Not sure if I follow. The 20min FTP test isn’t just taking 95% of 20min power - there’s an anaerobic depletion component that happens before the 20min effort classically. And that 95% scalar isn’t set in stone or very precise. Many people would use a correction factor closer to 90%.
If someone does 20min@300w during a test, their “actual FTP” may be only like 270w. In which case 3h @ 0.88 would be 3h @ 237w, which is pretty reasonable.
Can you expand on this a bit, it is not my understanding of the latest science, or my personal experience.
Going on the assumption that RPE is the main determining factor in performance (I only quit when RPE gets too high) and carbs reduce RPE, wouldn’t that mean that carbs do improve performance at nearly any length of time? Even considering it may be placebo or some unexplained phenomenon (the famous swish and spit a carb drink), it still seems carbs improve performance for endurance activities? Happy to see where my thinking is flawed. Thanks!
I’ve been looking back at some old rides (back when I was 45) and yeah I did a few 4+hour .85 intensity rides back then. I was carrying some very high ctl at the time. 
My question is, at a ramp rate of 5-7tss/d/week for 26 weeks would I see my approximate peak potential (from ground zero) barring marginal gains from specificity et al?
The more I look back (and this is an insight I gained from xert a few months ago) basically the more I see my ‘ftp’ is basically a function of my ctl. I’m not at the pointy end of anything, mostly just went out and rode my bike a lot, then trained, and I’m not convinced I’ve seen anything different between the two. Arguably I rode a lot harder when I was just out flogging the roads.
You said 0.88 of 20 or 30 min power. No one else has said that, they’ve said 0.88IF which is against FTP (itself at most 0.95 of 20 or 30 min power. Maybe even 0.9 as you suggest). Either way no one else is talking 88% of 20 or 30 min power.
And this would be the problem with your argument…
Alan Couzens has been saying this for many years, and has data on it: Volume vs Intensity | Alan Couzens
I’m saying that it is difficult to demonstrate a marked improvement in performance due to carbohydrate supplementation unless the exercise bout is sufficiently long.
IOW, despite all the hype you see here about pre-exercise feedings, ingesting beaucoup amounts of carbohydrate during exercise, etc., you shouldn’t expect any miracles during workouts under 1.5-2 h.
Man, my genetics must suck. My Weight and CTL numbers in his calculator produce an FTP prediction about 100 watts higher than my FTP. 
My current CTL is 65, but I’m coming back from some extended time off, so I’ll take the result with a pinch of salt, but it’s giving me 269 (71kg), which is almost bang on TR’s AI FTP (267).
My best FTP ever was 292, at a CTL of ~95. The model predicts 282 for that CTL (so within 3-4%).
This puts me broadly under the fat part of the bell curve, which suggests average genetic ability, perhaps a bit better than average given my age. That kind of fits with my life’s sporting experience, tbh: noticeably better than most ‘average’ peers during cross-country at school, getting selected for teams, etc, but equally obviously not an unusually gifted athlete.
I always think you know very quickly when you meet someone genuinely, unusually gifted because they’re just totally different animals.
Oh I see. I skipped a step because I thought it was obvious.
If you are estimating FTP from a ramp test, or even a classic 20min test (which involves a 5-min all-out effort before the 20min interval), there’s a very high probability that the “FTP” result you get is something you’ll only be able to hold for 20-30min. Ie, your “FTP” is your 20-30min power.
Hence holding 88% of your 20-30min power for 3 hours is pretty f**king hard if not impossible for pretty much everyone.
If you’re calling your 1-hour power your “FTP,” then holding 88% of that for 3-hours is relatively straightforward.
Hence my point that much of the disagreement on this issue probably stems from people using different methods to measure FTP.
That’s fun data, thanks for the link.
Much of this makes complete sense. That said, I’d disagree that holding 0.88 for 3 hours is ‘relatively straightforward’. I’m not saying it’s impossible - the studies above seem to show that it’s achievable- but I still think most trained cyclists would struggle to do it. If you’re a long distance Tt specialist, or just naturally a ‘turbo diesel’ perhaps, but that would be a bloody hard workout for most.
Perhaps I’m wrong on this and live in a different world, who knows, but if you can ride at sweetspot for 3 solid hours then you’re a bit unusual, and I’ll shake your hand after and say that’s a damn good ride.
Weird. At 81kg and 65 CTL, it gives me 307 watts…so I exaggerated a bit with 100 watts, but still…
That’s awesome. It’s another interesting data point in my 2024 goal setting. very cool.
If I put in what I’d consider to be a good racing weight the ftp is pretty much spot on. At my weight it’s off like 5%. But still pretty damn good for a number that is based on almost nothing.
That’s fair.
The only thing I’d ask you in follow up is to look at your own power duration curve, see what your max 1-hour power had been, and then think if you could hold 88% of that for 3-hours.
Because threshold work is my weakness and I do no TT at all, but doing 88% of my one hour power for 3 hours would be hard but not terrible.
I’m sorry, but this makes no sense at all. By definition, your maximal 20 min power is something you can sustain for 20 min. Knock 5% off of that, and you will be able to go significantly longer.
IOW, the fall off in power with increasing duration isn’t nearly as steep as you imply.
That 20-30min number was just back of the napkin math and I was mostly thinking about ramp-test-derived FTP when it popped into my head.
I’m completely open to it being longer given that I pulled it out of my ass after thinking about the matter for maybe a second. This is kind of casual conversation so I wasn’t really looking to be precise.
What do you suppose the average duration that a recreational cyclist can hold their ramp-test “FTP” is? For me personally it’s usually around 20-25min. This is what I’ve seen among other people I ride with too.
And then what do you suppose the average duration that a recreational cyclist can hold their 20min derived “FTP” is? For me personally it’s around 30min. Can’t comment on others for that one though.
If you’re training regularly, and yet you can’t maintain your ‘FTP’ for at least 40 min…
…it ain’t your FTP.
The ramp test with power-only measurements is the worst way to estimate FTP. That’s the first problem. CP testing is far superior. And even then you still need to do a long test to confirm.
I’ll use me as an example. I don’t do much structured training lately and I rode for an hour at my threshold on Monday. Used CP to guide me and lactate measurements as a backup. I could do 70 minutes if I fueled and was more motivated.
Gotta find the real threshold.
Yes that’s my point.
Remember, this side conversation started with me stating that the reason why people thought that riding for 3h at 0.88 was impossible was probably because they were basing their “FTP” off ramp test or maybe 20min test results, whereas you were likely basing “FTP” off 1 hour power.
This forum would be much better with nested comments on long threads like this one 