TSS - can this be right?

Bronzie
Bronzie Posts: 4,927
I’ve had a power meter (Powertap Pro+ used with my own Garmin 705) on hire for just over a week and I’m just trying to get to grips with it.

I seem to be getting some very high numbers for Training Stress Scores from the week of riding I’ve completed with it so far.

Allen & Coggan’s book indicates the following typical TSSs:
1hr Level 1 recovery ride = 12
2h30 Level 2 endurance on rolling terrain = 60
2h50 Cat 4 road race, rolling terrain = 185

And also the following guide for recovery:
0-150 = low – recovery complete by next day
150-300 = medium – some residual fatigue next day but recovery complete by 2nd day
300-450 = high residual fatigue may be present even after 2 days
450+ = very high – residual fatigue likely for several days

I did a 25 mile flat-out ride on Wednesday using a pretty much dead-flat test route that I’ve used over the years. It took me 1h13m with an average power of 250W (from the Garmin) and normalised power of 253W (from WKO+ for the CP60min figure). This was some 4 mins slower than I did in March which I put down to probably being a bit fatigued still from a hard road race the previous Sunday, and it being pretty windy on the night. I’m guessing therefore that my true FTP is around 270W but I obviously need to redo the test when I’m fully rested and with stiller conditions.

The rides I have completed and analysed with WKO+ (using 250W as my FTP) so far are:
Mon – 1h54m endurance ride – 204W AP – IF 0.894 - TSS 153
Tue – 0h40m recovery ride on rollers – 191W AP – IF 0.773 - TSS 39
Wed – 1h13m threshold (excl warmup/down) – 250W AP – IF 1.022 – TSS 127
Thu – 1h18m recovery ride – 171W AP – IF 0.752 – TSS 75
Sat – 3h30m easy club run – 176W AP – IF 0.856 – TSS 321
Sun – 3h02m endurance/tempo ride – 210W AP – IF 0.946 – TSS 304

So even if I have my FTP under-estimated by 20W or so, these TSS figures seem very high to me. Could there be an underlying issue with the data recording with the Garmin or is there something else I need to consider when looking at these TSS values?
«1

Comments

  • Anonymous
    Anonymous Posts: 79,667
    The TSS values are taken from the power and IF readings. Looking at the wednesday ride, it looks like your FTP is set slightly too low. If as you say, your FTP is 270 then enter that into WKO in the top right and back date it to the start of the week and see what happens. Your sat and sun TSS are immense!
  • Bronzie
    Bronzie Posts: 4,927
    Ok thanks for the replies - looks like I need to re-test when I'm better rested.

    As I understand it from reading "The Seven Deadly Sins" post, there are many options for testing but the most preferable in decreasing order of preference are:
    - 1 hr well-paced* TT (the Gold Standard)
    - the 20 min TT test protocol in Allen & Coggan's book
    - Normalised Power from a 1 hr crit

    Is this correct?

    * - presumably this is the tricky bit :wink:
  • Anonymous
    Anonymous Posts: 79,667
    What was the peak 20 mins NP in your hour long ride on Weds?
  • guinea
    guinea Posts: 1,177
    To get my FTP I just went to Richmond Park and gave it everything for an hour. It came up below 300. The next week I did the same and it was at 310. A week later I moved it up to 333.

    I don't think I was 10% better in two weeks, I just think I wasn't trying as hard as I could previously.

    I understand FTP to be the most you can possibly do in an hour.

    If I want to get high TSS scores 20 miles of intervals does that better than anything else I've found...
  • Anonymous
    Anonymous Posts: 79,667
    I know the 'best measure' of FTP is an hour all out but I find it very difficult to get anywhere that I can actually do that without getting junctions etc. unless I do it on the trainer, which I have done but it is horrible!

    Consequently I do mine from a peak 20 mins of a long climb and subtract the 5%. I always check peak 20 mins of hard rides to see if it has gone up...
  • liversedge
    liversedge Posts: 1,003
    I find that anything less than an hour and I start to doubt the number I use. You have to do an hour test.

