Huge difference in ftp in trainer vs hills

[Potuz V]

Hello, I'm new to power training but not new to cycling. I just got a quarq Elsa RS and wanted to start by setting my power zones for training. Here's my problem: I live in a very hilly place and essentially do not have access to flat stretches long enough to do a reliable FTP test. I can either do this on a trainer or on my daily ride which is a Hill that at top speed makes it a 19 minutes ride, but its steep at 9.5% average.

Now my problem is that on the trainer an FTP test following Coggan's method gives me an average of 253W (or 3.78 W/Kg) for my FTP (268W for the 20' part). However, on the said hill I can maintain 328W (or 4.90 W/Kg) for the 19 minutes climb. I only have a couple of weeks data with power but GoldenCheetah's Extended model is leaning towards a 311W value for the FTP which is no surprise since my normal ride is on steep hills and the only data it has is from these tests.

That's a huge difference. I am not sure if I should use Coggan's test which is repeatable or the hills that are affected by numerous factors let alone the fact that my average cadence is 71rpm on that hill and 95rpm on the trainer. So my first question is: which one would you use to set your power zones?

That leaves me to the second question: Should I attribute this difference to heat on the trainer (it's 38 Celsius and very humid here) or my bad power output at higher cadences?

Finally my third question and perhaps the only one that is relevant: does it make sense to use GoldenCheetah's model for my hills training and the trainer Zones for my flat trainings? I'm starting to lean towards this option. The biggest problem is that it seems that setting my FTP at 311W would make for very hard interval sessions on the flats (at least for me) but setting it at 253 makes for very easy (or even impossible) sessions on the hills which is my daily accesible training.

 

[RapDaddyo]

It really depends on why you want to know your FTP. If you want to know in order to define targets for training by level (e.g., L4, L5, ...), you can approach that in a different way and perhaps better suited to your training courses. For example, I use 90% of my maximum constant power for a given duration as a training target. So, if 328W is your maximum average power for 19mins, you could use 90% x 328W as a target power for regular efforts in the 15-20min duration range. If you want to know your FTP in order to define a target power for a ~1hr TT, then you would want to do some testing to confirm your maximum sustainable average power for an hour.

BTW, when you make reference to Coggan's test, I assume you are talking about taking 95% of a 20min effort. That's just one way to estimate one's FTP, and its accuracy will depend on one's power-duration curve. For example, in my case it would overstate my FTP because I have a relative high anaerobic work capacity. My preference is to do rides to exhaustion at a constant power for several durations and then use the Monod Critical Power Model to estimate MP at various durations including 60min.