Some time ago I wrote about recovery on demand. This is a method of training in which recovery is not planned in advance, but rather done when the need arises. For the athlete who is good at self-monitoring, this is a very effective way of recovering as it maximizes one’s use of training time. Unfortunately, many athletes tend to ignore their body’s signs of overreaching and with such a method would probably press ahead without ever allowing time for recovery, eventually resulting in full-blown overtraining.
“Periodization on demand” is a similar concept. Most athletes think that periodization is a rigid system in which a training plan is created which must be followed regardless of all other factors. My books have probably given that impression as everything on the plan is quite detailed and time-specific. I still think that’s a good idea as it provides a roadmap for where you are going. It doesn’t, however, mean that you must follow it without change. There can be several factors that require straying, such as illness, injury, periods of mental stress, unusual demands on your time, and more. One such factor that is seldom discussed is rate of adaptation – how quickly your fitness changes.
Physiologically, the purpose of training is stress the body with some combination of workout intensity, duration and frequency causing it to adapt. We call this adaptation “fitness.” The planning of periodization assumes the body will achieve a given level of fitness at a given time by following the plan. That may well be the case. Historically, the problem with this assumption has been that there was little in the way of data to confirm that the planned adaptation had occurred. That’s now changing due to technology.
A power meter for cycling or a GPS for running has taken some of the guesswork out of the measurement of adaptation. Such devices (there are more to come) allow you to more accurately measure performance changes – if you know what to measure. (I touched on this idea in my last blog post.)
Typically, in the Base period an endurance athlete wants to improve aerobic endurance, muscular force and speed skill. The first two can be measured using some combination of a heart rate monitor, power meter and GPS. Speed skill is still difficult for us to measure in a field test. But expect that to change when second generation power meters – yet to be released – provide more analytical information on pedaling mechanics such as individual leg contribution to power output and the range of effective force application to the pedal per stroke. In the mean time, we can easily measure aerobic endurance and muscular force adaptation.
I’ve previously written about the “Efficiency Factor” (EF) as a way of gauging changes in aerobic endurance. It’s based on the simple concept that as aerobic fitness improves, heart rate decreases at any given power output (or speed in running) [Lucia, et al, Heart rate and performance parameters in elite cyclists: A longitudinal study. Med Sci Sports Exerc, 2000, 32(10):1777-82]. Or, to reverse that, if heart rate stays the same, over time, power (or running speed) will increase as aerobic fitness improves. Heart rate by itself tells us absolutely nothing about fitness. It must be compared with something to have meaning.
This brings us back to the idea of periodization on demand. The optimal way to train, I believe, is to frequently measure your adaptation changing your training only when it’s evident that fitness has plateaued at a higher level or achieved a predetermined level. By doing this you take the guesswork out of training and move on to a newer form of stress only when your body says it’s time rather than when the plan changes. Again, this doesn’t mean don’t plan. Follow it, but be willing to change it when the time is right. This will usually require that you modify the plan going forward.
How about I give you an example of this from my own training.
I use a block training form of periodization. With this method the focus of training during any given period (“block”) is quite focused with generally only one or two aspects of fitness being addressed. In block 1 this fall I focused on aerobic endurance as measured by EF. The accompanying chart shows the progression over the course of four weeks (Oct 17-Nov 17). During this time I did the same EF workout 17 times. This involved warming up for 30 minutes and then riding one hour at a fixed aerobic threshold heart rate in low zone 2 (120-125, in my case), followed by a 30-minute cool down. I used the same two courses for these sessions and the same equipment with the workouts at about the same time of day. After the workout I compared normalized power for that hour with average heart rate for the same hour (NP/AHR). That produced a ratio that ranged from a low of 1.29 to a high of 1.45. You can easily see the progression of EF in the chart. Only the EF workouts are shown here (the blank days were missed workouts due to travel, easy recovery rides, or a cycling camp from Nov 1-3).
By the middle of November it was obvious that my aerobic endurance (as reflected in the EF ratio) had plateaued. Power had increased at the same heart rate but was no longer rising. It was time to move on to the next training block, which I am currently in.
