I did a 5 min and 20 min power test this week to estimate my current critical power for training and racing. Last time I tested was Spring of 2011, and in the time between then and now I tore my hamstring training for Vegas. As a result I spent about 6 months without doing any hard or serious training and have only really been able to get back to real training in the past 4-5 months.
Anyway, my tests this week resulted in a 3% increase in my 5min power but a 2% decrease in my 20min power. I was glad to see that I hadn’t lost too much during that time I spent injured, but I’m a bit curious how well I can compare the results. Last year I tested with a Quarq, whereas this year I did my tests with my older PowerTap, due to the Quarq being on a different bike that’s in the shop right now.
How much would this have affected my results? When I get the Quarq back on my TT rig should I go with the tested FTP or do I need to retest?
Assuming both the Quarq and PT are calibrated and torque zeroed you’d expect the crank based Quarq to read about 2 to 3% higher or typically about 6 to 8 watts higher than the rear hub based PT just due to drive train losses.
The fact that you didn’t see that much loss even after the time off is a good sign.
You don’t really ‘need’ to retest your FTP anytime soon. You shouldn’t really be training with your eyes entirely glued to the PM and regulating every power fluctuation. Instead do, what you can do when you train. Use the PM to set rough targets as in 90-95% of FTP for low Threshold efforts (like 2x20s, 2x30s, 1x60s, etc.) perhaps target a bit higher for shorter Threshold efforts (4x15s, 3x 12s or other variations). etc. But still take into account how you feel on a particular day and adjust as necessary including being willing to go a bit harder on days you feel great if you’re not intentionally holding back for instance while peaking for an event or saving some for a very intense session later in the week.
Basically train ‘with’ power not ‘by’ power it’s a meter to objectively measure what you accomplished during training but does not have to be a ‘training regulator’ or ‘effort governer’ that needs to be strictly followed unless there are specific reasons to keep a cap on things. Approached that way this stuff is pretty much self adjusting, as your fitness improves you’ll see things like your regular repeatable 2x20 power climb, on days when you’re more fatigued or stressed from other activities you may back have to back off but that’s totally normal and part of the ebb and flow.
If you rely on things like workout TSS and from it CTL and ATL in something like the WKO+ Performance Manager to help manage overall workload then it does help to have a reasonably accurate FTP estimate but it still doesn’t have to be nailed down to the nearest watt (which is basically impossible as most commercial PMs are only accurate to +/- a percent or two or in the vicinity of +/- 3 to 5 watts at typical intensities). So don’t overly stress having your FTP totally nailed down as long as you’ve got reasonable target ranges to roughly guide your workout intensity and then adjust them as necessary out in the field.
So no, I wouldn’t run out and retest when you get your Quarq back. I’d just get to work and then bump your FTP estimates up or down based on what you see day to day and week to week in things like long Threshold intervals during training. For instance if you’ve estimated your current FTP at say 280 watts but you start punching out 2x20 or 3x20 or 2x45 or even 1x60 sets at say 300-310 watts on a regular basis then it’s a pretty good sign that your FTP is due for a boost. If you struggle to get though even one 20 minute effort at 280 watts on a reasonably fresh day or can’t complete a 2x20 set at 255 watts or more (~ 90% of FTP) during weekly training then it’s a pretty good clue your FTP isn’t as high as 280. Pay attention, adjust as you go and as you see repeatable patterns and track your progress or lack thereof, lather, rinse, repeat.
Thanks for the thorough response. I’d like to think that the results indicate that I may even be a little ahead of where I was the last time I tested, but I’d rather stay a bit conservative with the training to ensure I’m getting proper recovery and able to stay consistent with my training.
As far as training ‘with’ power versus ‘by’ power, to me that comes down to what type of riding Im’ doing. Unfortunately with a family and a job that requires long hours, all my training during the week is done on the trainer. Training by power is very effective in those settings. Come Saturday’s however, power is nothing more than an interesting data point after the ride is over!
I tested them together last winter, PT wheel on the quark bike. I understand why the PT should be lower, but I did not observe a consistent difference. Your watts are your watts
“…all my training during the week is done on the trainer. Training by power is very effective in those settings…”
I think you’re missing the point of ‘with’ vs. ‘by’. I’m also a big fan of structured training, especially for those more limited on time. When I lived in the mountain states with long winters I’d do up to seven straight months of indoor riding when the snow was deep and an awful lot of that was 2x20 or 3x20 or 5x5 style interval work. But there is still no need to use the power meter as an effort regulator, you glance at it occasionally to make sure you’re not going way over or way under target but mostly pace by RPE and everything is self adjusting in the sense that your power numbers naturally move upwards as your fitness improves and you naturally self regulate on days when you’d bury yourself to try to hit numeric targets when carrying extra fatigue.
