Recovery Phase and Power Meters
Discussions of PM always seem to center on the "ON" phase of the interval. There is no mention of the second part of the interval, the recovery. There is obviously phyiological information there. Has it been studied and then dismissed or just not studied?
A variety of adaptations must occur if you are to make progress. Since the basic game plan is STRESSES - ADAPTATION - NEW STRESSES, gauging how well you have adapted to the current stress will let you determine when you can move on to the new stress. If you don't increase the stress after the adaptation, you don't make progress. So how do you determine when adaptation has occurred?
AFAIK, PM training requires periodic testing to measure adaptations. The 20 minute tt seems to be a std test. You take your average (or some other numeric meaure) of your power during the interval and that becomes the benchmark for determining your zones. Although it is based on a subjective determination (PE), it seems to be a good idea and people seem to believe in it and seem happy with how it works. After a certain amount of time, you re-test and see if you have made progress (adaptation) as measured by a higher power benchmark. The frequency of testing is important. For example, if you test every 6 months but your adapatation occured after 3, you missed the opportunity for 3 months of a new stress-adaptation cycle. Also, adaptation is different for different people and so using an out-of-the book testing schedule may not be best for everyone. So how do you determine when to re-test and how do you ensure your PE is the same on the re-test, unaltered by incomplete recovery, energy level, cadence (if wattage varies with RPM), pre-ride diet, external stresses, etc? It doesn't seem to get much discussion but I'm guessin' most people eyeball it. There is obviouvly plenty of room for mistakes in the timing of the testing and use of PE for benchmark determinations.
As noted in a different post, part of the utility of HRM is in measuring recovery. i.e. changes in the rate of recovery during the "OFF" phase of interval training. The recovery rate increases (recovery time decreases) with improved fitness. As most people find out empirically or know intuitively, better recovery measures progress (adaptation). For example, early in the season you can only do a set of 3/4 intervals x 3, twice a week. (I designate "on" in the numerator and "on + off" in the denominator. This reminds me to think of recovery as part of the interval.) As your recovery improves, you progress to 3/4 x 3; 3/4 x 1. i.e. you recover completely and add an extra set of 1 rep. After doing that for a while you add that rep to the first set and it's 3/4 x4. Then 3/4 x 4; 3/4 x 1 for awhile and then 3/4 x 5, etc. If you watch your HR, you might notice that your recovery rate increases increases though this sequence. For example, you find that your last rep of 3/4 ends with a HR of 130. After a couple of weeks it is at 125. Time to bump up the stresses.
Even those using HRM might not use this traditional approach and just rely on good ol' PE to determine when their adaptation has occurred. In either case, what advantage does a PM have in determinig when you ramp up the stresses when the adaptation is based on periodic testing with PE? No need to re-state how PM helps during the "ON" phase. No need to discuss what the adaptations are (ex. "lactate clearance").
I know ex fizz but admit to not having researched this specific subject. I'm actually looking for the real life experiences of experienced cyclists.
Originally Posted by lgh
This is a tricky thing, and one thing i believe I have gotten better over the last couple of years. Larger training load will introduce stresses with immediate lower performance, then a subsequent lower load will yield improvement/adaptations and higher performance. I believe I've never made a strong commitment to the "smaller-load" period. Now i let the "smaller-load" period go about a month and find that adaptations (i.e. speed) will improve for longer than I thought. I think I had a tendency to drive myself into the ground as soon as i had a good performance, when in reality, continuing the "smaller-load" period would yield even more improvement. Committing to the PEAK period can be a difficult thing.
That is why I prefer to test with the Conconi method. Easier to test (therefore you do it more often) logistically, and I believe it takes some of the external influences (such as fatigue) out. But I find that power testing doesn't really affect my interval power targets. Intervals are just inherently performed "to the highest of my ability on that day". I always try to get the best (and flattest) power numbers possible during intervals. In essence, intervals becomes my testing, and I always expect lower numbers during high-load periods; and hopefully get better numbers during lower-load periods.
Originally Posted by lgh
HR monitors are good for measuring post interval HR recovery rate; PMs not so much, especially when you have 100% recovery in between intervals (which I believe is a good thing). But it is good for Microburst/Tabatha type intervals since you can see how well you are holding the "recovery" power level. For example, you do 10s on, 20s off for 10 straight minutes. During the 20s off you try to hold mid Zone 2. The PM will indicate how well you are holding that Zone 2 power after repeated hard 10s efforts. During the post analysis you can also see what the normalized power is for those 10minutes. So if you are recovering quicker then you should have a higher NP; and this phenomena could have no corresponding improvement to your LT power.
