Homebrewed "TSTWKT"

Discussion in 'Power Training' started by kmavm, Aug 1, 2006.

  1. kmavm

    kmavm New Member

    Joined:
    May 16, 2005
    Messages:
    332
    Likes Received:
    0
    So, the Frank Overton article about TrainingManager, and the blog posts from folks who've been beta-testing it, has made me so hungry to use the thing that I've ended up trying to replicate it in a spreadsheet. I've played around with different metrics for "chronic training load" and "acute training load"; e.g., rolling averages vs. straight sums over a range of days, using 7 days for acute and 28 for chronic vs. 10 and 45, etc. None of the metrics seems to make too big a difference; the gross shape of the TSB curve seems similar regardless. I settled on rolling averages, since it seems to make more sense to me than having workouts count for 100%, and then suddenly fall off a cliff to 0% after 7 or 28 days.

    I'm considering "TSB" to simply by CTL - ATL. So, in my little model, positive TSB values indicate relative freshness, while negative values indicate relative staleness.

    A few observations:

    1. Grossly, the model "works." Looking back at mean maximal power data, strides in measured fitness tend to come after a month or two with high CTL followed by some sort of training interruption. For the most part, these mini-peaks have been unplanned on my part; they were more often the result of bad weather, not having time to train, etc., than a structured "recovery"

    2. Layoffs have unexpected consequences. The model predicts the huge bump in performance I saw coming back from my 10-day layoff in Colorado; my TSB at that time was the largest it had been thus far in '06. The ATL fell through the floor, while the CTL takes a good while longer to decay.

    A less obvious consequence of the layoff comes after resuming training. For a week or two, my CTL is still falling as it "digests" the long layoff. However, my ATL is back to my more typical 110-120 values, and this is causing a strongly negative TSB. However, I don't seem particularly stale; I'm putting up good, sometimes PB numbers for 3 minute power, 1 minute power, and 20 minute power, and I subjectively feel eager to train. Perhaps I'm not taking CTL over a long enough window?

    3. Most discussions of "TSTWKT" have envisioned an application to peaking for a target event. It seems likely that it would work for this purpose. But what does the model have to say about build periods? Getting a good "picture" of the gains you've made seems to require a brief drop in ATL; in the long term, this drop in ATL translates into a drop in CTL as well. So, during a build period, is it better to just keep plugging away, in an effort to achieve the maximal possible CTL before tapering? Or, is it better to be able to accurately assess the effectiveness of the training you've been doing so far, and throttle back a bit to see what you can do when you're fresh?
     
    Tags:


  2. RapDaddyo

    RapDaddyo Active Member

    Joined:
    May 17, 2005
    Messages:
    5,088
    Likes Received:
    41
    I'm curious as to whether TSTWKT or your model looks at volume by training level (L1-L7) or simply total ride statistics (duration, AP, NP, IF, TSS). Not having the benefit of TSTWKT yet, I have been working on a little different approach to training loads, based on durations by level. The first problem with that was developing the logic and algorithms necessary to parse ride files by training level, since I am not aware of any software that does that.
     
  3. kmavm

    kmavm New Member

    Joined:
    May 16, 2005
    Messages:
    332
    Likes Received:
    0
    TSTWKT as I understand it (and have been using it) deals entirely in TSS. It's intended as a unified, and so by definition coarse, metric for "form."
     
  4. RapDaddyo

    RapDaddyo Active Member

    Joined:
    May 17, 2005
    Messages:
    5,088
    Likes Received:
    41
    That was my impression. I guess where I'm headed is something else, what exactly I have no idea. It is based on weekly volume by training level. I guess I'm going down a solo path.:D
     
  5. frenchyge

    frenchyge New Member

    Joined:
    Apr 3, 2005
    Messages:
    4,687
    Likes Received:
    4
    Heh. I've thought about reverse engineering that thing myself, but unfortunately haven't had the time and inclination this time of year. Maybe in January when I'm starving for anything cycling related. Apparently, 1 person has successfully gotten close to Andy's model, but now that they're 'in the know' it's unlikely that they'll help the rest of us.

    The TSS rolling averages technique seems to be the typical "poor man's training manager." In that model, the ideal taper would seem to consist of 28 days of hard effort, followed by 7 days of complete rest, with maximal performance occuring on day 7. Unfortunately, that doesn't seem to be the case. The extra level of sophistication seems to be in the TL buildup/recovery area. Also, if you look at the screenshots carefully, you'll notice that TSB is typically in the +/- 0-20 range, as opposed to the +/- 150 range that one might expect from a simple CTL-ATL calculation.

    Anyway, I'm certainly willing to help crack that nut if my schedule frees up a bit. Count me in. :)
     
  6. acoggan

    acoggan Member

    Joined:
    Jul 4, 2003
    Messages:
    3,047
    Likes Received:
    9
    Dave Martin of/and the AIS already basically went down that path, but then couldn't figure out where it was leading them and sort of gave up...
     
