Power Meter Woes
#1
Senior Member
Thread Starter
Join Date: Jul 2012
Location: California
Posts: 128
Bikes: Holland Exogrid Jet
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Likes: 0
Liked 21 Times
in
9 Posts
Power Meter Woes
I have a Stages power meter which, since the beginning of the year, I've been using in conjunction with TrainerRoad's virtual training application for Windows. Another TrainerRoad user who had occasion to review my ride file thought that the power data it contained didn't quite add up. Initially, I didn't think much of it. After taking a second look, however, I began to suspect that he might have a point.
TrainerRoad's virtual training application includes a feature called VirtualPower. This feature uses speed data along with the power curve provided by the stationary trainer's manufacturer to calculate power. It is a feature designed to accommodate those using the application without an actual power meter. On those occasions when I used that feature, I noticed that TrainerRoad's VirtualPower calculator was reporting higher numbers for the same perceived effort. So, I conducted a little test. I paired the Stages power meter with my Garmin Edge 1000 cycle computer, and enabled the VirtualPower facility in the TrainerRoad application the objective being to record power data from both sources simultaneously. During the indoor workouts that followed, I noticed that there was a significant discrepancy between the power reported from each source. Specifically, the power output captured from the stages power meter on the Garmin head unit was 40 to 70 watts lower than that reported by the TrainerRoad application! Additionally, I could not detect a pattern. There was no evidence that the discrepancy changed in a predictable way for example, increasing or decreasing with the level of effort. A few watts I could ignore, but a difference of 40 to 70 watts is significant! I suspect that the power meter is defective. I was just wondering whether anyone else has had a similar experience with their Stages power meter.
TrainerRoad's virtual training application includes a feature called VirtualPower. This feature uses speed data along with the power curve provided by the stationary trainer's manufacturer to calculate power. It is a feature designed to accommodate those using the application without an actual power meter. On those occasions when I used that feature, I noticed that TrainerRoad's VirtualPower calculator was reporting higher numbers for the same perceived effort. So, I conducted a little test. I paired the Stages power meter with my Garmin Edge 1000 cycle computer, and enabled the VirtualPower facility in the TrainerRoad application the objective being to record power data from both sources simultaneously. During the indoor workouts that followed, I noticed that there was a significant discrepancy between the power reported from each source. Specifically, the power output captured from the stages power meter on the Garmin head unit was 40 to 70 watts lower than that reported by the TrainerRoad application! Additionally, I could not detect a pattern. There was no evidence that the discrepancy changed in a predictable way for example, increasing or decreasing with the level of effort. A few watts I could ignore, but a difference of 40 to 70 watts is significant! I suspect that the power meter is defective. I was just wondering whether anyone else has had a similar experience with their Stages power meter.
#2
Senior Member
Join Date: Oct 2013
Posts: 58
Bikes: 2015 Wilier Cento 1 Air, 2014 Trek Remedy 8 27.5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Likes: 0
Liked 0 Times
in
0 Posts
Been using a Kickr paired with Zwift recently due to an injury. I also have the stages and have noticed the exact same issue. A little research points to Stages in some case (not predictable) being noticeably low. Two issues. Stages does left side only so there could be a large discrepancy in leg bias? Second it appears most people report the low reading correcting with the new stages version. Not sure which you are running but I am running the original version meter with updated firmware. All told I am considering switching power meters in the near future.
#3
Full Member
I don't know about TrainerRoad, but in Zwift "virtual power" is consistently unrealistically high. I would trust your power meter over "virtual power".
#4
Senior Member
Join Date: Oct 2013
Posts: 58
Bikes: 2015 Wilier Cento 1 Air, 2014 Trek Remedy 8 27.5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Likes: 0
Liked 0 Times
in
0 Posts
FWIW Zwift at least on my part is reading the Kickr power so not virtual. Also, readings are considerably different using Garmin, Wahoo app, etc...
Stages has been reported as reading low vs quite a few other meters in many instances so not sure what causes it...
Stages has been reported as reading low vs quite a few other meters in many instances so not sure what causes it...
#5
Senior Member
Join Date: Oct 2013
Posts: 262
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Likes: 0
Liked 0 Times
in
0 Posts
Frankly, I wouldn't trust anything that's calculated (Trainer Road). I'd stick with what an actual power meter is reading. Even then, you'll see differences between products.
#6
don't try this at home.
Join Date: Jan 2006
Location: N. KY
Posts: 5,780
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 908 Post(s)
Liked 370 Times
in
269 Posts
You don't say what percentage of the Stages power that "40 to 70 watts" represents.
What were the wattage numbers?
What trainer model is this?
Fluid Trainers
Fluid trainers are supposed to have a predictable power curve, based on wheel speed. So, Trainer Road uses that curve to calculate power.