Originally Posted by lgh
Hope I understood your questions.
some intervals require a recovery that is more of a work intervals, i.e. riding a bit below FTP then doing a 1 minute hard effort every 5 minutes, then returning to the hardish pace. The PM is great for that kind of thing, as is the ergo.
I guess for most of the other stuff, your recovery is just milling around for a set period that allows you to recovery fully or partially and what you care about is how much work you can do off the recovery.
If I was more organized I would be interested in looking at the HR vs recovery. I use the garmin runners watch and have found few useful metrics in the software that comes with it.
I have noticed that in the summer, environmental factors seem to impact recovery quite a bit. One thing that helped me a lot when I was learning to use the ergo was doing the same warmup each time and doing 5 minutes at a set wattage at the end of the warmup. My HR after that 5 minutes was a very reliable indicator of if I should even bother with the workout.
It depends how you construct your intervals. Most power meter head units are also heart rate monitors. If you wanted to then you could use power output for the work period and wait until your heart rate dropped below a certain level before starting the next interval.
Intervals using power output are usually based on set periods of time. The work periods are defined as a set amount of time at a given wattage and then the rest periods between are a set amount of time at or below a particular wattage. The problem with power during the recovery period is that if you stop pedalling then power output is zero which doesn't tell you much.
This is an example of a 1 hour interval workout posted by Andrew Coggan on another forum where the rest periods are simply described as "easy":
"Seriously, the best season I've had in recent years followed a winter
during which I did the following '90/90'90' workout 3 d/wk:
5 min w/u
20 min @ 275 W
5 min easy
5 min @ 325 W
2.5 min easy
5 min @ 325 W
2.5 min easy
0.5 min at 500 W
2 min easy
0.5 min at 500 W
2 min easy
0.5 min at 500 W
2 min easy
0.5 min at 500 W
2.5 min easy
5 min warm-down
The '90/90/90' refers to the fact that (almost by chance) the powers
used were about 90% of the best that I could produce for that duration
when at peak fitness. That made the session challenging enough that I
didn't lose too much fitness over the winter, but not so hard that I
ever dreaded the workout or burned out from doing it." Andrew Coggan
The issue of getting useful and comparable data from your ftp tests is something which you have to work on. Finding a route which you can repeatably ride in a similar way is the important thing. Different routes will give you different results so its best to find one route and stick with it. perryr's very helpful advice on the subject is what I try to follow:
This year I've been doing an ftp test every 1-1.5 months, along with a few full 1hr time trials for a second opinion. When I do an actual 1 hour time trial I can't get anywhere near the predicted power output (95% of my best 20 minute power) however.
95% of 20 min for an hour ain't too shabby!
Let me try explaining recovery in a different way. If Cogan's intervals were so useful, why didn't he do 7 per 2 weeks instead of 6 or add on a couple more 500 watt efforts with each set? Or, why not just do one per week? Because if he can have more interval days per unit time he can advance faster but if he does too many .... well it won't work. That's intuitive. One of those intervals per month ain't helpin' much, two is doing more, three even more, 20 is too many. So what is the limit and how is it determined? The limit is determined by recovery. That will limit how many you can do per periodic cycle. The more you can do per periodic cycle without burning out, the better. That has led to one my favorite ex fizz expressions, "To get better, do difficult things, as often as you can, without getting tired or injured." Tired in this context means incomplete recovery between sessions.
Everyone who has ever done intervals knows that you can do more of the same type as conditioning improves. That is, improved recovery is one of the adaptations that occur with training and that allows you to crank 'er up and increase the stresses (intervals) for the next cycle. So, given the central role of recovery in development, how is it measured with PM training? To me it looks like PE, "I couldn't do another one." Nothing wrong with that. On a similar note, how do you determine frequency of testing if it's not linked to recovery changes? Again, PE or out of a book seems to be the way it's done.
Incidentally, a way to measure recovery with a HRM is as follows. In my 3/4 example it is taken at minute 3 and minute 4 OF THE LAST INTERVAL OF THE DAY. By definition, that is the last one you can possibly finish. If you can do more, you need to do them. (Rule #5) You might get something like 165/120 as the two numbers. Those go in the diary. The only other way I know to measure recovery in the field is PE changes.