The takeaway message here is not block training or aerobic endurance training, but rather making changes to training when your body says it’s time, rather than when the plan calls for it. Some day we’ll have software that does such monitoring for you and suggests when it’s time to change your training and what those changes might be. Until then you need to pay close attention to that which is appropriate and measurable.

Joe, it seems like the chart link is broken. Thanks.
Posted by: Fabiano Araujo | 12/06/2012 at 08:26 AM
Thanks for this reminder Joe, I tend to try to make and follow plans too closely. I noticed that in your Power Meter Handbook you also suggest the use of Normalized Power in the calculation for EF, but in your Training Bible (chapter 4 I think) you suggest the use Average Power (not normalized). Does it matter?
Posted by: Buzz | 12/06/2012 at 09:23 AM
Fabiano--It's not a link. For some unknown reason that wouldn't work today.
Posted by: Joe Friel | 12/06/2012 at 09:33 AM
Buzz--I don't recall that in any of my Training Bibles (which one were you referring to?). But, regardless, it's NP/AHR.
Posted by: Joe Friel | 12/06/2012 at 09:34 AM
I may have misread. It was in the Cyclist's Training Bible, 4th edition. Now that I think about it, I believe it was in the same section where you described decoupling, so you would have stated that VI wants to be below 1.05. If I keep my VI below 1.05 my AP should be close enough to NP that I could use AP, right? Reason I ask is that Strava doesn't provide NP.
Posted by: Buzz | 12/06/2012 at 09:39 AM
Hi Joe, always love the wisdom dished out here, thank you. I assume measurement of EF for running would be something like NGP/AHR as opposed to AP/AHR or similar? Most of mine are on a treadmill at the moment so doesn't matter for me but just curious
Also curious if you have any silver bullets for measuring stuff in the water. Only HRMs that work there and record seem to be either the older Sxxx Polars or the pricier current R models, and neither of those know anything about swimming, where the Garmin 910XT knows swimming but uses ANT+ for heart rate and that can't go through water...
Posted by: Mike Hardy | 12/06/2012 at 02:07 PM
Buzz--Yes, you're right about low VI. aP and NP are about the same.
Posted by: Joe Friel | 12/06/2012 at 02:17 PM
Mike H--Yes, correct about running. Don't have any good ways in the pool yet.
Posted by: Joe Friel | 12/06/2012 at 02:19 PM
Hi Joe, is it a safe assumption that most peoples AeT is also in LOW zone2? I'm doing most of my base riding at the upper end of zone 2 POWER which is a lot more difficult than lower zone 2 POWER if I do it for long enough (3,5h @ 215 watt is way more taxing than 3,5h @ 195 watt...).
Would you say that's worth it or would I get the same benefit if I stayed in the lower half of zone2? THANKS
Posted by: triprentice | 12/07/2012 at 10:31 AM
If you are measuring pw in training peaks is it also necessary to measure the EF?
Thanks Joe
Posted by: Joe Hamilton | 12/07/2012 at 04:06 PM
Joe H--TP does the EF determination for you. It's on the race graph page.
Posted by: Joe Friel | 12/08/2012 at 05:19 AM
Triprentice--I like to have athletes use low z2 in the early base period. Aerobic threshold is assumed to be, for this purpose, about 30bpm below LTHR. Testing with gas analysis would be a more accurate way of determining AeT.
Posted by: Joe Friel | 12/08/2012 at 05:22 AM
thanks for your answer! you did no longer rides of 3h+ as recommended in your books. aren't they THAT necessary as long as the total weekly volume in that zone is appropriate? I'm curious because I can't ride outdoors in base1+2 due to weather conditions and several 1-2h rides on the trainer are no problem but 2-4h is mentally impossible for me... ;-)
Posted by: triprentice | 12/08/2012 at 11:44 AM
Triprentice--Well, I can guarantee you that I wouldn't be doing 4h rides indoors. I've done them, many years ago, and it wasn't any fun.
Posted by: Joe Friel | 12/08/2012 at 01:07 PM
and what is your opinion on doing short&hard trainer sessions in the winter and just do the long rides as soon as they're possible outdoors? As far as I can remember you mentioned in several of your books that it's counterproductive to do ME workouts before AE is fully established...is it really so critical? my hope would be to train the "cycling legs" with low cadence threshold intervals and let the other workouts (swim, run, XC skiing) take care of the cardiovascular system - only until I can do long outdoor rides again
Posted by: triprentice | 12/08/2012 at 01:13 PM
Tripe entice--Threshold and sub-thresh work is ok.