    I know that I'm doing my test properly if I stop looking at the power number on my garmin and keep glancing down at elapsed time. Usually, I'll start clock watching at about 10 minutes and then every couple of minutes thereafter until cool down time. I swear that time moves more slowly the harder you pedal. I think that is consistent with general relativity too :?


    The hour test is great training too, for obvious reasons!
    --
    Obsessed is just a word elephants use to describe the dedicated. http://markliversedge.blogspot.com
  • Bronzie
    Bronzie Posts: 4,927
    liversedge wrote:
    I know that I'm doing my test properly if I stop looking at the power number on my garmin and keep glancing down at elapsed time.
    Interesting point - I made sure I couldn't see any power figures when I did my test as I didn't want to pace it on power. I just used the virtual partner on the Garmin from my last test ride on the course as my "carrot".

    I knew I was on to a stinker early on though when my virtual me went past real me after about 7 miles even though I had a stronger tailwind until then compared to the last ride. He even had the cheek to bid me a virtual cheery "Good evening" as he went past. :x
  • Bronzie wrote:
    Ok thanks for the replies - looks like I need to re-test when I'm better rested.

    As I understand it from reading "The Seven Deadly Sins" post, there are many options for testing but the most preferable in decreasing order of preference are:
    - 1 hr well-paced* TT (the Gold Standard)
    - the 20 min TT test protocol in Allen & Coggan's book
    - Normalised Power from a 1 hr crit

    Is this correct?

    * - presumably this is the tricky bit :wink:
    See here:
    http://alex-cycle.blogspot.com/2008/05/ ... -sins.html
    #2 on your list isn't one of the sins.
  • DaveyL
    DaveyL Posts: 5,167
    Bronzie wrote:

    And also the following guide for recovery:
    0-150 = low – recovery complete by next day
    150-300 = medium – some residual fatigue next day but recovery complete by 2nd day
    300-450 = high residual fatigue may be present even after 2 days
    450+ = very high – residual fatigue likely for several days

    Does that mean that with an HOP workout, you should be fully recovered the next day? The one HOP I've done so far had a TSS of 109.
    Le Blaireau (1)
  • liversedge
    liversedge Posts: 1,003
    Depends on your CTL!

    I can happily do 100TSS workouts day after day, but my CTL hovering around 90-100. When I had a CTL of 50 I needed a lot more recovery days.
    --
    Obsessed is just a word elephants use to describe the dedicated. http://markliversedge.blogspot.com
  • DaveyL
    DaveyL Posts: 5,167
    Ah. Just fired up the performance manager. With a CTL of 40 it is quite obvious why my pan is knocked out for a couple of days following the HOP...

    Must try harder! (or at least, ride more...)

    :oops:
    Le Blaireau (1)
  • Bronzie
    Bronzie Posts: 4,927
    liversedge wrote:
    I can happily do 100TSS workouts day after day
    Presumably a ride of 100TSS equates to approx. 1 hr at threshold?

    So a CTL (42 day rolling average of TSS) of around 100TSS/day would mean you'd be putting in "work weeks" of c.900TSS with an easy week every 4th week around 200TSS?
  • liversedge
    liversedge Posts: 1,003
    My weekly TSS is anywhere from 700-900. I take rest / easy days when I feel like it or work / rest of life intrude upon my obsession ;-). The only time I've chosen to take an easy week is as part of a taper or cause I had a crash :-(

    I will add however, that improvements in my FTP have plataeud almost exactly at the same time that my CTL plateaud. I struggle to get my CTL any higher now as I've run out of time I can train and my average IF is already .87 so can't push either intensity, frequency or volume. Then again, maybe I should do less. Christ, I don't know.

