What data set is GC using to calculate critical power?

[sciguy]

We have just started using GC after a number of years with WKO. It's
incredible
what a talented group of enthusiasts can/will do for the pure joy of
creation.

The critical power page is especially nifty but I'm confused in
regards to where it's drawing data from.
Is it just one particular ride or a composite of all rides within the
season. The value it calculates
has been consistent from the very first ride entered despite a
significant positive trend
in maximum trend in best 30 minute as well as 60 minute power over the
past two months.

Thanks for all the wonderful creative work on this project!

Best Regards,

Hugh

[Jamie Kimberley]

There are three curves in the CP plot.

1. solid black line is the cp curve from the selected ride.

2. solid red curve is the aggregate of bests for all rides of the
current rider.

3. the dashed red curve is the calculated CP curve. This is based
on a critical power model. I don't remember the specifics but I'm
sure that you can find some info on the wattage google group.
essentially we have a functional form for a riders CP. we try to
fit that curve to a few points from the riders "bests" (solid red
line). I'm probably wrong on the numbers here but we pick a point
in the 2-5 min range and another point in the 10-30 min range and
use those points to determine the fit to the CP curve. This may
explain why your calculated CP may not have changed despite an
increase in 60 min power. Furthermore, riders rarely give an all out
effort for an hour so looking at data on the shorter intervals makes
some sense. That said the method used above seems to work best
when you have "filled" out the riders CP curve over the 2-30 min
region.

I'm sure that someone else can offer up more details. if possible
I'd like to add this to the wiki as it is good to describe the
methods used when they are not trivial.

Thanks for using GC.
Jamie

> --
> _______________________________________________
> Golden-Cheetah-Users mailing list
> golden-che...@googlegroups.com
> http://groups.google.com/group/golden-cheetah-users?hl=en
>

"You don't use science to show that you're right, you use science to
become right."--xkcd
__________________
Jamie Kimberley
Postdoctoral Fellow
Department of Mechanical Engineering
The Johns Hopkins University
Office: 410.516.5162
Mobile: 217.621.8272
Fax: 410.516.4316
E-Mail:jamie.k...@jhu.edu

[d78]

Why does the CP=X watts and AWC=Xkj mean. I would have thought CP=X
watts would have been power at 60 minutes however it's not so I'm
therefore confused as to what it is trying to tell me. Help?

> E-Mail:jamie.kimber...@jhu.edu

[Matthew Perry]

CP is critical power as calculated by the Monod model. This takes the
total work produced over a short duration effort (eg 3 min) and a
longer duration effort (eg 20 min) and draws a line between them.

The slope is your critical power; ie the power or rate of work you can
sustain (theoretically) for a long long time. This has been shown to
correlate with FTP pretty closely although I've personally never been
able to sustain 60 minutes at my CP (maybe about 5 or 10 watts shy of
it).

The AWC is the y-intercept of this line and represents the finite
amount of work you can produce anaerobically.

Neither CP nor AWC have anything to do with the actual power than you
HAVE produced for 60 minutes ... its only a model based on shorter
duration efforts to PREDICT your 60 minute power.

- matt

--
·´¯`·.¸. , . .·´¯`·.. ><((((º>`·.¸¸.·´¯`·.¸.·´¯`·...¸><((((º>
"The best way to predict the future is to invent it."  -- Alan Kay
Matthew T. Perry
http://www.perrygeo.net
http://viedevelo.wordpress.com

[djconnel]

I wrote that extraction code.

Basically the CP model is that there is a fixed supply of anaerobic
work, which can be fully spent over any duration, and a sustainable
aerobic power, which can be sustained indefinitely. Obviously very
simplistic....

Most CP calculators as for two fixed time durations: for example 5
minutes and 20 minutes. But the problem is you then need to have
totally nailed an effort at both of those exact times. More common
with me is you go hard up a hill of a given length, and the time it
takes is variable. Maybe you had a super power for 4:30, but your
5:00 power isn't so hot, or you did a super 30 minute effort but your
20 minute effort wasn't your best. So what the code does is
iteratively pick pairs of time points which generate the "best" CP
curve for your data.