Structured training does not require strict adherence to prescribed power zones, it does require making the effort, sustaining efforts for durations appropriate to the targeted energy delivery system and making solid efforts and not simply getting complacent or lazy but it does not require being a slave to the meter or having a precise FTP determined.
I’ve used these methods myself and with all of the athletes I work with and it’s very sustainable and as long as they do the work, make the effort and give it their best effort for the day and stick with the program, the results come. Yes, I try hard to estimate both their and my own FTP by tracking regular and repeatable training power and with occasional dedicated testing (but typically only a few times per year at most even though FTP advances more quickly than that for most folks if they haven’t trained for many years) but the testing is not necessary to begin or progress with solid quality training and being indoors or out, time crunched or not does not really alter that.
I totally agree that structured training is very useful for someone with limited training time and indoor trainers are very efficient in terms of time vs. results but that does not require a strict prescriptive approach to intensity targets that requires a high precision FTP estimate. The difference with ‘by power’ vs. ‘with power’ isn’t a matter of structure or effort, it’s a matter of rigid ‘prescriptive’ target levels vs. doing to training to the best of your ability for that given day and then analyzing the resulting power file for an objective ‘descriptive’ look at what was actually accomplished, what that means in terms of workout quality, pacing, progress, or lack thereof and feeding that information forward to guide and improve the quality of future workouts.
If training were a single day, one time deal then it might be important to perfectly nail some prescribed training target based on a rigorous test, but training occurs over weeks, months and years and it’s the patterns, trends, and habits that determine if a program is successful or not. From that perspective the key thing is to do the best work you can do but the power data is still extremely valuable as a form of objective measurement and feedback and that’s training ‘with’ power.
How much would this have affected my results? When I get the Quarq back on my TT rig should I go with the tested FTP or do I need to retest?
No telling. I would retest.
But…but…testing is hard! I prefer to get faster without hard work and pain…
I guess I don’t see what the point of retesting would be given how the data is used… Yes, it might produce a slightly more accurate result, but I don’t see how that level of precision is necessary. Using the Monod protocol to determine FTP is an estimate, not an actual value. Also, any test is going to have a margin of error due to how it was executed. And when it comes to utilizing power for pacing in a race, I’m pretty happy if I can stay within +/- 5 percent of my target power under the best conditions.
With all that in mind, couldn’t I just tack on an extra 2-3% to my PT result and consider it to a very close estimate…probably about as precise as retesting at least!
Because there is a chance, given state of zero offset and calibration of both devices that the difference is actually huge, not small.
How much would this have affected my results? When I get the Quarq back on my TT rig should I go with the tested FTP or do I need to retest?
No telling. I would retest.
But…but…testing is hard! I prefer to get faster without hard work and pain…
I guess I don’t see what the point of retesting would be given how the data is used… Yes, it might produce a slightly more accurate result, but I don’t see how that level of precision is necessary. Using the Monod protocol to determine FTP is an estimate, not an actual value. Also, any test is going to have a margin of error due to how it was executed. And when it comes to utilizing power for pacing in a race, I’m pretty happy if I can stay within +/- 5 percent of my target power under the best conditions.
With all that in mind, couldn’t I just tack on an extra 2-3% to my PT result and consider it to a very close estimate…probably about as precise as retesting at least!
The powertap doesn’t measure cadence. It estimates cadence based off of the torque surges of each. The Quarq measures cadence directly. If you look at the raw data, the powertap cadence estimate jumps around quite a bit. As the data points out, over the long haul it gets pretty close.
I think you would probably get a good idea if they are much different real quickly anyway. Doing an interval set that suddenly is way too easy or hard, etc.
Because there is a chance, given state of zero offset and calibration of both devices that the difference is actually huge, not small.
I think you would probably get a good idea if they are much different real quickly anyway. Doing an interval set that suddenly is way too easy or hard, etc.
Because there is a chance, given state of zero offset and calibration of both devices that the difference is actually huge, not small.
Got it. Thanks for the input.
I’ve been riding with power for ~4 years, enough that I have a decent feel if it’s reading too far off of my RPE. It seemed to be pretty dead on.
Let me add another minor twist that I haven’t seen mentioned yet. Running with my quarq and my powertap, I measured about a 2% difference in reported power from the quarq between the two chainrings. This is the expected behavior. So in additional to all the other sources of error (drive train vs. not, different calibration, etc) there is another roughly 2% error depending on whether you stayed primarily in one chain ring or the other.
Let me add another minor twist that I haven’t seen mentioned yet. Running with my quarq and my powertap, I measured about a 2% difference in reported power from the quarq between the two chainrings. This is the expected behavior. So in additional to all the other sources of error (drive train vs. not, different calibration, etc) there is another roughly 2% error depending on whether you stayed primarily in one chain ring or the other.
+1 - depending on which crank you use, the difference can be up to 5% between the 2 chainrings.
