Using TrImp to calculate CTL, ATL and TSB
- Part 3
This is actually more of an update, as I finally figured out the formula and how to use it in Excel.
If you go back to Part 2, in the section where I explain what formula to plug into columns "H" and "I" for ATL and CTL, you need to change the formula. The correct formula for ATL, or column "H" (starting with the 7th row down, in this case it's "H9") is "=H8+((G9-H8)/7)". Then drag that down the rest of the column. You can also drag it up the column, but be sure to subtract 1 from the 7 for each row up you go till you reach the first row, which you can just reference the TrImp score for that day.
Now do the same thing for the CTL column ("I") using this formula: "=I44+((G45-I44)/42)" at the 42th row down (in my case it's I45). Drag up and down the same way (including subtracting 1 going up).
The basis for these excel formula changes is to use the real(?) formula for figuring out ATL and CTL scores. I found the formula on Middle Ring All the Way. Baisicly the formula for these are:
ATL = ATLy + ((TSS - ATLy)/TC)
Where ATLy is yesterdays ATL score, TSS is todays TSS (or in my case todays TrImp score), and TC is the time constant (in my case 7 days
CTL = CTLy + ((TSS - CTLy)/TC)
Where CLYy is yesterdays CTL score. Everything else is the same, but the time constant is for the 42 day length.
What this does is place more weight to the more recent workout sessions. Eventually the workout you did 8 weeks ago, will really only be a memory. And the workout you did this morning will have the most impact on your fitness. Now it's not going to make a huge difference, but there is a difference that can be seen.
Here's a chart using the old method (click to see it bigger):
And here it is using the new method (again, click to see it bigger):
The data is the same, but the formula is different. The difference is CTL and ATL isn't huge, but the difference in TSB (training stress balance) is bigger. There's less places where that number is in the positive.
Update: I downloaded some spreadsheets from the Google Wattage Group to check my formula against thiers. It seems to be pretty accurate. Remember that Using TrImp will give you different values. So you can't compare someone's power based CTL & TSB values to your TrImp scores. One of the next things I'm gonna do though is see if there is some relationship between them. At least with the power score values I get from SportTracks (not an accurate power measuring device, but at least an estimate).
Click here for Part 1.
Click here for Part 2.
ATLATLCTLCTLTSBTSBTSSTSSTrImpTrImpTraining+ImpulseTraining+ImpulsecyclingcyclingtrainingtrainingscorescorepowerpowerAcute+training+loadAcute+training+loadChronicChronicBalanceBalanceWeightedWeightedaverageaverageformulaformulaexcelexcelspreadsheetspreadsheetanalyseanalyseGunnarGunnarSoroosSoroosbikebikeracingracingpower+scorepower+scorebike+trainingbike+trainingSportTracksSportTracksGarminGarminEdgeEdge
Saturday 06/07/2008 | 10:17 am
Matt Rossing on Friday 06/13/2008 | 09:47 am
Nice work--similar to what i was toying with in my head. works well with my data--looks similar to what i had managed to get wko+ to do by manually entering my trimp data in place of the TSS score.
Write a comment