FTP : Hills versus flats

stevepflanagan said:

...My question is would this be an accurate FTP for on the flats?...

Well there's always debate about whether the 95% of 20 minute approach yields an accurate FTP estimate whether on flats or hills, but I realize that's not really your question.

Many folks differ in terms of the power they can sustain in different terrain and it goes both ways, some do better on flats, many do better on moderately steep climbs. But most still only use one FTP estimate to evaluate their training regardless of terrain. But if your riding goals involve more flats than climbs (e.g. flat time trials) then you should test in that terrain. If your riding is varied then go with the current estimate but in the end your training levels will reflect what you can reliably do and they're pretty much self adjusting so a perfect FTP estimate isn't necessary and only really becomes important when you start tracking TSS and CTL.

There's some great information on FTP testing here:

Alex's Cycle Blog: The seven deadly sins
Alex's Cycle Blog: The Sins of Sins (Testing FTP #2)

-Dave

stevepflanagan said:

I have just recently started using a Polar CS600 Power Meter. I did a 20 minute climb (steady 8.6% gradient the climb) at an average of 370 watts, (I weigh 79 kg). As per the Hunter & Allen book, my FTP would be about 350w.

as Dave says, i'd tread lightly with the assumption of 20MMP * .95 = FTP. Most people seem to be in the .92-.97 range. doesnt sound like much but it can be especially with what looks like a super high FTP like yours.

stevepflanagan said:

My question is would this be an accurate FTP for on the flats? Thanks

most peeps i talk to consistently say their 60MMP is higher in climbs than flats. i'm that way too. other posters here seem to say that also. if you ride mostly flats, then i would do a dedicated test on the flats. otherwise if you ride in a variety of terrain, then i would use the best figure (for FTP) that i came up with.

stevepflanagan said:

I have just recently started using a Polar CS600 Power Meter. I did a 20 minute climb (steady 8.6% gradient the climb) at an average of 370 watts, (I weigh 79 kg). As per the Hunter & Allen book, my FTP would be about 350w. My question is would this be an accurate FTP for on the flats? Thanks

Maybe, maybe not.

Go out and do some testing and find out. Either do an hour flat out or two Monod tests - ~5 mins and ~20 minutes and plug the values into the spreadsheet.

I'm an oddball. I find that I can put out more power on the flats than I can the hills. Strange really, since I love climbing.

I thought that the 20" (x0.95) FTP estimation was only valid if you pre-stressed with three 1" near FTP muscle 'wake ups', then a 5" maximum effort. Then after a short recovery, ride the 20" TT which should be ~1.05% of your 1 hour FTP. 😕

bwbike said:

I thought that the 20" (x0.95) FTP estimation was only valid if you pre-stressed with three 1" near FTP muscle 'wake ups', then a 5" maximum effort. Then after a short recovery, ride the 20" TT which should be ~1.05% of your 1 hour FTP. 😕

That's the protocol most often recommended and helps to diminish the impact of anaerobic energy production but it's still a swag at the individual rider's AWC to CP relationship. It might be very close but it's just a broad guess that doesn't have a lot of published data to back it up. It's probably a good starting point but there are better methods for estimating FTP: Alex's Cycle Blog: The seven deadly sins

-Dave

bwbike said:

I thought that the 20" (x0.95) FTP estimation was only valid if you pre-stressed with three 1" near FTP muscle 'wake ups', then a 5" maximum effort. Then after a short recovery, ride the 20" TT which should be ~1.05% of your 1 hour FTP. 😕

Those "1" near FTP muscle wake-ups would be nothing more than part of your warmup (with other effort at FTP & above to hit the target energy systems). For a dedicated 20MMP test, & I did one recently, I go with a longer warmup - about 20 minutes - to help create enough of a buffer in my system. my perception is it makes the pain level increase more moderately than jumping into a full-blown 20MMP with a short 5 min, below FTP warmup.

20 * 0.95 is what dave says it is: a SWAG.

in actual fact FTP could be anywhere between 88-97% of 20MMP. it all depends on how well trained you are, what your AWC is, yadda yadda. IOW a whole lotta stuff. that's why it is a glorified 'eduguesstimate'.

so do your 20MMP alright, it is excellent training. add in a 5MMP test and now you have enough to plug into the monod spreadsheet. the CP value from monod plus recent power on long intervals will give you enough to get a hella good read on your FTP.

The protocol in TRWP is Hunter Allen's. My opinion is that it is reasonably accurate if you do the 5 minute maximal effort first, otherwise it tends to overestimate FTP. I do race a few 1 hour TTs during the year, but typically I will review ride files using #6 to tell me when it's time to retest using critical power testing, #5.

Alex's Cycle Blog: The seven deadly sins

Quoted post said:

"the seven deadly sins....

...er, ways of determining your functional threshold power (roughly in order of increasing certainty):

1) from inspection of a ride file.
2) from power distribution profile from multiple rides.
3) from blood lactate measurements (better or worse, depending on how it is done).
4) based on normalized power from a hard ~1 h race.
5) using critical power testing and analysis.
6) from the power that you can routinely generate during long intervals done in training.
7) from the average power during a ~1 h TT (the best predictor of performance is performance itself).