So if the standard power curve isn't exactly correct for your trainer, the calculated watts could be off. I've heard that fluid trainers have different resistance once they are warmed up.
Tire inflation can affect the power due to tire flexing losses, but then I'd expect the Stages to be higher.
Kurt Kinetic power calc
See their page discussing their power calculations.
From that page:
What were the wattage numbers?
What trainer model is this?
Fluid Trainers
Fluid trainers are supposed to have a predictable power curve, based on wheel speed. So, Trainer Road uses that curve to calculate power.
So if the standard power curve isn't exactly correct for your trainer, the calculated watts could be off. I've heard that fluid trainers have different resistance once they are warmed up.
Tire inflation can affect the power due to tire flexing losses, but then I'd expect the Stages to be higher.
Kurt Kinetic power calc
See their page discussing their power calculations.
From that page:
Limitations
While training with speed-based power is affordable and accurate and works amazingly well for longer intervals and overall ride averages, it's not strain-gauge based and does not measure torque, so there are a few things we cannot do. Unlike much more expensive systems, the Kinetic inRide will not register a spike in wattage from a sudden acceleration. The wheel must turn faster before a higher wattage reading is observed, so readings are not instantaneous.
While training with speed-based power is affordable and accurate and works amazingly well for longer intervals and overall ride averages, it's not strain-gauge based and does not measure torque, so there are a few things we cannot do. Unlike much more expensive systems, the Kinetic inRide will not register a spike in wattage from a sudden acceleration. The wheel must turn faster before a higher wattage reading is observed, so readings are not instantaneous.
#8
pan y agua
Join Date: Aug 2005
Location: Jacksonville
Posts: 31,250
Bikes: Willier Zero 7; Merlin Extralight; Calfee Dragonfly tandem, Calfee Adventure tandem; Cervelo P2; Motebecane Ti Fly 29er; Motebecanne Phantom Cross; Schwinn Paramount Track bike
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 1415 Post(s)
Liked 658 Times
in
341 Posts
Have you calibrated the power meter?
I'd trust a properly calibrated power meter over an extrapolation .
Heck, that's why you buy a strain gauge based pm
I'd trust a properly calibrated power meter over an extrapolation .
Heck, that's why you buy a strain gauge based pm
#9
Newbie
Join Date: Apr 2016
Posts: 4
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Likes: 0
Liked 0 Times
in
0 Posts
I have a Stages power meter which, since the beginning of the year, I've been using in conjunction with TrainerRoad's virtual training application for Windows. Another TrainerRoad user who had occasion to review my ride file thought that the power data it contained didn't quite add up. Initially, I didn't think much of it. After taking a second look, however, I began to suspect that he might have a point.
TrainerRoad's virtual training application includes a feature called VirtualPower. This feature uses speed data along with the power curve provided by the stationary trainer's manufacturer to calculate power. It is a feature designed to accommodate those using the application without an actual power meter. On those occasions when I used that feature, I noticed that TrainerRoad's VirtualPower calculator was reporting higher numbers for the same perceived effort. So, I conducted a little test. I paired the Stages power meter with my Garmin Edge 1000 cycle computer, and enabled the VirtualPower facility in the TrainerRoad application the objective being to record power data from both sources simultaneously. During the indoor workouts that followed, I noticed that there was a significant discrepancy between the power reported from each source. Specifically, the power output captured from the stages power meter on the Garmin head unit was 40 to 70 watts lower than that reported by the TrainerRoad application! Additionally, I could not detect a pattern. There was no evidence that the discrepancy changed in a predictable way for example, increasing or decreasing with the level of effort. A few watts I could ignore, but a difference of 40 to 70 watts is significant! I suspect that the power meter is defective. I was just wondering whether anyone else has had a similar experience with their Stages power meter.
TrainerRoad's virtual training application includes a feature called VirtualPower. This feature uses speed data along with the power curve provided by the stationary trainer's manufacturer to calculate power. It is a feature designed to accommodate those using the application without an actual power meter. On those occasions when I used that feature, I noticed that TrainerRoad's VirtualPower calculator was reporting higher numbers for the same perceived effort. So, I conducted a little test. I paired the Stages power meter with my Garmin Edge 1000 cycle computer, and enabled the VirtualPower facility in the TrainerRoad application the objective being to record power data from both sources simultaneously. During the indoor workouts that followed, I noticed that there was a significant discrepancy between the power reported from each source. Specifically, the power output captured from the stages power meter on the Garmin head unit was 40 to 70 watts lower than that reported by the TrainerRoad application! Additionally, I could not detect a pattern. There was no evidence that the discrepancy changed in a predictable way for example, increasing or decreasing with the level of effort. A few watts I could ignore, but a difference of 40 to 70 watts is significant! I suspect that the power meter is defective. I was just wondering whether anyone else has had a similar experience with their Stages power meter.
