how to measure CTL

Since 2001, over 90,000 cyclist's have joined Cycling Forums to discuss topics from general cycling to equipment, training, racing and travel or vacation destinations (especially in europe during the tour de france). We also feature an great deals in our online store, 100's of articles, classifieds and product reviews.

Hi, i'm a pretty big newbie when it comes to training and terms. I know CTL has to do with your training load, right? Do you need one of the cycling programs to figure it out? Or is there some way i can do some math and figure it out? Also, what is the main benefit of knowing it, just so you can either stay consistent or increase training without just adding time/intensity but switching the workouts up?

... I know CTL has to do with your training load, right?...Yep
...Do you need one of the cycling programs to figure it out? Or is there some way i can do some math and figure it out?...CyclingPeaks WKO+ sure makes it a heck of a lot easier. But you can hand calculate everything or use a spreadsheet. Here's what you need to calculate:

- Normalized Power using 30 second smoothing and pwr^4 averaging
- Intensity Factor = NP/FTP
- TSS = IF^2*hours*100
- CTL = CTL(d-1) + [TSS-CTL(d-1)]*[1 - exp^(-1/42)]
- ATL = ATL(d-1) + [TSS-ATL(d-1)]*[1 - exp^(-1/7)]
- TSB = CTL-ATL

All of that is easy to drop into a spreadsheet but it's much easier to just use WKO+. A spreadsheet version can still be useful for forward looking CTL, ATL and TSB estimates for instance when planning a race taper.
... Also, what is the main benefit of knowing it, just so you can either stay consistent or increase training without just adding time/intensity but switching the workouts up?The main benefit is to monitor your training load and have a measure of how deep your training base is. Also the benefit of being able to predict when you'll peak for events as opposed to hitting those peaks randomly. A lot of folks have also found their weekly CTL ramp rate during builds helps them predict overtraining and illness. So it's a good way to make sure you don't overdo things and get yourself sick during heavy training. It helps you figure out when you need additional rest or when you can afford to press a bit harder and lets you quantify the overall stress created by any individual workout and lets you compare that to the short and long term stresses your body has already adapted to.

In short, it's another training metric. Just as valuable or usesless as any others depending on how much time you take to understand it and how you apply the results. Check out this piece on the PMC: http://www.cyclingpeakssoftware.com/power411/performancemanager.asp

Good luck,
Dave

thanks as always to you dave for the super informative reply!

Hi

A long time ago Dave Ryan said:

Yep
CyclingPeaks WKO+ sure makes it a heck of a lot easier. But you can hand calculate everything or use a spreadsheet. Here's what you need to calculate:

- Normalized Power using 30 second smoothing and pwr^4 averaging
- Intensity Factor = NP/FTP
- TSS = IF^2*hours*100
- CTL = CTL(d-1) + [TSS-CTL(d-1)]*[1 - exp^(-1/42)]
- ATL = ATL(d-1) + [TSS-ATL(d-1)]*[1 - exp^(-1/7)]
- TSB = CTL-ATL

I get all my data off a CompuTrainer or PowerTap and I'm finding that my TSB on the PMC does not always conform to the TSB=CTL-ATL formula, any ideas why that might be? Sure beats me! It will be spot on for a couple of days and then be a bit different, for example my TSB on 01.06.2006 was -21.9 with a CTL of 65.1 and an ATL of 91.4 so in theory my TSB should have been -26.3. Just trying to understand. :confused:

Hmm.. Think I've worked it out, the CTL is the figure for the NEXT day? What an idiot :D

Thanks.

PBUK

Hmm.. Think I've worked it out, the CTL is the figure for the NEXT day?Sounds like you figured it out, but to be more clear: TSB is the predictor of how you will feel on today's ride based your training load through yesterday.

Today's TSB equals yesterday's CTL minus yesterday's ATL.

Sounds like you figured it out, but to be more clear: TSB is the predictor of how you will feel on today's ride based your training load through yesterday.

Today's TSB equals yesterday's CTL minus yesterday's ATL.yep, like frenchyge said, but my apologies I should have written that as:

TSB(d) = CTL(d-1) - ATL(d-1)

-Dave

Yep
But you can hand calculate everything or use a spreadsheet. Here's what you need to calculate:

- Normalized Power using 30 second smoothing and pwr^4 averaging
- Intensity Factor = NP/FTP
- TSS = IF^2*hours*100
- CTL = CTL(d-1) + [TSS-CTL(d-1)]*[1 - exp^(-1/42)]
- ATL = ATL(d-1) + [TSS-ATL(d-1)]*[1 - exp^(-1/7)]
- TSB = CTL-ATL

All of that is easy to drop into a spreadsheet but it's much easier to just use WKO+. A spreadsheet version can still be useful for forward looking CTL, ATL and TSB estimates for instance when planning a race taper.

Good luck,
Dave

Dave, what value/cell reference/operator should I be using when you indicate "exp" in this formula? (I'm trying to set this up in Excel)

Dave, what value/cell reference/operator should I be using when you indicate "exp" in this formula? (I'm trying to set this up in Excel)Actually, that is Excel notation for e^x (sorry can't write decent formulas in straight text) or "e" taken to the "x" power.

So for instance CTL based on CTL from the previous day, TSS, and using a 42 day time constant can be entered into Excel as:

"=ROUND(G2+(D3-G2)*(1-EXP(-1/42)),1)"

That adds a 1 decimal place rounding and in the case of the spreadsheet I pulled it from G2 represents previous day CTL, D3 represents current day TSS and this formula was pulled from cell G3 so copying the formula down(ctrl-d) the G column calculates running CTL over many days or weeks. You seed the sequence by putting your starting CTL into cell G2 in this example.

