So just starting back on the bike after a 2 yr hiatus. I have about 10wks of rides in system but the mFTP is ~60 watts above what I know it is on the bike. Because of this huge difference using the program to actually plan my workouts is impossible. Is there a way to temporarily adjust my mFTP to my bikeFTP or just get usable info from WK05??
Youâll need to be a bit more specific - which âsystemâ and what FTP figures are you referring to, i.e. where is your mFTP figure coming from? What is your bikeFTP and how are you measuring it?
Using WKO5, mFTP from hero bar. bikeFTP that I can manually adjust from FTP testing.
wko needs some max efforts over different durations in order to be accurate. Perhaps its worth performing some tests and then having another look at your mFTP.
I read a thread on the forums stating that 1-10min max efforts can effect the mFTP dramatically but the users had very little data in the WKO5 system. Me having 10wks of data I thought I would have overcome that statistical skew. I was off the bike for medical and have trouble still with the max efforts so will have to just use training peaks until I can get more reliable data from WKO5 I guess.
mFTP is your FTP based on WKOâs power profile model. bikeFTP is one you set manually. So I guess my question is, how do you âknowâ your FTP? What testing have you done?
You canât manually set mFTP⌠itâs based on the data you feed it. If it truly is that far off, maybe you have bad power files in there?
It isnât necessarily that you have bad values.
You need to set the model up and maintain it. You need to start with a week or ride of max efforts.
Get a 20 minute, 5 minute, 1 min and sprint max effort in. Then youâll probably also need to do a longer 1 hour type ride. Then check your residuals and see if you need to test those ranges.
I do understand from your post that you have an mFTP value from WKO5. But where are you getting your set FTP value? It is strange to see that you have such a large discrepancy. I have maintained my WKO5 model and my sFTP and mFTP are within 5 watts.
When I do a 35-45 minute ftp test, my mFTP usually jumps up and is spot on. WKO uses the FTP that you give it so if your mFTP isnât fully modeled, you donât really have to worry about it.
Iâve actually been on a peak lately turning in my best performances and having an all time high eFTP on Intervals.icu but my mFTP is 25-30 watts too low because I havenât done a test in a while.
I recently set new PRs for 1hr to 3.5hrs so I have a low mFTP with a long TTE.
The modeled FTP is based on your power duration curve over the last 90 days. Itâs garbage in garbage out, especially If any of the usual suspects falls out (roughly 5s, 1 min, 5 min, & 30-80 min @ MLSS but only Trainpeaks will know). If itâs been well feed and itâs still off, you are not alone but 60w is a lot. I use the PD curve to supplement and guide my testing and training, not to replace it.
Others have addressed the question, but how are you determining your bikeFTP? I think the ânot feeding the modelâ and âefforts falling outâ only makes sense if your mFTP is lower than what you think it should be. If it is that much higher, either your mFTP is pulled way up by skewed data (should be very apparent by looking at the curve) or your bikeFTP is set too low.
Can you post a screen shot of your PD curve with values for reference?
I am in the same position as the OP. I track my FTP well and have just manually adjusted it to 230 (after doing 2 months of base only with another month to go). My mftp is 268. There is no way that I can hold that for more than a few minutes currently. It was up to 278 at one stage which do match up with what I consider my real FTP pretty well.
I understand that the WKO model needs feeding with accurate and up to date data.
What I do not understand is how WKO aknows that me doing 1, 5,10,20 hard efforts over the course of a week are full on efforts. By doing them will my PD curve lower, which then impacts my mftp. Does WKO know that a specific combination of efforts is me trying to update my curve or will it just think that I am not doing full on efforts.
WKO doesnât know anything. Itâs trying to fit one of four models to your power data, and then it calculates mFTP from that. Sometimes you get unrealistically high mFTP values if you havenât done any longer efforts, or if you havenât done sprint efforts.
As you said, itâs about the data that goes in. You can post your curve here and we can look at it to help if you like. My guess is your curve fit probably isnât very good somewhere.
Please select Last 90 Days and repost. Leave the screenshot above for comparison.
Might also help to also post one for Last 30 Days. Thanks.
1 Like
Iâll give an example of how changes in max power at different durations will change how the curve fitting works.
5 days ago mFTP at 257W, and even this was not correct because I aborted my last 20-minute test about 11 minutes into it (had been holding 285+ watts):
and 4 days ago I had one hard effort around 20 seconds, and mFTP dropped to 240W
The only power durations that changed over the past 90 days - those from 9 to 37 seconds:
and that caused the NEW curve in RED to shift up for durations below ~4 minutes, and move down for durations greater than ~4 minutes:
3 Likes
First look I think you need to do some true sprint efforts unless your PMAX is really 650W. I think if you went out and did some short sprints, 5, 10, 15s or so on the road, youâd get a pretty substantial correction as @WindWarrior pointed out above.