    Edit - yes 100 TSS is an hour at threshold.
    --
    Obsessed is just a word elephants use to describe the dedicated. http://markliversedge.blogspot.com
  • a_n_t
    a_n_t Posts: 2,011
    Will babelfish translate this thread? :lol:
    Manchester wheelers

    PB's
    10m 20:21 2014
    25m 53:18 20:13
    50m 1:57:12 2013
    100m Yeah right.
  • DaveyL
    DaveyL Posts: 5,167
    So is there a minimum CTL below which there is little point in doing workouts above a certain TSS, or for example the HOP workout?
    Le Blaireau (1)
  • a_n_t
    a_n_t Posts: 2,011
    Manchester wheelers

    PB's
    10m 20:21 2014
    25m 53:18 20:13
    50m 1:57:12 2013
    100m Yeah right.
  • doyler78
    doyler78 Posts: 1,951
    How long do you have to wait before an unseeded PMC loses the effect of the unseeded values (training due to poor bike fit too erratic prior to this to seed properly)? I've now got nearly 12 weeks of data.
  • Bronzie
    Bronzie Posts: 4,927
    a_n_t wrote:
    Will babelfish translate this thread?
    And I swore I'd never become a "power nerd" - hey ho! :oops:
    NapoleonD wrote:
    What was the peak 20 mins NP in your hour long ride on Weds?
    261W but it was in the last 20 mins of a 73min ride (pushing into a strong headwind and getting angry that my time was so pants) so by definition I shouldn't have been able to achieve this if my FTP really was 250W (I still think it's more likely around 270W).


    Having had chance to play with the WKO software a bit more, I've found part of the reason as to why the Sat / Sun rides scored so highly on TSS is that when I stopped at the cafe (both rides), although the Garmin auto-paused, when it's analysed in WKO it uses the time of day from the beginning of the ride file being started as the baseline. I therefore have a period of around 1 hour on the Saturday ride where I was stationary, but the data has done some sort of weird interpolation so that I'm actually scoring TSS while I'm drinking coffee 8) .

    So presumably to stop this happening, anytime I'm stationary for a significant period, I need to stop the Garmin and reset it (holding the Lap button for 3 secs) to start a new file and then upload both to WKO and add the TSS's together?
    Is their a setting in WKO that I need to alter to prevent this error?
    Can I retrospectively alter the data files to correct this error by removing the dead time?

    Secondly, I had the Garmin on the factory setting of:
    Settings > Data Recording > Record Data Points = Smart Recording (records data each time it alters)
    when it would presumably be better to set this to record "Every Second" (although obviously the storage capacity of the unit will be reduced this way). [As advised in "Training and Racing with a Powermeter].

    The corrected TSS's are now:
    Sat (FTP 250W) = 250
    Sun (FTP 250W) = 247

    And at my "true" FTP (subject to actually pinning it down a bit more):
    Sat (FTP 270W) = 214
    Sun (FTP 270W) = 224
  • APIII
    APIII Posts: 2,010
    I just switch the 705 off if I'm stopping for any length of time. When you switch it on again it will start from where you left off.
  • Alex_Simmons/RST
    Alex_Simmons/RST Posts: 4,161
    edited June 2009
    doyler78 wrote:
    How long do you have to wait before an unseeded PMC loses the effect of the unseeded values (training due to poor bike fit too erratic prior to this to seed properly)? I've now got nearly 12 weeks of data.
    About 4 months. But why not just clone the chart and seed the clone to see the impact?
  • Bronzie
    Bronzie Posts: 4,927
    Switched the unit off when I rode into work this Monday.

    The ride home is appended to the morning ride's file as "Lap 2", but when analysed in WKO I've got 8 hrs with the unit switched off that have accumulated TSS points as the software has linked the two "laps" together (even though there was an 8 hour period with zero readings). I've ended up with a 10 hr ride with c.1000 TSS when all I did was pootle into work and back!
  • Bronzie wrote:
    Switched the unit off when I rode into work this Monday.

    The ride home is appended to the morning ride's file as "Lap 2", but when analysed in WKO I've got 8 hrs with the unit switched off that have accumulated TSS points as the software has linked the two "laps" together (even though there was an 8 hour period with zero readings). I've ended up with a 10 hr ride with c.1000 TSS when all I did was pootle into work and back!
    Well that sure ain't right.

    There seems to be persistent problems with the Garmin management of power meter info.

