"The 33"-Road Bike Racing - TSB-numbers

I just began using the PMC in WKO and IM trying to get an idea of normal build and race TSB scores. I wasnt using my power meter for all rides so mine is kinda off a bit. After a hard training week and 2 crits I was looking at a -32. After a rest day it came up to -26. Is this normal?

WKO+ needs at least 28 days of data before TSB becomes even SLIGHTLY meaningful.

it took me three months before it started to understand what "normal load" is for me.

WKO+ needs at least 28 days of data before TSB becomes even SLIGHTLY meaningful.

it took me three months before it started to understand what "normal load" is for me.

I have a years worth of data, just missing a ride or 2 from each week.

You can "seed" it by estimating TSS scores for prior rides based on other data like time and avg hr. I just did an ad-hoc one in a spreadsheet but I think there is a formula around. If someone knows where it is I'd appreciate it.

I use that to fill in the TSS for times when I don't use my PT, like when the wheel is broken or I use race wheels.

It took me a couple years to figure out what a too high CTL is for me. It's not the same for everyone.

You can "seed" it by estimating TSS scores for prior rides based on other data like time and avg hr. I just did an ad-hoc one in a spreadsheet but I think there is a formula around. If someone knows where it is I'd appreciate it.

I use that to fill in the TSS for times when I don't use my PT, like when the wheel is broken or I use race wheels.

It took me a couple years to figure out what a too high CTL is for me. It's not the same for everyone.
http://home.trainingpeaks.com/articles/cycling/the-science-of-the-performance-manager.aspx

Did you seed your ATL and CTL start values based on the directions? As skipmcne said, it takes a while to get enough data for it to be meaningful. I tried to backfill mine using estimated TSS values from HR data and it gave me a good idea of the shape although the magnitude was a good bit off (which I got a good amount of grief on here for :p)

I have a years worth of data, just missing a ride or 2 from each week.

In that case, just estimate it. I've since my initial attempy further calibrated my HR-based TSS estimates and periodically check rides with known TSS against it and it comes up pretty close. But even without HR, I've gotten pretty good at estimating what the TSS should be based on perceived IF.

1. Your training stress balance is slightly positive (+5 or so) on race day: most competitors perform best with a slightly positive TSB.

2. Your chronic training load reaches its highest level two to three weeks before race day: your CTL is a measure of your fitness level, and you want your fitness level to peak close to your peak racebut not so close that you dont have time to reduce your fatigue level with a proper taper.
http://blog.trainingpeaks.com/2008/09/managing-your-training-stress-balance.html

http://blog.trainingpeaks.com/2008/09/managing-your-training-stress-balance.html

Thanks, thats good info.

so my TSB of -80 and -66 is bad.

If you have a year of power data (even missing a ride or two a week) your seed values will be useless as they only effect the values for the days in your selected ATL and CTL constant (10 and 42 days usually). So just expand your PMC to something like twice that and you should have accurate values for the last 42 days. Filling in the missing workouts with an estimate based on IF is a good idea for completeness though. I usually just find a ride that "felt" the same and copy it to the missing workout.

About the original question, TSB is relative to the user. Some people (myself included) can't really stand a TSB less than -40 for more than a couple days without coming up to a more manageable value near -20. On top of that everyone needs to come back "positive" at some point. But that point is different for everyone. Friel posted about an athlete who stayed negative TSB for an entire season.

Lastly a TSB that stays that negative for long without symptoms of overreaching might be a sign of an improperly set FTP. Not saying it is, just saying its possible. TSB/ATL/CTL are based on NP vs. FTP. If FTP is inaccurate, everything else is.

As a test of my above comment I set my CTL and ATL seed values to 500 on 10/15/08. It shows me at 87 CTL today. This is within 10% of my actual value. Setting it to 1000 is within 15%.

I go 4 weeks on, 1 week off for the most part and you can see my TSB go from -40 to +10 during that time...... Fun stuff when you are deep in the - *like I am right now*. I have figured out that my race performace starts to take a hit around -20 TSB and I might as well stay home if its below -30.. Was boarderline last week with a TSB of -28 and my results showed.

Starting in May I will ride my base and will be more stable.