration levels, nutrition and and … all that has an impact on me, but changes all the time.


#13

I am no maths expert but there needs to be a multiplier in the WKO4 formulas.
Andrew Cogan posted these as the original formulas :

From the original spreadsheet (ca. 2003):

CTLtoday = CTLyesterday + (TSStoday - CTLyesterday)(1/CTL time constant)
ATLtoday = ATLyesterday + (TSStoday - ATLyesterday)
(1/ATL time constant)

TSBtoday = CTL yesterday - ATL yesterday (since you typically want to know how “fresh” you’ll be before today’s training ride or race, not after)


#14

Fair enough.

No point trying to fit to a model that you don’t think works for you.

Mike


#15

x 1/42 is the same as /42

Mike


#16

I did say that maths wasn’t my forte. :slight_smile:


#17

Thanks for this Mike!


#18

No worries. Any suggestions for improvement welcome.

Mike


#19

The WKO4 equations appear very different, with no exponential in them, but while they’re a lot simpler they actually give very similar results, generally within 0.1 for CTL and 1.0 for ATL of what the “proper” formulae give.


#20

The WK04 equations are the proper formulae, as written by Andy Coggan in the original PMC spreadsheet.

Not sure where the others originated. They do give very similar values and are very, very close if you change the exponents to 41.5 and 6.5 (or thereabouts, from memory). That’s the reason my spreadsheet has inputs to three decimal places! I just couldn’t get it to match WK04 until I found out about the simpler equations.

Mike