Hello,
May be someone can help me to understand.
My eFTP on Rouvy is 372
My eFTP on intervals.icu is 328
There is a big gap, which one should I trust more ?
Both have same activities.
It depends on how the algorithm in each platform is assessing estimated FTP…I wouldn’t compare them or trust either of them.
Do a proper FTP test if you want to know your ‘real’ FTP.
I suggest a KM Progression 1 Test: The FTP Test: Physiology and New Protocols
4 Likes
higher. the right one, is always the higher one! 
17 Likes
You could model both power curves and choose which one you like. Go to (left menu) power → (top menu) power curves → Define Curve. Add 2-3 data points to add a modelled power curve to the diagram.
My instinct is the Intervals eFTP is more correct, partly because, maybe, Rouvy have more incentive to calculate higher (happier customer metric), and maybe because your Intervals account is pulling in data from other rides in addition to your Rouvy trainer rides? For example, Intervals has a nice feature to filter between indoor and outdoor rides—if all your indoor rides are on Rouvy I’d be curious if applying the filter in Intervals gives you a closer number.
What @Olly_Thomas said - the most reliable way to find out is to try to ride at your supposed threshold for 30mins+. Providing you are not starting off way too high, Kolie Moore’s tests are way easier than a 20min test.
With the default settings, the intervals eFTP can be based on a 3 min max effort. In my case that will overestimate FTP by 20-30W. I have my eFTP min duration set to 30 mins (1800 secs) most of the time, which means that the eFTP is pretty close to what I can actually maintain for 30mins+. The main downside is that eFTP only updates when I do long FTP test or a race, so can be a bit low at times.
Interesting reading, but did I miss something or does it not actually explain how to then calculate the ftp number from these tests?
From the article…
Your FTP will be equal to the average power of the entire test…
1 Like
Duh. I guess I was expecting something more complex as it mentioned using ‘WKO’ software.
I might try this next time.
What’s interesting looking at my power curves in intervals is I’ve never actually hit my ftp power for 60 minutes (I don’t race or try that duration, just 8 or 20 ftp tests) - regardless of level of fitness. And if I’m honest I’d say I probably couldn’t hold my ftp power for 60 mins
It you’re being honest, then chances are your estimate from an 8-min or 20-min effort is too high. So your easy rides become harder than they should as a result of a higher level/zone 2. Perhaps look at your 20-min power at about 90-93% and see how that relates to what you could do for an hour.
1 Like
Then, if you’re honest, it’s not your FTP😉
Well, am curious - does everyone else also have a 1 hour ftp test to backup their shorter estimated ones to affirm they’re the same? Or do you just believe you could?
I guess to expand I should say I’m unsure if i could do it. Some of that is psychological because I just don’t really desire to kill myself to that degree. Also harder to find a route long and flat and traffic-free enough - I’m already a bit compromised in the 20, the 8 is easier obviously. But actually just talking about it has made me kind of curious to see what I’d get for a hard hour. Will put it on my list for this year and also the other advice given and the longer test.
I wouldn’t be in too much of a hurry to do that 1 hour test if I was you!!! FTP doesn’t actually really matter, in the sense that you can’t actually base your ‘zones’ off a single number, namely FTP. FTP as a number to base zones off will soon go the way of the Dodo.
The absolutely essential number you need to identity is your top of Z2 (in 5 zone model). That’s the number you need to base most of your training off IMO in order to get the most aerobic gains from your training. Keeping 75 or 80 percent of your time below that first rise in blood lactate is the key to sustained long term aerobic gains.
1 Like
I feel I’ve got a pretty good sense of that - both in rpe and hr and power. They all track reasonably well and I can pretty much feel when I’m going to high end zone 2 and into zone 3. Well I think I have anyway!
I guess the ftp thing was interesting just because i was looking through all the data available on intervals. For the years I’ve had a PM I was just curious to see the ftp from Garmin, and intervals, and looking at my 1hr power numbers too. I thought it odd that my ftp estimates was always higher than my 1 hour power but thinking more about it and these useful comments, I suppose it makes sense that my ftp is higher as I never ride flat out for an hour.
Equally I don’t know whether it would be the same if I could. I guess I was more curious whether people that are more competitive/serious cyclists - who race and train harder - would have longer intervals if an hour + at their ftp power. Presumably they would? But if not, I guess no one can say they could definitely hold that power for an hour unless they tested it?
This is more the (long unused) science background in me and I’m just naturally curious about these things.
I’m actually surprised this year with how much more riding time I’ve had, and armed with much more knowledge and at least a little bit of structure, I seem to have improved little - certainly less than I hoped!. And some weeks I end up in a spiral where my Garmin is telling me low/unbalanced hrv and strained, yet I’m way under my average weeks training. Specifically after long events or rides of 5+ hours where I push myself. Maybe this is just age or genetics or I need more expert input to hone my progress….
I appreciate ftp is just one of many numbers. At least ftp is universal and kind of easy to understand and does give you some indication of progress over the months/years.
Re the 1 hour test thing. Our league races are often typically around an hour or so. When the race goes from the gun and it’s flat out full for an hour, the NP for that hour is a great estimate of FTP. That’s how I used to confirm my actual tests. Previous life now though 
No, I don’t have a 1 hour test to back up the number I use as my ftp. I am best at holding a constant power on a steady 7-8% grade, and I don’t have one near me that’s long enough to last an hour. If I did, it would be too high above sea level by the time I got to the top and that would start to affect my power. And if I had a 7% treadmill to ride on I would not enjoy monthly one hour tests on it. For all these reasons I just use eftp. I also never tell anyone my ftp, so it’s not really relevant what others might think of my methods.
I really don’t get why people fetishize the 1 hour thing, which is totally arbitrary. What I care about mostly is having an accurate estimate of the two numbers I need to graph my power curve: steady aerobic power and AWC. If you want to call that first number something else besides FTP so be it, but those are the ones I care about as it tells me how I can pace efforts of various lengths.
Generally…take a KM Baseline Test…at the very least achieve the 30mins mark…take that as FTP and build our TTE from there to over 60mins time in zone with a view to being able to hold new FTP for 1hr continuous…then build FTP…rinse and repeat.
This is specific to me and my goals though…40m TT where I need to be able to hold it for around 55-60mins continuous to be competitive in road bike category in the UK.
1 Like
The follow up question would be:
What protocol do you follow when doing either the 8-min and/or the 20-min test?
I’ve lost count of the number of tests I’ve seen people do that aren’t done as intended.
This is similar to what I did. I have a good feel for my FTP based on threshold and sweetspot efforts. My anaerobic contribution inflates my eFTP by quite a bit so I tweaked the eFTP min duration until it matched more closely what I believed my FTP to be.
1 Like
Can’t remember the exact details - I usually grab them from the web.
20 min x 0.95
Average of the 2x8 x .9
Both with some sort of hard effort before.
But l haven’t done one of those for ages as I just look at the intervals eFTP of 1x8 min (as I noticed it takes the better of the two rather than the average)