Posted by: Joe Friel | 12/08/2012 at 01:39 PM
Joe, I listen to the TB podcast every week and have read your training bible. Love them. I have two questions:
1. Did I read in your "bible" that it could be advantageous for some new athletes to train in the base period for up to three years?
2. How does Maffetone fit in your base plan and/or AeT?
Thanks
Posted by: Gene | 12/09/2012 at 05:59 AM
Gene--A newbie may need to 2-3 seasons to build a good fitness base given how little volume they can manage initially. Maffetone and I often see things the same way I've found. But not always.
Posted by: Joe Friel | 12/09/2012 at 06:53 AM
Thanks Joe. If I understand you, my MAF puts me in zone three. You would suggest that I back down a few beats to develop a good base, right?
Posted by: Gene | 12/09/2012 at 08:33 AM
Joe, I really enjoyed this timely post. How do you figure in a rest period during a longer ride with regards to decoupling? Yesterday I went out for 3 hours. Minus warm up and cool down I rode for 1:13 with NP @ 84% of FTP while HR was 86% of ATHR and Pw:HR was 4.35%. After a mini cool down, 5 min off the bike to eat a bite and a 3 min warm up my trip back took 1:17 with NP @ 75%, HR @ 88% and Pw:HR @ 0.35%. I have two fairly good periods, but if I include the rest period my Pw:HR plunges to 8.4%. If the rest period is included, I have a ways to go to get my decoupling down. If it's not included, which period would you give more relevance to, the harder first period or the easier 2nd period? With regards to the subject of this blog, if you are in early Base 2 and staying coupled at the longest distance/time you expect to race, should you add some Base 3 workouts and basically do 6 or 7 weeks of Base 3? Thank you very much for your answer.
Posted by: George Coffey | 12/09/2012 at 11:42 AM
Gene--Well, of course it depends on whose z3 we're talking about. A good estimate of the number is to subtract 30bpm from your LTHR.
Posted by: Joe Friel | 12/09/2012 at 12:23 PM
George--I'm afraid I don't fully understand. Your decoupling was well under 5% both before and aft the break you took. But you say something about it "plunging" to 8%. That's kind of high and wasn't reflected in the numbers you gave me. I'm probably missing something obvious here. As for second question, I really can't say how much base 3 an athlete should have as there are several variables.
Posted by: Joe Friel | 12/09/2012 at 12:29 PM
Joe, I'm sorry I wasn't more clear. What I was trying to determine was whether it's more accurate to look at decoupling for two separate steady state efforts or to look at the two efforts plus rest period as a single long effort. As far as base 3 training it's not how much, but like in your example using your AE as measured by EF, if you reach the goal of base 1 and 2 (remaining well coupled for the maximum time you intend to race)would you recommend moving into base 3?
Posted by: George Coffey | 12/09/2012 at 01:46 PM
George--Regardless of whether you call it two sessions or just one, it is still only one or two data points. It takes many such data points to make a decision about whether or not to move on in your periodization. So I don't think it's significant. Once you have achieved all you intended to achieve in a period/block its time to move on to the next. That's an easier decision in block training as there is usually only a single focus, such as aerobic endurance. In traditional periodization a period has several abilities to focus on, such as AE, muscular force, and speed skills. So such a decision is just a bit harder to make.
Posted by: Joe Friel | 12/09/2012 at 04:22 PM
I've been tracking decoupling for 6 months now. My decoupling (np/hr) varies from 1.19 to 1.45. If I really spanked it on a hilly it would be more like 1.60. So I think this needs to carry a health warning that to interpret this as meaningful data you really need to ride the same flattish route with similar wind conditions, temperature and similar NP. Even when I do that, I'll get 1.3 one day and 1.4 on another day. Or perhaps restrict it to turbo (I don't use turbos or rollers so thats out). Still I'm always looking out for decoupling PB's and if its a highish figure, or if the 1.4s start appearing more often, thats a positive, but qualitative, subjective, not quantitate and objective !
Posted by: Paul | 12/11/2012 at 02:12 AM