« Build Period Overview | Main | Confidence »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Kelownagurl  (Barb)

Two questions:

1. If I don't have a power meter, is my avg speed on the trainer a reasonable enough number to use as data, since I can keep gearing and tension constant and don't have to worry about weather/terrain etc?

2. Following your formula, if on a LT bike test I rode an avg speed of 31 km/h with an avg HR of 174, then my output/input ratio would be 31/174 = .18?

And my goal would be to increase this ratio? (.19, .2 etc?)



Hi Joe,
I have followed my P:HR ratio over the last year not only because my heart rate floats one zone or so below my power zone (the usual tables matching HR and Power zones don't let me see if the call I am placing on the muscles is yielding the expected heart rate response) but also because I thought it might be a way of monitoring imrpvoements in pedaling efficiency - getting a few more watts out of the top and bottom stroke for the same HR.

But when I read chapter 2 of Allen & Coggan's new edition last summer, the error rates for the various meters out there caught my attention and I developed some uncertainty about whether I was overinterpreting data. The accuracy of the power meter and whether or not you remember to manually zero before every ride has a strong impact on the statistical significance of digits at and past the second decimal place.

(If I get the math right anyway - under ideal circumstances, if your power meter has a claimed accuracy of 1.5%, a ride with a reported 200W average power may actually be 204W or 196W. If one's average heart rate was 150bpm, P:HR can range from 1.30 to 1.36 and simply reflect the error range of the meter. Meters with 2%, 3%, and 5% would produce very wide P:HR ranges without significance)

Since the changes I see in my P:HR ratrio for long steady state rides are at or beyond the second decimal place, I haven't been confident about making any conclusions regarding fitness from them. I do, though, know that when I see a drop of >0.05 (e.g. from 1.35 to 1.29) on a route that I have done several times before under similar steady state circumstances I am either carrying alot of fatigue in the legs, or (in one notable case with an inexplicable drop of 0.1) I am in a presymptomatic phase of getting the flu.

Have you used P:HR as a metric of excessive fatigue or risk of overtraining? I expect that absolute values would be unique to the individual.

Thanks for your thougths and travel safely !


I think an ooops is in order - I didn't get the math right in my example. My example was for my Quarq Cinqo, which has a stated accuracy of 2.0%. The ideal circumstances with a PowerTap and 1.5% accuracy would yield a power range of 197 - 203, or a P:HR range of 1.31 - 1.35. Still broad enough that I don't know what to do ... unless I have made another math error some where ... Sorry !


Hello Joe,

I wanted to verify what I should be using as my output metric since I do not have a power meter but I do have a Garmin Edge 500; I am assuming I should use average speed. Is my assumption correct?




Any development in movement-economy could make a big (or huge, depending on sport) contribution to an increase of the output-input ratio.

While I know that you are big on sport specific performance, if the goal is aerobic endurance AND fat-loss, it may be a good idea to, some of the time, abandon sport specificity and sport specific movement-economy: any new aerobically demanding activity will be more taxing for the participant as he/she hasn't developed any level of economy yet.
Being a new, not adapted-for, activity, it may also be more efficient for metabolising body fat.

My question to you is, roughly, in percentages, how sport-specific do one need to be? (my guess is that you'll answer with the good ol' 80-20 rule, please don't).

Gary D.

I have a hard time understanding that HR is an input tool. I believe it to be the dependent variable in a situation, if you run harder then your HR increases. If you run slower, the HR decreases. It is similar to 'If you run longer, your distance increases' or 'if you give more effort, your pace increases." Whereas I accept that comparing one day to the next in HR may not always be accurate, HR is still dependent on effort. The same can be said for pace, distance, and watts. Maybe I misunderstand something here about a relationship between variables and input/ output.

Ryan Rodman

What do you think ultimately determines the leveling off point for the output:input ratio? Do you think it is the overall weekly time spent on the bike working on aerobic fitness? That is, if an athlete increases their weekly training time 10% one year, their ouput:input ratio will increase by 10%, or by a proportional level? Proportional because, I do not know, but realize there are likely diminishing returns the more time is spent training. Also, this assumes I suppose, that their bodies can tolerate an increase of 10% in training time.

richard Sterry

This post describes something I’ve been doing for a while, but I add another dimension. I assume Joe wants us to do a similar circuit for consistency, where I tend to vary my off road routes from time to time.