The Andrew Coggan intervals were just showing what he had written down for the recovery periods. The length of the session was restricted to an hour long workout. I'd guess that not doing more frequent or longer interval sessions during the winter might be due to not enjoying turbo training much, or maybe it had to be fitted in around other commitments.
In the book "Racing and Training with a Power Meter" their guidance for recovery is that if you can't manage the prescribed power output when nearing the end of a set of intervals then that is a sign you've done enough for that session. Progression is seen as being able to do the same duration interval at a higher power output. As you improve you increase the target power output for the interval.
If you haven't read it then "Racing and Training with a Power Meter" by Hunter Allen and Andrew Coggan (second edition now) is well worth a look through.
Training and Racing with a Power Meter: Amazon.co.uk: Hunter Allen, Andrew Coggan: Books
Yeah, I gotta read that. But I like talking to the people who have actually tried this stuff and match their experiences with what I know of principles of ex fizz. I've noticed through the years that people who actually do this stuff (try to go fast on bikes) seem to fall into 2 categories. Those who like to use the data because they are curious and like the added interest of using data and those who use it and sort of dismiss it because they already figured it out without the investment in monitoring equipment or the equipment was used for a while and is now gathering dust on a shelf. My questions usually center on why are both goups successful.
You have been following my posts on a few of these topics and probably figured out I'm curoius or curious about the data and pretty heavily rooted in science. But I love these guys who use PE and I'd rather buy a Moots than a $1200 rear wheel.
A powermeter has its uses but is still a tool. It's not something which guarantees an instant performance boost just by having one fitted to your bike.
You might have seen this study from 2009 before. They had two test groups doing the same training, one using power meters for pacing and one using heart rate. There wasn't a statistically significant difference in outcomes between the two groups.
Here's the full article as a PDF file:
Original 19 page thread with bad tempered arguments.
Study which show us that training with PM does not make you faster - CyclingNews Forum
Reply by authors on specific points
CyclingNews Forum - View Single Post - Study which show us that training with PM does not make you faster
Effects of high-intensity training by he... [J Strength Cond Res. 2009] - PubMed - NCBI
Effects of high-intensity training by heart rate or power in well-trained cyclists.
J Strength Cond Res. 2009 Mar;23(2):619-25.
Swart J, Lamberts RP, Derman W, Lambert MI.
Department of Human Biology, Faculty of Health Sciences, UCT/MRC Research Unit for Exercise Science and Sports Medicine, University of Cape Town, The Sport Science Institute of South Africa, Newlands, South Africa. email@example.com
"The aim of this study was to determine whether the performance of cyclists after 4 weeks of high-intensity training improved similarly using either heart rate or power to prescribe training. Twenty-one well-trained men cyclists (age, 32 +/- 6 years; peak power output, 371 +/- 46 W) were randomly assigned to a power-based (GPOWER) or heart rate-based (GHEART) high-intensity training (HIT) group or a control group (GCONTROL).
Training consisted of 8 repetitions of 4 minutes at either 80% of peak power output (GPOWER) or at the heart rate coinciding with 80% of peak power output (GHEART), with rest periods of 90 seconds. A 40-km time trial and VO2max test were performed before and after 8 training sessions. There were significant improvements (p < 0.05) in peak power output (GPOWER = 3.5%; GHEART = 5.0%) and 40-km time trial performance (GPOWER = 2.3%; GHEART = 2.1%) for both of the high-intensity groups.
Although there were no significant differences between groups for these variables, when the data were analyzed using magnitude-based effects, the GHEART group showed greater probability of a "beneficial" effect for peak power output. The current general perception that prescribing training based only on power is more effective than prescribing training based on heart rate was not supported by the data from this study. Coaches who are unable to monitor progress frequently should prescribe training based on heart rate, when intervals are performed under stable conditions, because this may provide an additional advantage over prescribing training using power."
Last edited by WR304; 09-10-2011 at 10:49 AM.
Yes, I get that journal but completely forgot about that article.
In a sense, intervals are intervals. You kinda do them the same regardless whether or not you're looking at a number. Their success is determined by a couple of things. One is your PE. Did you or did you not suffer properly during the "on" phase? Also, as Eddy B said in his book, "The key to intervals is knowing when to start the next one."