Note the key words "hard", "routinely", and "average" in methods 4, 6 and 7..."

stevepflanagan said:

I have just recently started using a Polar CS600 Power Meter. I did a 20 minute climb (steady 8.6% gradient the climb) at an average of 370 watts, (I weigh 79 kg). As per the Hunter & Allen book, my FTP would be about 350w. My question is would this be an accurate FTP for on the flats? Thanks

I have just migrated from Polar to PT and it's interesting to compare the two systems, though I didn't take the opportunity to run both together. In my experience the Polar is VERY accurate as a power measurement device over a longer period as confirmed by theoretical calculations based on my weight and an approximation of Cd. However as the time period decreased the Polar becomes increasingly inaccurate. In addition, the displayed power seems to vary depending on the gear selected. I confirmed this several times by placing the bike on a resistance trainer and cycling at a fixed speed and varying only the cadence and gear. There were quite distinct steps in displayed power at each gear change, though the actual power was clearly unchanged. I spoke with Polar about this but it was never resolved. Some other users put it down to a trainer specific aberration however I very much doubt that as it was also clearly evident in the real world. I used the earlier wired Polar power equipment and the problem MAY have been fixed on subsequent versions.

The bottom line of all that, is that if you're climbing a steep hill I'm guessing you were in pretty low gears and the power indicated may not have been accurate. So to answer your question, no it wouldn't be an accurate FTP on the flats. You can probably get a bit of an idea if they're the same by comparing your heart rate on the climb to that you see holding the same power for the same time on the flats.

While I didn't find the Polar especially accurate in these comparative situations, it was, however, consistent. Most of us train in the same area most of the time and you'll soon learn to interpret what the numbers say to what they really mean. I'd suggest the actual number is unimportant and the more significant is how your current performance relates to what you've determined from the past. ie if you determine your FTP at "350W" in that configuration and you're currently pushing "375W" you're over threshold. The fact those numbers aren't actually accurate is immaterial.

In the end it seems no power measurement system is perfect, and you may have just found one of the limitations with the Polar. As mentioned, the problem may have been fixed in later versions, or it may have even been specific to mine (though I've certainly heard similar stories from others). Sadly there is a lot of false information on the net about the Polar (difficult installations requiring extreme accuracy etc which is complete nonsense if you read the patent as to how it actually works) that makes these things difficult to share information amongst users as to what we can do to fix it. In the end I never bothered about it and just accepted that I was going up a hill pushing out 300 "apples" when I knew I was good for 330 "apples".

Pete

jollyrogers said:

The protocol in TRWP is Hunter Allen's. My opinion is that it is reasonably accurate if you do the 5 minute maximal effort first, otherwise it tends to overestimate FTP. I do race a few 1 hour TTs during the year, but typically I will review ride files using #6 to tell me when it's time to retest using critical power testing, #5.

Alex's Cycle Blog: The seven deadly sins

Quoted post said:

(from Dr Coggan)

"the seven deadly sins....

...er, ways of determining your functional threshold power (roughly in order of increasing certainty):

1) from inspection of a ride file.
2) from power distribution profile from multiple rides.
3) from blood lactate measurements (better or worse, depending on how it is done).
4) based on normalized power from a hard ~1 h race.
5) using critical power testing and analysis.
6) from the power that you can routinely generate during long intervals done in training.
7) from the average power during a ~1 h TT (the best predictor of performance is performance itself).

Note the key words "hard", "routinely", and "average" in methods 4, 6 and 7..."

I posted this elsewhere last week and it may be appropriate in the context of margin or error in FTP estimation among the SDS:
What I meant was that #7 has a small margin of error and #1 has a larger margin of error in estimating FTP. I would give a WAG that, in general, #7 is possibly +/-1-2% (depending on pacing, TSB, environmental factors), #6 is possibly +/-3%, #4 is +/-5% (although generally NP works very well for me - maybe better than +/-5%), and #1&2 are in the +/-10% range (depending on the type, effort level, and duration of the ride for #1 and depending on composition of relatively recent training and racing for #2)

I agree Jollyrogers in that #6 is a solid go-to (it is for me, at least). I also agree with Dave in that the validity of estimations of FTP are higher when the testing environment/terrain/bike position/etc are specific.

For the OP, a 20min test is not on the original list of SDS, but is #5b on Alex's list: Alex's Cycle Blog: The seven deadly sins
Let's say, for example only, that the margin of error in estimation of FTP with a 20minute test is +/- 4%. Then lets also say the your capacity to hold power on a 20min hill is maybe 3-4% different than on the flats. How valid is your estimation FTP on the flats, based on a 20minute test on a hillclimb?

Quoted message said:

stevepflanagan

It also depends if there was lots or not standing climbing. Try fully standing climbing sometimes and you'll find it's different :p

There was no standing at all on the climb. Since I did that test I have increased my average watts to 382 on that climb, but routinely produce an average of around 370, so at least I know the Polar CS600 is consistent.

I haven't done a flat test yet but I have done a 1 hour Tempo interval averaging 305w, with an average heart rate of 164bpm, (whereas the 20 min test on the climb was at an average heart rate of 172bpm). I therefore think that an FTP of 350, from the climb test, may be slightly overstated for the flat, maybe my FTP would be around 330w on the flat.