The formula I use is: Average Speed (including stops) divided by Average Heart Rate. Then I multiply the result by the number of feet climbed per mile. Using this latter dimension factors in the hill climbing. This Performance calculation enables me to track improvement over the year, and to compare year on year results.

It may all sound complicated, but by entering in the basic details to a spreadsheet, all the calculations and charts are automatically performed.


Hi joe tks for your hard works.
I have read all article with great attention but i havent understand what kind of speed or power must use in the formula. You write " It’s simple math. Just divide your output (power or speed – not pace) by your average heart rate for the workout"
What speed or power must use? Average or Max? I think average but in not sure.


Hi Joe, congratulations for all your posts, they're just amazing.
As mountain biker & trainer of myself, I'd never used these kind of methods, but they are in somewhat similar to those used in oder technical disciplines. As an engineer, this imput/output comparison is done almost everyday. Totally agree that every measurable variable is more likely to be improved.

Are there any ways of measuring progress at oder stages of the training ?

Thanks a lot for sharing your knowledge. Keep educating us !!

Ed Schaffer

Thanks, Joe. As a someone who, while not a racer, takes training to heart, I've been looking for a single measure that helps me get a sense of general improvement. Any chance of putting this in TrainingPeaks?


Harry Mattison


Wish I could be joining you in the Canary Islands, but in the meantime I will keep riding my bike in the basement here in rainy & snowy Boston.

I was wondering how many sets/session you reccomend for squats during the MT phase.

Also, could you share your opinion regarding box squats vs. free squats?


Joe Friel

Harry--2-5 sets depending on a lot of variables. Don't know you well enough to give a more precise number. By box squats I assume you mean squat till butt touches box. Ok so long as not going below knee bend when seated on bike and pedal at 3 o'clock.

Joe Friel


Joe Friel

Ryan--What determines improvement in O-I ratio is what always determines changes in fitness--aerobic capacity, anaerobic threshold (as % of VO2max) and economy. As any of these improve (assuming the others at least remain the same) O-I increases.

Joe Friel

Gary--aerobic fitness improves whenever power or speed improves relative to a given heart rate or HR declines relative to a given power or speed.

Joe Friel

Chris--For competitive athletes, performance is all about specificity in the last 12 weeks or so before a race. The farther away from the race one is the less specific training may be.

Joe Friel

Cor--for running yes. For cycling no.

Joe Friel

Madeleine--what you're looking for is trends over a long period of time. This will take such tolerances into account and minimize their effects.

Joe Friel

Barb--1. Only works if you can calibrate speed (resistance). 2. Yes.

Harry Mattison

Thanks Joe for your quick response and everything you do do encourage cycling excellence.

In case it helps more specifically determine how many sets I should do, I am 37 yrs old, 6'1" tall, 175 lbs, have never lifted weights before, have raced for 1 year (cat 4), and hills are a big limiter for me.

Yesterday I did a session of 100 lb squats, 15 per set, 3 sets. After that session I felt like I couldn't do much more. Tomorrow I was planning to try 110 lbs.

Joe Friel

Harry--3 sets is probably best for you now given that you are new to it. Be sure to allow for recovery for at least 48 hours.


sweet! While I read your post, I realized that I figured this one out independently a few months ago. In my training log I compute what I call "economy-factor" for every run i do.

I divide pace by average heart rate.

I have this set up in excel. There also is a nice interpretation for the unit of the resulting value: If you write pace in Meters per Minute, and then divide it by heart rate (beats per minute), you get Meters per beat.
So it tells you how far you go, while your heart beats once. Typical values I've seen in my runs are around 1.4 m/b.



Based on your answer to my earlier question I assume that you are saying that the only way to properly compare input and output on the bike is to have a power meter; is my assumption correct?

Joe Friel

Cor-Yes, probably best. But could be done on an indoor trainer if speed is calibrated.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name is required. Email address will not be displayed with the comment.)

Joe Friel's Sponsors

Blue Competition Cycles
"Specialized Bikes is my newest sponsor. I’ve been impressed with their bikes for a long time so I tried one. Had to have it. What a great ride!"

D2 Shoes
"I’ve been using custom-made D2 Shoes for years. Currently on my fourth pair. And, yes, they have midsole cleats as described in my blog."

ADS Sports Eyewear \
"ADS Sports Eyewear makes all of my glasses for casualwear, reading, cycling, golf and more. Hundreds of models from all the major manufacturers."


Subscribe via Email

Enter your email address:

Delivered by FeedBurner