Question for the Ex.Phys Boffin(s) (AWC, Power-Duration)



Alex Simmons said:
OK - I know that's another question but I just wanted to be clear what you mean by long term CTL.

How long it's been up there in the clouds and not having just ascended to that level perhaps?

I suppose "long term CTL" is a bit redundant, eh? :)

What I meant was that in addition to moving the power-duration curve upward (well, the aerobic portion, anyway), endurance training also tends to flatten the curve (i.e., decrease the negative slope of the terminal component), and it is this latter effect that I would hypothesize is most impacted by how many miles/hours/years you have under your belt.

At the extreme, I think you can see this in the performance of young pursuiters (e.g., Tyler Phinney), who can often approach the times of the best in the world at 4 km, but lack the stamina to compete at the same level in 3 wk stage races. As such talents continue to train (and reach maturity...development undoubtly plays a role), they may (or may not) get a little faster at shorter durations, but they get a lot faster at longer durations.
 
amartinez said:
are there real studies about this ?

I know that there have been various attempts to develop equations describing the speed-duration (or power-duration) relationship over very long periods of time (e.g., in migratory birds). I'd have to dig around a bit, though, to see what has emerged from such efforts.
 
acoggan said:
At the extreme, I think you can see this in the performance of young pursuiters (e.g., Tyler Phinney), who can often approach the times of the best in the world at 4 km, but lack the stamina to compete at the same level in 3 wk stage races. As such talents continue to train (and reach maturity...development undoubtly plays a role), they may (or may not) get a little faster at shorter durations, but they get a lot faster at longer durations.
Ummm...dare I say, "raise the left and then fill the right"? :rolleyes: ;)
 
i'm not one of the "boffins" but i'll chime in on what i've observed anecdotally..

i'm not going to quantify it because i don't know how to separate which portion is coming from AWC and the degree of the effect is dependent one's AWC ability but... i know that as long as for the preceding period i keep intensity slightly below FTP... the ending period i can bang out a good ~10 min a fair bit above FTP... and the last 5 mins far in excess of FTP... the ave power for the total period far in excess of what i could do for an isopower interval (dependent on length of intervals)... in fact for me, if i had started the interval at the ave power for the total time period i probably wouldn't be able to complete the interval at that power...

but i if i'm interpreting you question properly (and not sure if i am)... AWC has less potential to effect ave power for the total duration the longer the duration mostly because that ~10 mins that you can actually use you AWC ability effectively represents a smaller and smaller percentage of the total time the longer and longer you go out because that number is fixed.. 10mins/20mins is 50%, 10mins/240mins is 4%... so AWC is of little effect for a 2hr interval..

i've found if you are riding an isopower interval at FTP or above your ability to even utilize AWC ability diminishes rapidly... if you ride all but the last ~10mins or so slightly under FTP you can more than make up for it on the back end as long as the 10min is significant portion/percentage of the entire interval and you AWC ability is such that, that 10mins contributes significantly to the ave power for the interval...
 
amartinez said:
Nice, lets call it ei=endurance index:), would be an insteresting value to track ?
I think that you have to be careful about what your intentions are and data you apply to the curve. The original idea (or at least what the guys who wrote about horse racing and human running tried to do) was to predict times at various distances. Alex applied his MMP data to make a curve for us here. As a prediction of performance, that might not work too well. If I started an effort at my 5-minute best power and tried to hold it for as long as possible, I doubt that I would be able to do my best power at succeeding durations. OTOH, if you are simply using the MMP data to derive "EI" (as defined above) as an index to track and not using it as a predictor of performance, that's something else.

Do horses understand proper pacing? Maybe we should ask one. :rolleyes:
 
doctorSpoc said:
in fact for me, if i had started the interval at the ave power for the total time period i probably wouldn't be able to complete the interval at that power...
That probably suggests a NP for any duration far in excess of the AP for that same duration, which is not unheard of. Is there a motivation factor at work here?
 
Steve_B said:
That probably suggests a NP for any duration far in excess of the AP for that same duration, which is not unheard of. Is there a motivation factor at work here?

i think some of it might be motivational... but i think it's physiological as well... i blow up doing isopower intervals at ave powers less than 'ramped' ave power for ramped intervals and the RPE is far greater for the isopower intervals than for the 'ramped'

e.g. my FTP is in the 250 range... just from memory, if i was going to do a 20min interval at say 260W i would do something like this.. 1st 5mins at 240W, next 5mins ramp to 260W... ramp to 270-280W... ramp to 300W+... if i started at 260 i'd blow up at about 10mins... wouldn't have a snowballs chance in hell of completing that interval..
 
doctorSpoc said:
i think some of it might be motivational... but i think it's physiological as well... i blow up doing isopower intervals at ave powers less than 'ramped' ave power for ramped intervals and the RPE is far greater for the isopower intervals than for the 'ramped'...
I've found the same thing, particularly in training, less so during long TTs. My best AP for long efforts this winter all happened during ramped workouts like you describe. More than once I'd try a subsequent workout (after some easier days) right at the AP for the ramped effort and usually blew early but then did very well in another ramped effort a few days later.

