Tuesday, January 15, 2008

FTP and Power

Training with a power meter is much like training with a heart rate monitor. You need a reference point and, based on that reference, personal zones are set up. Then workouts are prescribed using those zones. With a heart rate monitor the reference point is your lactate threshold heart rate. I describe how to find this heart rate reference point in my books, especially the latest--Total Heart Rate Training.

The reference point for power comes from the work of Andy Coggan, PhD. He calls it Functional Threshold Power (FTP). FTP is the power you would average with your highest possible effort during a 60-minute race. Once you know your personal FTP you simply use Andy's system to find your zones as described in Hunter Allen's and Coggan's book, Training and Racing With a Power Meter:

Zone 1 <55% (active recovery)
Zone 2 56-75% of FTP (endurance)
Zone 3 76-90% of FTP (tempo)
Zone 4 91-105% of FTP (lactate threshold)
Zone 5 106-120% of FTP (VO2max)
Zone 6 121-150% of FTP (anaerobic capacity)
Zone 7 >150% of FTP (neuromuscular power)

The key to all of this is discovering your FTP, and making sure it's current as your power changes relatively quickly, at least as compared with heart rate which hardly changes at all over the course of a season. Expect FTP to change about every four to eight weeks as your training load increases or decreases.

There are many ways to find your FTP. My favorite is simply to have the athlete complete an all-out, solo 30-minute time trial which is NOT part of a race. I've found the average power for this 30-minute, solo effort to be quite close to what the athlete does in a 60-minute race, such as a 40k time trial. The reason is that when riding alone you'll feel sorry for yourself and not go as hard as you would should it be a 30-minute race where you are held accountable to the world for your results and motivation is therefore considerably higher.

Allen's and Coggan's book describes several other ways of finding your FTP. And it's a great reference for understanding other aspects of power-based training.


At January 15, 2008 8:21 PM , Blogger Dinger said...

Joe, thanks for posting this. As luck would have it, I just got Allen and Cogan's book over the holidays, but just haven't got to it yet.

Again very helpful. Thanks for your thoughts.


At January 16, 2008 11:28 AM , Blogger d2g said...

Hi Joe, how would you compare a modified Conconi test result with Coggan's FTP protocols? Is there a correlation between them that you know of??

Thanks for your work and your willingness to pass along your knowledge!


Don Gabrielson

At January 16, 2008 8:40 PM , Blogger transplant said...

Mr Friel, why do you recommend the 30min TT over the (more common/popular) 20min minus 5% variation?


At January 16, 2008 9:22 PM , Blogger Joe Friel said...

Quite simple - it works.

At January 18, 2008 11:39 AM , Blogger rick said...


love this stuff

trying to correlate LTHR zones w/ power zones for myself.

i find if cadence is low (55) I can maintain "tempo HR" @ 153 for me and power averages of 210-215

as cadence is increased to say 90, i can only output power averages of 170-180 w/o hr jumping over 155 (above tempo zones)

does this show a deficiency of aerobic capacity vs power or is it purely a function of higher cadences using less afterload on the heart ?? or something else ??

ie:.....is this normal ??

or how can someone use this information to improve ?

At January 18, 2008 2:42 PM , Blogger Joe Friel said...

Rick--HR rising with higher cadences despite no change in power is common. Simply moving your legs faster--even if you removed the chain--would place a greater demand on the heart. Research shows that this is so and that low cadences (around 50rpm) produce the highest economy rating using the measures typically used in a lab (lactate, HR). But when asked to pedal at their most comfortable cadence at a give power *experienced* cyclists choose much higher cadences. Generally around 90 rpm. This tells me that there are economy determiners that we don't know how to measure in the lab yet.

At January 18, 2008 4:19 PM , Blogger rick said...

thank-you Joe

so i guess
? keep working till a higher cadence at a given power output is more comfortable

thanks again


At January 20, 2008 7:38 PM , Blogger Joe Friel said...

d2g--Sorry I missed your question above. I'd prefer to call it a graded exercise test (GXT) rather than a Conconi test. The latter implies that there is a deflection in the HR as power increases and that this deflection would be indicative of LTHR (or could be referred to as FTP). I have found that is rarely the case and the limited research I've read on this also does not support Dr C's test.

But to answer your question...you should roughly similar results for a GXT and most every predictor of FTP. It's doubtful they would be the same but would be ballpark close. Probably +/-10% which is quite a bit, I guess.

At January 26, 2008 9:12 AM , Blogger d2g said...

Joe, thanks again-- I'll modify my testing protocol to the 30-minute test from the GXT we've been using. I'm sure my athletes will love going from a 20- to 30-minute test... ;-)

I agree that 10 percent is more variation than we'd like to see in a result.

Best, d2g

At February 3, 2008 11:46 AM , Blogger Dessa said...

Do you know, what could cause lactate threshold (as well as FTP) to happen at 95% of VO2Max? Is it not enough strength or not enough aerobic training, or something else? I wonder if you have seen this to happen before.

Thank you in advance,

At February 4, 2008 8:49 AM , Blogger Joe Friel said...

dessa--There could be several reasons. My first reaction would be that either fatigue set in before completion due to poor aerobic endurance or lack of rest before the test or an unusual test protocol, or that the athlete was not motivated enough to continue the test. I'd compare the results of this test with others to see what the upper end numbers were. If lacking previous data you may consider a follow up test. Of course, knowing VO2max is of little importance for training.

At March 30, 2008 9:18 AM , Blogger rick said...

i like the 30 and 60 min for determining LT

however ,, using a power meter i have been doing speed intervals
15 sec all out, 30 sec recovery
today i did a set of 16...

this took @ 12 min

calculated my ave. power during sprints.. about 500
and ave power during recovery about 150
(HR varied b/t 168-175......w/ my nl LT around 168
the overall average was ...500+150+150=800/3 =266

interestingly.......almost identical to my CP 12.. ..continuously pacing myself steady at max for 12 min

is training one way of the other more beneficial ??
or does it train different systems ??
since in both ave HR's are about the same ??

i've read your training bible ,and Allen + Oggan's book on power and really can't get a grasp on this

i think it would make a great separate blog topic.



At March 30, 2008 2:56 PM , Blogger Joe Friel said...

Rick--The workouts you do are specific to the outcomes you want. If you want to sprint well you need to do sprints in workouts. If you want to climb well you need to climb in training.

At September 20, 2008 6:22 AM , Anonymous GiantDrag said...

Hi Joe, great article and to the point. I have a question. Just done a 20min FTP test (wish I had done 30min now) Anyway I did this on training bike on a flat road and averaged 315watts with HR pretty much at my normal race heart rate, so i'm confident its a fair result.

I think the 20min test states minus 5% of the wattage which puts me on 300watts. Thing is im pretty sure that had I continued for another ten minutes I could have held the wattage at 315 so it kind of makes the 20min value minus 5% a bit rubbish.

That isnt my point though really. Lets say I take the 315watts as my FTP (my estimated 60min race value)I agree with your theory about doing this in training and not in racing as well so the figure should be pretty accurate. I have also measured recently a fairly hilly 25mile TT, Aero, on TT bike in a race and it came out at 338 watts! Quiet a considerable difference - 23watts - Do you think this is infalted because I put big effort into the hills on this course (4 in total about a 45secs- 1 minute long each)and is it a safer bet to work with the FTP test result on the flat road I did today?

Thanks Joe. Drag.

At September 21, 2008 1:35 PM , Blogger Joe Friel said...

HI Drag--If you have a 60-minute mean max power from a race then by all means use that as your FTP regardless of the hills you mention.


Post a Comment

<< Home