I think I'm going to rent a PM from FastCat and play with it. Steve Johnson once said you should be able to sustain 250 watts for prolonged periods of time. I really need to see where I am with that.
This is the Nimmerichter study they refer to in the Cyclingforums thread:
"Longitudinal monitoring of power output and heart rate profiles in elite cyclists":
Longitudinal monitoring of power output and hea... [J Sports Sci. 2011] - PubMed - NCBI
Study on the usefulness of a power over other metrics - CyclingNews Forum
When it comes to shorter intervals then there isn't really much difference between what I did before I had a heart rate monitor, with a heart rate monitor and now with a powermeter. I like this quote from the above thread as it sums it up quite well.
"I've said it once, I'll say it again:-
Before power or HR I did:-
3hr + LSD
After power or HR:-
3hr + LSD
It's just now I have more quantifiable data from said rides. The physiology and it's, and more specifically "your", adaptations from stimuli don't change, just the ability to track it." Tapeworm
Where I've found having power data particularly useful is being able to directly compare between rides, regardless of whether the bike has offroad tyres or slicks fitted. I couldn't believe just how poor the power outputs from some of my offroad rides were. It's something I've been working on to try and improve.
It's also been good for looking at the big picture via the WKO+ 3.0 PMC (Performance Manager Chart) for tracking training load, intensity and recovery over the year. That's showing some interesting results.
One thing to bear in mind is that the software you download your recorded data into makes a big difference as to how easy it is to interpret any power data. Cycleops Poweragent and Garmin Connect/ Garmin Training Centre for example aren't great.
WKO+ 3.0 has some good points (being able to quickly search through all your files on keywords is great) but then spoils it with some incredibly shoddy sections (most of the report options, multi-file analysis) and essential missing features (you can't overlay one ride on top of another to read off the individual data points for example). It's what I've been using as it seems to be the least worst option but for a lot of things you'll still need to export the raw data into Microsoft Excel and create your own reports.
Whatever program you use it's worth typing out a detailed ride description for each ride file (route, wind, conditions, how you felt etc) as it's difficult to remember these things. Adding plenty of notes and lap markers is useful as well so that you can compare exactly the same place later. I'll always put a manual lap at the start and end of each climb and also for sections of road/ trail that I ride regularly. A well laid out file with plenty of detail means that when you return to it a few years later you'll immediately know what it represents.
Thanks. Without having read the book, I'm seeing its utility more for tracking data, finding weaknesses, and periodic testing than for actually executing a workout. I already know how to do that. Time and recovery limits me more in that regard than anything else. You're finding some interesting applications regarding equipment testing. I have some time off in a couple of weeks and will try to get a rental by then. Looking forward to it.
Will a Garmin 500 get the job done from a computer point of view?
If you have a Garmin Edge 500 then you can import the .fit files (set to 1 second recording) from it straight into many of the popular software programs which you would also use to look at power. The information displayed will be exactly the same, apart from missing a power trace.
These are all free downloads so you can look at the programs using your own data and some of the example power files around.
WKO+ 3.0 Trial
Analysis Software for Training Files from: Power Meters, GPS Devices and Heart Rate Monitors | TrainingPeaks WKO+
Golden Cheetah: Cycling Performance Software for Linux, Mac OS X, and Windows
Lots of power files from races that you can download and look at in WKO+ 3.0:
Files & Analysis of Races
I'm using a Garmin Edge 500 with my Powertap hub at the moment. It works ok so long as you never stop the timer and manually cut out the stationary time post ride.
Post #120 on for my settings and comments on Garmin Edge 500 with a Powertap
The Garmin Edge 500 display whilst riding isn't as good as the CycleOps Joule 2.0. The Garmin's 3 second display smoothing is slightly too low. It will take you at least a few rides to get an idea of what to look for in the jumpy power numbers.
This post is quite a good example of how I've been using power data as a way of learning lessons and improving for the next time I ride a route.
By road dirt in forum GPS, HRM and Bike Computer
Last Post: 01-22-2008, 01:38 PM
By mbabaracus in forum XC Racing and Training
Last Post: 07-17-2007, 09:06 PM
By cgee in forum XC Racing and Training
Last Post: 09-04-2006, 06:49 PM
By Beezer in forum XC Racing and Training
Last Post: 11-28-2005, 06:00 PM
By Ultra Magnus in forum XC Racing and Training
Last Post: 09-28-2004, 01:55 PM