  7. RapDaddyo

    RapDaddyo Active Member

    Joined:
    May 17, 2005
    Messages:
    5,088
    Likes Received:
    41
    Well, it's good to know that at least one other person had that idea. I think at the moment all I will do is track the data. Then, maybe one day I'll be staring at the charts and a light bulb will go on. Anyway, it's been fun figuring out how to parse a ride file.:cool:
     
  8. acoggan

    acoggan Member

    Joined:
    Jul 4, 2003
    Messages:
    3,047
    Likes Received:
    9
    If all goes according to plan, you won't have to wait nearly that long:

    http://www.jeff-mbh.blogspot.com/
     
  9. frenchyge

    frenchyge New Member

    Joined:
    Apr 3, 2005
    Messages:
    4,687
    Likes Received:
    4
    C'mon Andy. I've been around here long enough to know better than that... :p
     
  10. acoggan

    acoggan Member

    Joined:
    Jul 4, 2003
    Messages:
    3,047
    Likes Received:
    9
    Hey, you think that on June 27 I got up at 4 a.m., flew to Denver, drove (was driven) to Boulder, drove (was driven) back to Denver, and flew home to tumble into bed at midnight just for grins? :D
     
  11. Woofer

    Woofer New Member

    Joined:
    Dec 31, 2004
    Messages:
    599
    Likes Received:
    0
    Assuming the algorithm has not changed since that image has been posted - TSB has to be CTL - ATL. This matches the curves displayed exactly if the left side legend applies to both CTL and ATL and the right side legend only applies to TSB.

    Unless I am misinterpreting, I am not seeing the effect you are claiming.

    Here's another image that shows the same behavior.
    http://www.fascatcoaching.com/training_power/popup-6.htm
     
  12. cclarke

    cclarke New Member

    Joined:
    Jan 22, 2005
    Messages:
    35
    Likes Received:
    0
    Train high, sleep little? The new paradigm! I'll have to call my travel agent.
     
  13. Tom Anhalt

    Tom Anhalt New Member

    Joined:
    Dec 9, 2003
    Messages:
    239
    Likes Received:
    0
    Close...oh so close... :)

    I actually did "crack the code". But, now I'm sworn to secrecy so I can't help... :p

    I embarked on my path of a homebrew TSTWKT because I was sick of being teased by reports of it's use and having to wait...However, I have it on good authority that the wait will truly be a VERY short time from now.

    To quote Yoda; "Patience...you must have patience". :D
     
  14. frenchyge

    frenchyge New Member

    Joined:
    Apr 3, 2005
    Messages:
    4,687
    Likes Received:
    4
    You're right. It wasn't in Frank Overton's article, it was in a screenshot that I've seen since. If I get a chance I'll try to dig up where that was and post it.
     
  15. WarrenG

    WarrenG New Member

    Joined:
    Sep 8, 2003
    Messages:
    1,063
    Likes Received:
    0
    Your approach is better since not all levels of intensity need the same amount of work, nor recovery time. You'd be wise to allow the user to provide some input about their abilities in these various levels including nm so the software can predict their outcomes in a more personally useful way.

    Does 6 hours of sleep for a few nights in a row affect recovery? Extra work stress, hot temperatures during training, travel time to races, etc? All of these things matter for me and are given consideration in my training plans.
     
  16. Tom Anhalt

    Tom Anhalt New Member

    Joined:
    Dec 9, 2003
    Messages:
    239
    Likes Received:
    0
    Hence, the use of NP and TSS. That's the whole point.
     
  17. frenchyge

    frenchyge New Member

    Joined:
    Apr 3, 2005
    Messages:
    4,687
    Likes Received:
    4
    Here's the chart I was thinking of: http://www.twowheelblogs.com/2-old-2-go-slow/training-manager-its-real-and-it-rocks . I'm reading the right hand scale as the +/- TSB scale.

    Addtionally, if you look at the image on the link that Andy provided, it looks like the TSB scale is +/- 36.
     
  18. Squint

    Squint New Member

    Joined:
    Jul 27, 2003
    Messages:
    351
    Likes Received:
    0
    It doesn't use rolling TSS. The formula isn't something you can just enter into the formula bar.

    [​IMG]
     
  19. kmavm

    kmavm New Member

    Joined:
    May 16, 2005
    Messages:
    332
    Likes Received:
    0
    Well, since that looks an awful lot like OpenOffice output, it must be something that someone can enter into a formula bar.

    I must admit, all this "secret sauce" nonsense drives me a bit bonkers, and seems counter to the spirit I'd expect from this sort of enterprise. If TSS, IF, VI, etc., were all opaque and unexplained, I'd treat them with hugely more suspicion. If the formula is "black magic," how am I supposed to critically evaluate the model in question? How can I understand the assumptions that went into it, and its possible limitations?
     
  20. Squint

    Squint New Member

    Joined:
    Jul 27, 2003
    Messages:
    351
    Likes Received:
    0
    A friend of mine initially made it in OpenOffice with some of my input and then I translated it into Excel. The screenshot is from Excel. We are not part of the inner circle so we didn't have anything more to go on than any other member of the public.

    The formula could not be entered into the formula bar. We had to program it in VBA and the OpenOffice equivalent.


     
Loading...
Loading...