-Dave

Dave, what value/cell reference/operator should I be using when you indicate "exp" in this formula? (I'm trying to set this up in Excel)

If you get there, would you mind sharing it? I'm an Excel klutz, and the Training Peaks software is only PC. I've tried to find an Excel spreasheet, but no luck so far.

Not to mention, I only have hand-copied Power data from my Tacx trainer, not anything I can upload from a Powertap or SRM.

B

If you get there, would you mind sharing it? I'm an Excel klutz, and the Training Peaks software is only PC. I've tried to find an Excel spreasheet, but no luck so far.

Not to mention, I only have hand-copied Power data from my Tacx trainer, not anything I can upload from a Powertap or SRM.

B

Dave's 11:10 post solved it perfectly. Right now I'm just experimenting with how much info I want to include, but I could send you a working copy as I think I've got it working.

Dave's 11:10 post solved it perfectly. Right now I'm just experimenting with how much info I want to include, but I could send you a working copy as I think I've got it working.Sounds good. You can also download a very nice PMC spreadsheet from:
http://groups.google.com/group/wattage/topics
You've gotta join the wattage group which is free but full of good information. Once you join, go to the files page and look for:
PMC Template_Modified.xls (http://wattage.googlegroups.com/web/PMC%20Template_Modified.xls?gda=UTxtXEoAAADIKNpTkP4lQt0bsxNG6vGAyhoUB6KyVyzB4s_nO--2bGG1qiJ7UbTIup-M2XPURDTR3hR1V29Oykn-waV6DBw0zrYofg7E6Na7YPZK5qFX3g)

It's a very nice PMC tool that supports forward looking load planning.

-Dave

Actually, that is Excel notation for e^x (sorry can't write decent formulas in straight text) or "e" taken to the "x" power.

So for instance CTL based on CTL from the previous day, TSS, and using a 42 day time constant can be entered into Excel as:

"=ROUND(G2+(D3-G2)*(1-EXP(-1/42)),1)"

That adds a 1 decimal place rounding and in the case of the spreadsheet I pulled it from G2 represents previous day CTL, D3 represents current day TSS and this formula was pulled from cell G3 so copying the formula down(ctrl-d) the G column calculates running CTL over many days or weeks. You seed the sequence by putting your starting CTL into cell G2 in this example.

-Daveone shorthand way of calculating this (and avoiding the EXP bits), is to simply replace:

*(1-EXP(-1/42)) with /(tc+0.5) where tc=42 for CTL, tc=7 for ATL

try it :).

Should be quite close unless you drop tc very low and that's not pertinent for pseudo-PMC purposes.

IMHO I don't believe it's appropriate to carry a decimal point for any TSS, CTL, ATL or TSB factor. With the inherent degree of inaccuracy present firstly in the PM readings and secondly in how we all establish FTP, I think it's unjustified.

...IMHO I don't believe it's appropriate to carry a decimal point for any TSS, CTL, ATL or TSB factor. With the inherent degree of inaccuracy present firstly in the PM readings and secondly in how we all establish FTP, I think it's unjustified.Yeah, can't argue with that. I did it to check my predictions against WKO+ that carries a single decimal point. But you're right, you'd be hard pressed to notice the difference between a CTL of 90 and 90.1 :)

-Dave

one shorthand way of calculating this (and avoiding the EXP bits), is to simply replace:

*(1-EXP(-1/42)) with /(tc+0.5) where tc=42 for CTL, tc=7 for ATL

try it :).

Should be quite close unless you drop tc very low and that's not pertinent for pseudo-PMC purposes.

IMHO I don't believe it's appropriate to carry a decimal point for any TSS, CTL, ATL or TSB factor. With the inherent degree of inaccuracy present firstly in the PM readings and secondly in how we all establish FTP, I think it's unjustified.

Wow. That's cool. Now I want to figure it out.

- Normalized Power using 30 second smoothing and pwr^4 averaging


Any way of actually calculating this? Trying to get some TSS figures from my trainer sessions - my Tacx Flow gives me power and then average power per session.

My understanding is that on a trainer doing long intervals, Normalized Power and Average Power should be about the same?

I realise that this isn't going to be as accurate as going the whole Powertap/Cyclingpeaks route, but would at least give me some figures to monitor.

Any help/info appreciated.

B

Any way of actually calculating this? ...If you collect second by second power data(or data sampled much faster than 30 seconds) then you could calculate NP accurately. If you ride fairly steady efforts on your trainer you can get a very good estimate by skipping the 30 second averaging. It's what I do when I ride gym ergs. I block my workouts into 2.5 or 5 minute chunks and ride a steady power for at least that long before bumping it up or down. So a workout might look like:

min. pwr
2.5 150
2.5 190
30 245
2.5 190
2.5 150

That's typical for a short low SST day on the gym erg I use with the power steps it gives.

To calculate NP, just take the pwr column and raise it to the fourth power (pwr^4), multiply each raised power term by its time in minutes. Add them all up and divide the total by total time in minutes and take the result to the fourth root (average^0.25). That's a very good estimate of NP for relatively smooth data.

...My understanding is that on a trainer doing long intervals, Normalized Power and Average Power should be about the same?...If you ride fairly steady intervals even with steady ramps in power the AP and NP for the intervals are often the same or very close. But the AP and NP for the overall workout including warmup and cooldown and rests between intervals is usually quite a bit lower. If you get tricky with things like microintervals they'll be quite a bit different.

-Dave