OK my final piece of FB research!
From Kevin Williams:
The general form is filter(yourdataseries, {1;30}, 2), where yourdataseries could be weight, 30 is the length (in days if using weight), and 2 means average 15 on the left and 15 on the right. If you want to average 30 to the left, use 1 instead
Thank you!!!
Thank you 
Thanks again, that was really helpful and I updated some charts.
@redlude97 Just over a week ago on another thread:
We were talking about FRC and HIE modeling differences, these were the values from 9 days ago:
On Saturday I sessioned a short ~30 second steep hill, did 4 hard efforts. Impact on 90-day PDC was pushing up MMP from 7 to 31 seconds:
and with respect to FRC/HIE, saw these changes:
Minor movement in some of the other values, but nothing of interest w.r.t. modeling and estimation.
As others have said, it would be amazing to have a forum off Facebook. I donāt have an account and wonāt create one, and I know there are a lot of other people in the same boat that are missing out because thereās only a Facebook option. Please. Facebook is such a terrible site.
Yes come on already.
Making progress on the WKO5 webinars and am planning on using the test protocol that Tim Cusick presents (āInitial Structured Testingā), which is basically warmup + max 5 min, warmup + max 20-30 min and warmup + max 1 min on consecutive days with rest as needed.
How would you pace these efforts?
Iām hitting 60-80mins TiZ per Threshold (FTP set to 317, so hovering between 307-317) workout currently, but donāt have any idea where my max is for the 1 and 5 min tests are.
WKO5 gives me these targets/durations:
Additional question: What time range are you looking at when looking at residuals/testing targets? Last 30 days? Last 90 days?
My A event is a 100mi MTB race, so 3s power is pretty much irrelevant to me - how should that influence how Iām utilizing the PDC testing protocol?
For the 1 minute, you donāt pace it. Just go all out and hang on for dear life. If should be a dying quail effort. That is: if your power is even for the whole 1 minute, you didnāt go out hard enough. At least that is the advice from my coach - we are also using WKO5 mFTP, so Iāve had practice feeding the model 
Haha, I like that - totally makes sense and good to know that decreasing power is somewhat expected. Thanks!
I think the targets/durations from WKO5 confused me, but maybe Iāll ignore them until Iāve completed this initial set of testing to feed the model. Havenāt done any all-out efforts over 5 min so far, so hopefully will develop a feel for how hard I can go for that time.
Anything longer than 3 minutes should be paced as more of an even effort. Itās only the short efforts that should be done with an all out start and then just hang on.
Once you get through these initial max efforts, about every 4 weeks look at the power duration curve vs WKO5ās model of your power duration curve, and you want to target where the modeled PDC is most divergent (higher) than your actual PDC.
Basically you are trying to make sure that WKO5ās modeled PDC best fits your actual PDC.
And just like this week, when filling in the PDC always go longest to shortest
Thanks for clarifying, thatās how I did understand it as well. For the longer ones Iām more comfortable having some idea for pacing.
From my understanding of Cusickās tutorial, thatās pretty much how the duration/watt targets are derived in WKO5. Makes sense.
The initial structured testing Cusick showed was 5 min, 20 min, 1 min - are you saying 20-5-1?
Looking at my first set of max efforts, they were done in this order:
But after the first week, it was longer to shorter
I usually look at WKO5 @ 90 days.
I keep my eye on the model and feed it now and then. I donāt do a set 2-3 days of feeding though that might be good the first time around. Itās pretty easy to throw a 5 second to 3 minute interval into any training ride when the model needs it.
I often ignore the modelās request for long duration efforts. For example, it currently wants 55 minutes at 90% of FTP. I just donāt want to do that.
Every 4-6 weeks Iāll do a Kolie Moore longer FTP test and the WKO5 model seems to like that.
For me, the model is always looking for shorter - less than 2 minute - intervals. And since Iām trying to only feed with inside workouts, really sucks. For whatever reason, I find it easier outside to do really hard short intervals. But since all of my structured work is inside on the trainer, Iām try to primarily build my PDC model inside.
There are a lot of ways to skin that cat. I will have my athletes do a test week to get the initial numbers, then I like to do short tests at the end of a recovery week and hold the longer test until theyāve got a workout in their legs so theyāre not stale. I had all kinds of issues trying to FTP test during recovery weeks, and Iāve found a lot of people experience the same thing.
But I can hammer out short and sprint tests after recovery and some opener efforts pretty well. So essentially you get all the curve feeding done once every four weeks, but donāt have to muddle up too many workouts to get there.
For the test week, I do 5-1-20-sprint. I think thatās what Tim recommends in his webinar. But reason I do that is the same as above - Iām stale for long efforts if I do those first.
JMO.
Go find a slightly longer than 1min KOM hill near you and just go test on that. Ones that get steeper as you go i find easiest because it forces you to keep the power on. We have a local uphill TT hill that I use that runs up to a public dog park.
Always fun to go do an effort, get to the top and start dry heaving and feel like im going to cough up blood while karen is out taking her yorkie for a walk
I just did a 2:30 test a few weeks back on a hill like this near my house. Took the KOM from a club mate who was literally on a plane flying to move back to Japan at the time I did it. 
But set a slew of all time power PRs doing it so it worked.
Nice!!
Iām excited for that testing week - have 2 weeks with lots of travel and little riding ahead, but then will get after it.
What time frame are yāall looking at for PDC / mFTP?
Noticed earlier that my last week mFTP is much higher (331 vs 316) than my current year mFTP.
From Cusickās videos, sounds like you should retest every 30-90 days, so itās that also the time frame to evaluate the PDC for?
Side Q: Can I make charts depend on each other? Iād love to use the gearing chart (bar chart of rear gear on the workout view) as selector for a power and cadence distribution chart of when I was in that gear. Iāve done this manually now, where I just hard code the gear I want to look at, but obviously fun to optimize 
thanks!