I definitely think some of it is mental for me and some is probably related to warmup. I try to get a really solid and fairly long warmup for time trials and that seems to help me get into target pace more quickly. But regardless of the reason ramped efforts do seem to work really well in training and I almost always finish the last 5 to 10 minutes well above my target AP which is a nice kick to a long steady interval.

-Dave
 
Steve_B said:
Ummm...dare I say, "raise the left and then fill the right"? :rolleyes: ;)

You can, but "lift and flatten" is much more appropriate (esp. since you can maximize glycogen stores - i.e., "fill" - by training in a manner that actually steepens the curve). :)
 
acoggan said:
since you can maximize glycogen stores by training in a manner that actually steepens the curve.
Which would mean boatloads of L1/L2, i.e., LSD rides for long durations?
 
Steve_B said:
I think that you have to be careful about what your intentions are and data you apply to the curve. The original idea (or at least what the guys who wrote about horse racing and human running tried to do) was to predict times at various distances. Alex applied his MMP data to make a curve for us here. As a prediction of performance, that might not work too well. If I started an effort at my 5-minute best power and tried to hold it for as long as possible, I doubt that I would be able to do my best power at succeeding durations. OTOH, if you are simply using the MMP data to derive "EI" (as defined above) as an index to track and not using it as a predictor of performance, that's something else.

Do horses understand proper pacing? Maybe we should ask one. :rolleyes:
Well, similar pacing consideration also aplies to any other TT events, if you run your 10k race at your best 1500m pace are in a similar problem...

I don't known much about horse racing but, in our country, they are run with a rider (Jockey) :)

Talking more seriously, for prediction purposes would be interesting to try other cases to see the goodness of fit and if the slope remains reasonably constant when we extend the duration.
 
OK, for some fun, here are two more charts. This time I have compared Normalised Power and Average Mean Maximal Powers for the durations 5 minutes to 3.75hrs.

One for each of the two years of data I posted earlier.

2006:
RiegelsEquationNPvAP2006.png


2007:
RiegelsEquationNPvAP2007.png
 
Steve_B said:
Which would mean boatloads of L1/L2, i.e., LSD rides for long durations?

I was actually thinking the opposite: heavy doses of level 5 and 6 training would strongly stimulate glycogen "supercompensation", but would likely result in a steepening, rather than a flattening, of the power-duration relationship. Ergo, it is a mistake to equate flattening of the power-duration curve with the "filling" of anything.
 
doctorSpoc said:
i blow up doing isopower intervals at ave powers less than 'ramped' ave power for ramped intervals and the RPE is far greater for the isopower intervals than for the 'ramped'
I can understand and I've done this myself riding the trainer, though not to the same degree of undershoot/overshoot. I find it makes it easier mentally sometimes. Outdoors, I find that I often "perk up" and ride harder on the last interval, basically because I know that I am almost done, as opposed to having to slog through 3 more or whatever. There might be a similar motivational factor at work.
 
acoggan said:
I was actually thinking the opposite: heavy doses of level 5 and 6 training would strongly stimulate glycogen "supercompensation", but would likely result in a steepening, rather than a flattening, of the power-duration relationship. Ergo, it is a mistake to equate flattening of the power-duration curve with the "filling" of anything.
well filling one's stomach after a nice long ride comes to mind :D.
 
acoggan said:
I was actually thinking the opposite: heavy doses of level 5 and 6 training would strongly stimulate glycogen "supercompensation", but would likely result in a steepening, rather than a flattening, of the power-duration relationship. Ergo, it is a mistake to equate flattening of the power-duration curve with the "filling" of anything.
OK on the filling. I was was simply mimicking Kirk Willett as an attempt at humor.

We're getting quite a bit off topic here (but since when has that ever stopped us?) but since I'm a curious type, I have a question.

This explains the definition of glycogen supercompensation and a "conventional" way of achieving it. So what you are discussing above doing the depletion phase of the glycogen supercompensation cycle with L5/L6 training, which would also have the effect of raising "the left", the part of the curve at the shorter durations. However are you saying that supercompensation by itself has an effect upon the slope of the line?

Thanks.
 
This graph is with my own data (last 9 months), I'm an AG triathlete and my FT was around 240watts in this period, instead of log-log I've used a more familiar linear graph with duration in Hours (from 5' to 6hr), the exponent of the curve is the "endurance index" (slope in log-log graph) and coefficient is (around) 1h (avg or norm) power:

20pbl3c.jpg


Just in case anyone want to play with his own data I've attached the excel file (zipped).