TrainerRoad + WKO5

Hi all,

Long-time user of TR, I like to think that I’ve helped in this forum as much people as others have helped me. I’d like to present to you my plan to have some feedback.

My plan is to start the low volume plan changing each workout with the alternates to increase it to 1h30min (if necessary) making sure to match the level of “productive/stretch/breakthrough” of the 1h workout. In addition to that, I will do a 2h endurance ride and the long ride on weekends.

I will use the mFTP rather the result from the ramp test and also doing a 20min test rather the ramp test. My first question: does this makes sense? or would I receive bad results on the TR model for not doing the ramp test?

Moreover, if the workout is above threshold, I was thinking of adapting the workout and use iLevels from WKO5.

What do you guys think? does it makes sense? or the adaptive training model would not me well feed?


To mFTP work properly you have to maintain your PdC pretty regularly- this means maximal efforts around sprint, 1 min, 5min and longer efforts around 20-40 min.

WKO is amazing software but rather for long term analysis rather than acute changes (my personal opinion). You could try wko for 30 days to check if this suits you but personally I never use it for workouts prescription, only for analysis and trends. But maybe it will work for you.

In theory that would work great and is basically a form of adaptive training. But it reality it takes some effort, literally.

I’ve had WKO for several years and find mFTP to be a pretty decent match for what my FTP is from tests but ONLY IF I am doing hard all out efforts regularly over varying time frames either from group rides or races or buckling down and doing periodic WKO focused test efforts. If I just feed WKO structured workouts, and particularly base phase sweet spot workouts with little or no above threshold work, I find that my mFTP is significantly below my “real” test based FTP.

That is not a flaw in WKO, its just the nature of models, they need good data to work.


@bbarrera - ping, as you are the only TR user I know who also uses WKO5 :grinning:

As said above, mFTP decays unless maintained. You don’t want to use it to set training zones and I don’t think you need or want to change your FTP every day.

Just do your 20 minute test every 4-6 weeks and use that number to train by. You can look at your 20 minute aerobic/anaerobic contribution percentages and use their actual percentage for your FTP. For example, you may be 93% aerobic or 96% as opposed to the average 95%.

in particular if all done indoors on a trainer, in Erg mode. Same is true for other models (e.g. Xert).

@Zirbio you can use FTP estimate from 20-minute test with TrainerRoad.

For work above threshold, I’ve used power duration curve directly to help set power targets. Its discussed in many WKO webinars, for example near the end of “Individualize Your Training with WKO5” or “Training and Coaching with WKO, Lesson 3: Establish Training Targets.” Keep in mind Coach Tim Cusick is responsible for majority of the content, and likes to start many webinars by saying it is his opinion.

The Coaching and Training with WKO playlist is a good place to start learning about using WKO to self-coach.


Very true. Xert does have a setting to remove the decay, and go with training load, when on a plan, but it’s not something that you can toggle on the fly and have it recalculate from the start.

Thank you all for your replies, much appreciated

1 Like

OK, I picked up WKO5 last year for this purpose (as well as some general use). WKO5 has a huge learning curve, and I have not really put in the time yet. That said, I’ve mainly been looking at the PDC and mFTP, and most of the time it has been in line with tests, estimates, as well as where I would estimate my FTP. (ie when over/unders are clearly over and under, I know my FTP is in that range, likely nearish the middle.)
Recently it gave me a number that was 20ish watts higher than it was trending, and I am quite sure my FTP is not that high. I don’t know enough to diagnose why that happened, but it is likely that the PDC is messed up from not having maximal efforts. I do the vast majority of my training riding indoors currently, so my max efforts at different time periods are likely inadequate. I suspect that some essential (for good estimates) aged out of the 90 day window, and that is what gave me the weird estimate.
I think that WKO5 can be useful for this, but you will need to be disciplined to feed it regular max efforts of different lengths to give it the data it needs for a good mFTP estimate. You will not get these efforts as part of a TR plan - you’ll get one with the RAMP test, and maybe some short sprints that are in some workouts, but that is about it.
I suspect that this lack of max efforts is likely an impediment to TR doing estimated FTPs from following their plans. Maybe some magic ‘ML dust’ will help with this :slight_smile: (In all seriousness, if they can individualize these estimates based on individual HR response they might be able to do a decent job. I expect this to be at least years away.) have a rolling block of 42 days that they estimate FTP from. You just have to do one max effort within that time for the algorithm. Can you use WKO like that? This is what I’m doing now with TR.

1 Like

WKO uses your PDC from last 90 days to estimate one’s FTP. It’s a black box. With Golden Cheetah, you can use any time duration and select from one of four models. Three of the model are standard in the business based on CP. You can set your own search intervals (curve fit parameters) on all four of the models. is also a black box and by the look of it, basically plagiarizing Golden Cheetah’s Extended CP Model.

It uses mFTP from whatever range you want. There is no problem to take last 365 days or any other time range.

1 Like

Or more precisely, it automatically uses 90-days for mFTP however you can use any date range. I’ve modified charts to use a not-the-last-90-days range.


I much prefer Golden Cheetah to WKO5, but both can be used to accomplish basically the same data analysis. However, I dislike WKO5 based on the fact that the devs think they know how to look at my data better than I do. No real comparison window, forced smart segments, data filtering, etc.

But for most of TR users, understanding the CP concept and adding 1-2 max efforts at 1-3min (for MAOD) and 12+ min (in addition to a ramp test for MAP) would alleviate lot of the confusion about FTP. Dr. PF Skiba even released a spreadsheet to do simple CP analysis and performance modeling with Banister’s IR model.

mFTP might change around in WKO5 but you are not obliged to use it or pay attention to it.

The FTP you enter in the rider profile is the FTP that the software uses for calculations. So, you are free to update your FTP based on the mFTP or not update it if you haven’t done an effort in a while.

1 Like

… some calculations like CTL, TSS, IF, and classic zones. Let’s call that set-FTP or sFTP.

But you can get those metrics anywhere.

The interesting physiology modeling stuff like iLevels, anaerobic work (dFRC), relative vo2max within a workout, absolute vo2max, energy contributions, etc, etc, all use PDC and ignore sFTP.

1 Like

And the hero bar and all the modeled physiological parameters is updated to reflected the self selected end date/range? That’s new or are you saying you can create a customize report (which is not).

Some standard and some custom reports. The custom reports are tweaks to the formula in standard reports. That and a bunch of date ranges.