The issue with it is it uses all of your history, while physiology is
constantly changing. We talked about putting in a time limit that it
would use data only from the previous N days for some N. But
unfortunately other tasks got in the way of me contributing further.

I like the CP curve because at least within the duration of around 2
minutes to around 30 minutes it gives me an idea if a "PR" over a
given time duration is actually a quality result or not. For
example, if all I do are 5 minute and 20 minute intervals, then I go
out and PR a 7 minute interval, is that really the result of superior
fitness, or just the result of doing an interval duration I'd
previously neglected? The guide for this is if the 7 minute effort
"pushes up" the CP curve. If it does, it really represents a new
level of performance. If not, it might just be a "gap fill" effort.

[d78]

Ok that sets the context however what I'm still no closer to
understanding is how the CP=X figure below the curve is different to
what the curve is telling me. How can it be right that my CP=X isn't
even on the curve. It is actually 1 watt below the furtherest CP point
(3 hours) on the curve. Surely it has to lie somewhere on the CP curve
and if it is to be used for zone setting surely it would logical to
set at ~60 min power as on the curve. What am I missing?

[SteveI]

On Aug 2, 9:00 pm, djconnel <djcon...@gmail.com> wrote:

> So what the code does is
> iteratively pick pairs of time points which generate the "best" CP
> curve for your data.

What are the constraints on where each time point has to come from?
The reason I ask is that my best 7 minute power is 20W above my CP
curve, so I'm thinking that the two ranges must be either side of that?

[djconnel]

Here's the relevant code:
CpintPlot.cpp :

// bounds on anaerobic interval in minutes
const double t1 = USE_T0_IN_CP_MODEL ? 0.25 : 1;
const double t2 = 6;

// bounds on aerobic interval in minutes
const double t3 = 10;
const double t4 = 60;

USE_T0_IN_CP_MODEL allows use of a 3-parameter model which I rejected
because it didn't do any better than the simpler, 2-parameter model.
The third parameter is a time offset added to the duration of all
efforts so power at zero time goes to a limiting value rather than
infinite.

So the answer to your question is it uses 1 to 6 minutes for the first
point, 10 to 60 for the second. Why the gap? The reason is I
wanted the first point to have a high fraction of anaerobic power but
a low fraction of neuromuscular power (which dominates in sprints).
The second point I wanted to be mostly aerobic. 7 minutes is sort of
in the intermediate region which doesn't give good resolution on
aerobic nor anaerobic power: if the two points used to derive the
slope (which is AWC) are two close, the slope becomes sensitive to
small variations in either point.

But you can always change the value and recompile....

Dan

[sciguy]

Dan,

Thanks so much for the very clear explanation of your method as well
was
work that went into creating it. So many of use appreciate the effort
more
than you would know.

Hugh

[Gene Raphaelian]

Dan -

I realize this thread has been dormant for a long time but I do have a question so thanks for any help you can provide. 

I'm wondering why my CP can be,,,

289 in September

285 for the Year

285 for All Dates

When I select Current Selection in chart settings. Shouldn't All Dates and Year also be 289 as that is my best for All Time and my best for the Year?

Last, what does the "Rank" column signify and why does it read "N/A"?

Thanks.

gene r 

[Karl Billeter]

On Mon, Sep 26, 2016 at 04:00:16PM -0700, Gene Raphaelian wrote:
> Dan -
>
> I realize this thread has been dormant for a long time but I do have a
> question so thanks for any help you can provide.
>
> I'm wondering why my CP can be,,,
>
> 289 in September
> 285 for the Year
> 285 for All Dates
>
> When I select Current Selection in chart settings. Shouldn't All Dates and
> Year also be 289 as that is my best for All Time and my best for the Year?

Your MMP should be be best for All Dates. Perhaps your W' is higher to an
extent that less of your MMP is coming from CP.

K

[Gene Raphaelian]

Thanks Karl. 

In the month of September, where GC has my CP @ 289, my W' is 10.6 kj. 

For This Year, where GC has my CP @ 285, my W' is 14.3 kj. 

Best,

gene r