WKO5 is here and it looks... different

[kurt.braeckel]

What I do right now is look at EF for each interval, and I can do that straight off my reports page on Training Peaks. I can also look at things like EF for the last minute of the intervals and see when they really blow up. But those trends are good to watch for overall fatigue. In my experience with these blocks myself and coaching these blocks, the athletes subjectively know when they’re toast before they can’t physically squeeze out more quality intervals anyway, so it comes out in their training comments, or in strava comments.

One of my athletes commented before his last VO2max workout on Strava, saying, “When I started the block I was motivated and wanted to smash all of it. Now, I’m kinda over it.”

I replied and said, “Yep, and we’re done.”

[WindWarrior]

At the athlete level, this expression

workoutrange(meanmaxfrom(power,300),meanmaxfrom(power,300)+300,ef)

Will find the best 5-min average power for each workout, and calculate the EF.

It does not find “all” the 5-min intervals, and I don’t know how to do that. By “all” I mean there is nothing in the WKO expression reference that allows searching a workout’s interval list, or finding all non-overlapping 5-min MMP power bests.

[kurt.braeckel]

It’s probably not worth the time to do it considering how easy it is to do on TrainingPeaks or even with WKO itself using highlight functions anyway. It takes me about 30 seconds to review a set of VO2max intervals and see if the athlete is progressing or hitting the wall.

[ojtCycling]

Lots of WKO5 threads knocking around. Which one shall we use for chart building discussions…this as good as any?

Anyway, I’m trying to build an expression that will allow me to show the % difference in 2 numbers.

My use case is I am looking at my peaks across various PDC durations during my base phase and I want to see what % difference they are to my all time best. My thinking on this is to assess if I am doing enough “intensity management” in my base phase to make sure I don’t drop off too much in this period from last season and can build back up stronger this season.

I have managed to copy a chart and understand the logic (and get values) for the peak power and all time best power for a duration but would now like to express the % difference as a annotation plot on a graph. Thoughts please?

[WindWarrior]

Something like this?

30-day vs All-Time PR (for me it starts October 1, 2016):

and 90-day vs All-Time PR:

[WindWarrior]

and variation on a theme, best weekly 5-sec power vs all-time PR:

Last week (week 49 of the year) my max 5-sec power for the 4 workouts completed was 621W and that was 54% of my all-time PR of 1155W.

Last 90 days is the selected date range.

A few configure screens:

and somehow this ended up not showing a red line on the chart…

and

the X Axis Label is blank in all 3.

[ojtCycling]

Excellent! Yea thank you this is the type of thing I am after. Particularly the % against all time expression

[WindWarrior]

Create a new chart…

make a gauge with % RHE (right hand expression) date range vs all-time PR. I’ve hardcoded that to October 2016 when I received my first power meter.

add a blue bar for power-RHE:

add a line for all-time power PR:

Final chart with the configure menu still open:

you can copy the chart and change meanmax(power, 5) to meanmax(power, 60) for 1-min, and meanmax(power, 1200) for 20-min, etc. etc. You must make those changes to all uses of meanmax(power, DURATION) for the new chart.

[ojtCycling]

Fanatic thank you very much @WindWarrior for taking the time to set that all out. I will try and go through it tomorrow

[WindWarrior]

Combining the two ideas, now the grey bar is the 90-day (selectable, right-hand range) max, and the blue bars are weekly max 5-sec efforts:

Make sure all the “X Axis Label” on the 4 chart elements - All Time PR Gauge, 5-sec by week, 5-sec max, and All Time PR Line - are using week.

Now we can see the current max in that 90-day window was the week of 7 October 2022 (10/3/22).

And a family of charts for 5-sec, 1-min, 5-min, and 10-min:

Hope that helps. Ask if you run into issues.

[WindWarrior]

Same concept on a PD Curve:

90-day MMP vs all-time MMP

• dashed line is all-time
• dotted line is 90-day % of all-time
• yellow highlights where 90-day is within 99.9% of all-time

Of course instead of 90-day, you can select 30-day or 180-day or whatever in right-hand range.

[kurt.braeckel]

Another option that could be easily modified with different time ranges, but one I like. Kinda similar to TR’s season charts, and you can look at this year vs. Last year vs. last 400 days. If you wanted to make it “all time” you could just enter that argument for the 400day plot, etc… and view current vs. All time PDC. This is data from someone who hasn’t done much testing recently, hence the 28 day plot is low across the board.

[RONDAL]

This is great as I have been wanting to build a version of this from TP for a while but have been too lazy to really get into it.

[juanperro]

I would agree with STP and i recently said that WKO5 is the best value for money. You buy a program once and there isn’t any subscription.
The iLevel power zones are very helpful and you do have to keep “feeding the model” (WKO5 have youtube videos) and this only helps your training.

[RONDAL]

The one thing I wish you could do with WKO is have the data in the cloud so that I can utilize on laptop or PC depending on location. Its the one thing that has me keeping my TP subscription, is the fact that remotely I cannot do workout/race analysis.

[WindWarrior]

Have you tried using remote desktop? In the past I used it for other reasons, worked well enough 15 years ago.

[smlring]

Keeping the pc up just for wko can be too much waste for some.

[WindWarrior]

hmm, even 15 years ago the computers in the lab were sleeping until I launched a remote screen sharing session and started working. Has technology regressed?

[WindWarrior]

I’ve been running my own IPSec VPN tunnel for over 20 years, and Mac based. It all just works but I’m not a typical user. I’ve also remotely managed a Mac lab and it all just worked - technology stack supported by Apple included wake-on-LAN packets and use of remoteComputerName.local which resolved via Bonjour/Rendesvouz with all computers using DHCP. No static IPs, no special magic, just connect via remoteComputerName.local and sleeping computer wakes up, and I remote desktop into it over a VPN tunnel.

My dad used to use a commercial app to remotely access his Mac server.

I thought you either pay a little money and make the problems go away with a commercial solution, or you roll your own for free.

[WindWarrior]

yes, I speak several different versions of Klingon For instance, WKO5 expression language.