When I have athletes do PD testing for their initial profile, I have them do 15s sprints (twice), 1m, 5m, and 20m, and then we fill it in as we go from there. I believe thatâs what Tim Kusick recommends in the webinars as well.
The application of the âTime Trialerâ profile curve will usually err on the side of a much higher mFTP than the other models, and youâll get that one if you either donât have any kind of sprint, or you havenât done any sprints in a while. Iâm an âAll Rounderâ most of the time, and when the model switches me to âTime Trialerâ, I usually know the mFTP number is too high and I need to populate some data for it to be accurate.
And also as @WindWarrior said, letâs see the last 90 days. (I actually use a time period of âLast 60 Next 7â quite a bit, too).
3 Likes
It would still help to see the 30-day and 90-day curves, instead of an 11 month curve (year to date). I also have a 42 day range and use that sometimes.
also @kurt.braeckel - John started post by mentioning recently completing 2 months of base so it possible there is still 1 long/max effort between 60-90 days ago is pushing mFTP up even though fitness has dropped.
Even easier than Kurts suggestion is to start with the 3 test targets shown in WKO, and spread them out over 1 week. Here are mine:
I have the opposite issue of @John_Hallas - mFTP is too low. I have been EASILY completing 30-min tempo efforts just under 250W, so a mFTP of 240W is too low but given my 90-day power curve you can see why WKO has mFTP so low. The only recent long effort ended early at 11 minutes.
2 Likes
Yeah, I added the part about the time period after I saw your post. Youâre right about that. The YTD data is kinda meaningless.
My process for finding FTP using WKO:
- Look at PDC and see the curve fit, determine if mFTP is good and compare to inflection points. This usually gets me 98% of the way there.
- Stole a chart from Trevor Connor where I can use 30 or 90day measures that kind of give me a range to work within for athletes who might have something else up. Most of the time, I can get numbers that agree within the ranges. (The shot below is my data for the last 90 days - Iâm lazy and donât change the title)
- Then a lot of it is a bit intuitive by subjective ride data and such.
Based on my data including my curves, I have been using 280W as my FTP for the last two months. It might be a little low now, but Iâm about to test now coming out of my VO2max block so weâll see.
Another helpful thing can be looking at a âzoomed inâ version of your PDC or multiple overlaid PDCs to see where inflection points fall at various points. If I look at my PDC with sprint data truncated off, my FTP inflection point is bang-on 280W.
With my most recently onboarded client, she came to me with a Zwift ramp test FTP of 205W and said she was getting burned out and really struggling to complete her training plan.
I looked at four weeks of rides for her and told her I thought she had her FTP set 20W too high. Before testing, I set her TP FTP at 185W and had her do some riding. She felt better. We did some testing, and her FTP inflection points and the charts above all had her between 175 and 185W. I put her at 180W, and her workouts look great.
More than 10% error in FTP from a ramp test⌠think thatâll change your training a bit? 
1 Like
Max power this year was 650 ish - not something I ever really test out. I do most of my riding indoors. I was a TT rider until I had an accident 18 months ago. Now I just really train to up/maintain my FTP. I could change it to another phenotype I suppose.
Last 90 days
Last 30 days
Last 60 next 7
Use the 30 day (or 42 day if it looks good). You need to wait until earlier efforts flush out of the 90-day windows. Yes, until then you have to ignore mFTP in the hero bar at the top, because mFTP is based on 90-day window. However for yourself just use 30- or 45- day to better match what you know.
Iâd recommend doing some short max efforts. The easiest way to update your 30-day PD curve is to do the short/medium/long test targets as in the screenshot I posted above. You might be able to use the 42-day now, that is the past 6 weeks (6 weeks x 7 days/week). Over time you can extend out from 42 to 60 to 90.
@kurt.braeckel my HR zones havenât changed since last tweaking them in early 2017. I can get within a couple percent of FTP by just going out and pacing an 8-min effort between 150bpm (about 90% ftp) and 160bpm (100% ftp). HR for last 2 minutes.
To use terms from the TR AI FTP FAQ, using HR-to-Power on 8+ minutes intervals since 2017 has been âMAGICâ for 5+ years 
and reliably gives me my ftp.
And FWIW my Garmin on endurance and tempo efforts is pretty close, here is the most recent one:
and Iâve got my ftp set to 272W, last updated 11 months ago.
3 Likes