    In any case, you can cut and paste any section of a ride file into a new file. and of course cut out any section of ride file you like. just select and cut.
  • DaveyL wrote:
    So is there a minimum CTL below which there is little point in doing workouts above a certain TSS, or for example the HOP workout?
    I don't think you can think of it in those terms.
  • DaveyL
    DaveyL Posts: 5,167
    I think the point I was trying to make was, if you have a low CTL, like in the 30s-40s, is it better to build that up by initially doing efforts at a level where you can come back and do them day after day, rather than do an HOP (or similar) which might give you a TSS of 100, but which you might need a day or two to recover from?

    I suppose this kind of relates to the "building a base" vs "doing high intensity" debate.
    Le Blaireau (1)
  • Different people can sustain a different rate of CTL growth over the medium term (number of weeks/months).

    But there is a practical limit and experience has shown that attempting to ramp CTL by more than 8 TSS/day per week is not sustainable in the medium term. What an individual can ramp at is very dependent on many factors. Many find much less than that to be a useful limit.

    How any individual workout fits into that is only part of the picture. CTL doesn't tell you composition. So if you have a CTL of 30, but do a HOP once a week, are an experienced trainer but have limited time then you'll probably be able to do it. If you're a newbie, then probably not a good idea and it would be worth working your way up there.

    Either way, I think doing a HOP is probably better left until you have worked your way up to efforts of that duration and intensity. When I did my first HOP this past season I was at a CTL of ~60 TSS/day.

    End of day, if you have a CTL of 30 and are hoping to be in good shape, well you'll need to train more.
  • doyler78
    doyler78 Posts: 1,951
    doyler78 wrote:
    How long do you have to wait before an unseeded PMC loses the effect of the unseeded values (training due to poor bike fit too erratic prior to this to seed properly)? I've now got nearly 12 weeks of data.
    About 4 months. But why not just clone the chart and seed the clone to see the impact?

    Missed that one - that's what I get for tacking on to someone else's thread :lol:

    That's a good idea - hadn't thought of that. I'll have a play around with as I got a bad cold a few weeks ago and it will be interesting to see if I can find clues in there for it (then again it could be sardine tin I share with 6 other people at work). My guess is too aggressive a ramp rate. Perhaps I have found my breaking point (fitness wise for once and not bad setup) which if I have should prove very valuable info.
  • For those that have a long history of power meter data, and then plug it into the Performance Manager for the first time (provided they have a reasonable history of FTP) it can be exceptionally enlightening.

    In fact it was scary the first time I looked at my data at how well it showed my previous season.

    I tagged it the "Retrospect-a-scope" when Dr Coggan was beta testing it all.

    I see even Joe Friel is impressed with how good it is based on his recent blog entry:
    http://www.trainingbible.com/joesblog/2 ... ining.html
  • James_London
    James_London Posts: 530
    Bronzie wrote:
    I've got 8 hrs with the unit switched off that have accumulated TSS points as the software has linked the two "laps" together (even though there was an 8 hour period with zero readings). I've ended up with a 10 hr ride with c.1000 TSS when all I did was pootle into work and back!

    WKO+ does some irritating interpolation because it doesn't appear to recognise the markers when the Garmin auto-stops (if you have it set to stop the timer below a certain speed). Just select and Ctrl-X (cut) the offending data in WKO+. It's a bit annoying in town though not a problem on a proper ride where stops are infrequent. It's easy to spot this in WKO+ as long as you are viewing with time as the X-axis and not distance.

    You definitely want 1-second recording and you can now (with the 2.70 firmware) choose to display Average Power instead of NZAP (non-zero average power) on the 705.

    Your FTP would be 95% of the 261W rather than 261W but as you say 20 minutes at the end of a ride is not a good way to test it. Follow Coggan's protocol or do an exceptionally hard race and check the 20 minute max average in WKO+. Given the numbers you've mentioned, 250W is definitely too low anyway...

    It took me a few months to get into the whole power thing but once you're happy start the workouts in the back of the Coggan book and you will really get the most out of it.