Dan,
I was staying out of this to see how it shook out and how people viewed different things, but since we've moved into subjects that are in my direct area of expertise / research / publication...
from what i understand, and i'm somewhat behind the curve on this, i think you've created a TSS analog for running, no? this would be helpful for those who're using training peaks for triathletes, and this sort of sportswide extrapolation of what you've helped them develop for cycling is the direction i think they're going.
As Andy mentioned, I brought power and a TSS-like metric to running in 2005 when I released GOVSS. (The guys at Peaksware have been talking about their rTSS ever since, but have yet to release it...though they are again claiming they are close.) GOVSS is currently in use in a number of settings (you have seen thrown it around here enough, no doubt), including at least one national/olympic training center (to my knowledge). Ask MarkyV how he likes it :-)
It has already been done. I spent the last year finishing development on and validating SwimScore, which works by the athlete inputting distance, pace, rest and reps. The user must also input a threshold distance/time. I hope to publish on this in the next few months, and the white paper will be released shortly. If you are interested in playing with it, head over to http://www.physfarm.com/php-client. It is included in TriUtilities. The demo is free. You will be surprised at how well it works. (I was.)
Can't add them together...you need to treat each sport separately. You can prove this to yourself by using my performance modeling and prediction program, RaceDay. You can also get the free demo at the above address. You can very nicely (R^2 > .9 in some cases) model performance in individual sports, but you lose almost all predictive / modeling ability (in most people) once you start adding scores together. The principle of specificity reigns here. (Evidence based coaching at work :-) ).
Again, this is where modeling work can be very helpful, because you can begin to QUANTIFY the recovery costs of different workouts in different sports.
On the contrary, I think this is a situation where coaches can learn a lot from science...what many GOOD coaches find is that this sort of modeling work actually supports them. The benefit to using models is that the coach can learn how to wind up the athlete just right in a matter of months, rather than years. Using these types of models / metrics helped me push a veteran Ironman pro (at 39 years old) to a sub 9-hour performance after a lot of VERY heavy hitters brushed him off and said, "You want something your physiology cannot deliver." I'd like to say it was some coaching magic, but it wasn't. It was based entirely on scientific first principles.
Phil
Dr. Philip Skiba
Scientific Training for Endurance Athletes now available on Amazon!
I was staying out of this to see how it shook out and how people viewed different things, but since we've moved into subjects that are in my direct area of expertise / research / publication...
Quote:
from what i understand, and i'm somewhat behind the curve on this, i think you've created a TSS analog for running, no? this would be helpful for those who're using training peaks for triathletes, and this sort of sportswide extrapolation of what you've helped them develop for cycling is the direction i think they're going.
As Andy mentioned, I brought power and a TSS-like metric to running in 2005 when I released GOVSS. (The guys at Peaksware have been talking about their rTSS ever since, but have yet to release it...though they are again claiming they are close.) GOVSS is currently in use in a number of settings (you have seen thrown it around here enough, no doubt), including at least one national/olympic training center (to my knowledge). Ask MarkyV how he likes it :-)
Quote:
you've got to pretty much nail down a swim-TRIMPs or swim-TSS that's analogous to the bike and the run versions, in order for coaches and athletes to measure stress. It has already been done. I spent the last year finishing development on and validating SwimScore, which works by the athlete inputting distance, pace, rest and reps. The user must also input a threshold distance/time. I hope to publish on this in the next few months, and the white paper will be released shortly. If you are interested in playing with it, head over to http://www.physfarm.com/php-client. It is included in TriUtilities. The demo is free. You will be surprised at how well it works. (I was.)
Quote:
then you've got to decide whether you just add the two TRIMPs or TSS scores together for a triathlete's double workout in two sports on a given day. if not, you need to determine whether the stress score X 2 is then modified by the fact that it's a brick (back to back) or separated by several hours (and of course, how is it modified? is a run and a later swim worth more, or less, than simply the two scores added?) Can't add them together...you need to treat each sport separately. You can prove this to yourself by using my performance modeling and prediction program, RaceDay. You can also get the free demo at the above address. You can very nicely (R^2 > .9 in some cases) model performance in individual sports, but you lose almost all predictive / modeling ability (in most people) once you start adding scores together. The principle of specificity reigns here. (Evidence based coaching at work :-) ).
Quote:
then you need to ask whether you've got to adjust for distance. for example, let's stipulate for a moment that at, oh, Zone 3, a ride is worth about the same as a run a fourth of that distance (a 5mi run = a 20mi ride, a 10mi run = a 40mi ride). but i don't think we can extrapolate this forever. few coaches would agree that 5 days of riding 500 mi equals 5 days of running 125mi, at least not for a triathlete, because of the stress of a weight-bearing exerciseAgain, this is where modeling work can be very helpful, because you can begin to QUANTIFY the recovery costs of different workouts in different sports.
Quote:
accordingly, i think this is where science might have to have its come to jesus with tradition and myth and lore.On the contrary, I think this is a situation where coaches can learn a lot from science...what many GOOD coaches find is that this sort of modeling work actually supports them. The benefit to using models is that the coach can learn how to wind up the athlete just right in a matter of months, rather than years. Using these types of models / metrics helped me push a veteran Ironman pro (at 39 years old) to a sub 9-hour performance after a lot of VERY heavy hitters brushed him off and said, "You want something your physiology cannot deliver." I'd like to say it was some coaching magic, but it wasn't. It was based entirely on scientific first principles.
Phil
Dr. Philip Skiba
Scientific Training for Endurance Athletes now available on Amazon!