Simply put, power meters are pricey. Not everyone can afford one, so we decided to create a feature that'd make sure everyone can still benefit from training with precise, comparable data. When we train with power, we're able to monitor our progress. It's the consistency of the data that's super important here, so that means switching between the power meter and VirtualPower for your training probably isn't the best idea. This is because your FTP is based off one or the other's readings. In addition to that, there quite a few variables that influence the accuracy of the VirtualPower data. The power-to-speed ratio (power curve) for each trainer has been calculated with certain tire pressures, different rolling resistances, different trainer wheel tensions, and in some cases different power meters to calculate those values (which we know can have discrepancies from device to device). So it should be kept in mind that the discrepancy between VirtualPower and real power could range from just a few, or in your case a little bit more.

So, even if your VirtualPower settings aren't 100% representative of your outdoor power, you can still reap the benefits of training with power as long as you do so consistently with the same trainer, bike/trainer settings and VirtualPower curve. Doing so means your results are comparable, and you'll be able to track your progress.
I hope this helps!
Ride hard cheers,
Community Manager at TrainerRoad Cycling's Most Effective Training Tool
#10
Senior Member
Not exactly apples-apples scientific comparison but my FTP with Stages this year is at least 20 watts lower than my what my TrainerRoad/Kinetic Road machine virtual power number was last year. More or less the same amount of structured trainer riding.
#11
Senior Member
Are you really trusting an estimated value over your power meter? You got the power meter so that you no longer have to do such things. Enjoy your power meter and forget about estimations.
Wait a second, 'virtual power' is 40 to 70 watts higher? No wonder why you rather believe that one is true.
Wait a second, 'virtual power' is 40 to 70 watts higher? No wonder why you rather believe that one is true.

#12
Senior Member
Join Date: May 2013
Posts: 367
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 57 Post(s)
Likes: 0
Liked 0 Times
in
0 Posts

#13
Senior Member
#14
Senior Member
Join Date: Oct 2009
Posts: 2,853
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 1067 Post(s)
Likes: 0
Liked 259 Times
in
153 Posts
So you are suggesting that for example he may be making 100w with his left leg and 150w with his right. Giving him 200w according to Stages and really 250w in reality. I don't think that much of an imbalance is very likely at all.
#15
Senior Member
Join Date: May 2013
Posts: 367
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 57 Post(s)
Likes: 0
Liked 0 Times
in
0 Posts
No I'm not suggesting that at all. I'm commenting on the absurdity of trying to draw conclusions based upon two devices that both estimate power.
#16
serious cyclist
Join Date: Apr 2009
Location: Austin
Posts: 20,666
Bikes: S1, R2, P2
Mentioned: 115 Post(s)
Tagged: 1 Thread(s)
Quoted: 9021 Post(s)
Liked 3,511 Times
in
1,907 Posts
For all the people talking about Virtual Power being inaccurate - I've tested it. It's pretty effing close on a good trainer (KKRM). I'd trust it much more than Stages, certainly, with all the things that can affect that.
#17
Senior Member
#18
serious cyclist
Join Date: Apr 2009
Location: Austin
Posts: 20,666
Bikes: S1, R2, P2
Mentioned: 115 Post(s)
Tagged: 1 Thread(s)
Quoted: 9021 Post(s)
Liked 3,511 Times
in
1,907 Posts
#19
Senior Member
I have not. All I am saying is that between trusting a Stages power meter 'with all the things that can affect that' and trusting virtual power, also with all the things that can affect that, I would trust the Stages power meter. If in your particular case the virtual power was closer to a power tap, which makes it better in your eyes well, that's good.
In any case, the main point was that when looking at large discrepancies such as the one found by the OP, I would look first at the virtual power as the one more likely to be wrong. The Stages is a device designed and sold exclusively to measure power. Virtual power is an estimation offered for free by a software company, based on data supplied by a third party. Sure, the Stages has the whole left/right issue, but then the virtual power has many things to consider too: variability between trainers of the same model, ambient temperature, tire used, tire pressure, weight of the rider, etc., just as Nick posted up there:
Again, I am not suggesting virtual power is useless or always inaccurate or anything like that, simply that I would assume it to be wrong before assuming the power meter is wrong when seeing 40+ watt difference. I am sure there will be cases where the virtual power will be more accurate (such as the one you described, if we assume the PowerTap reading is the 'real' power) but I bet than more often than not, the Stages would be closer to the 'real' value than the virtual power.
In any case, the main point was that when looking at large discrepancies such as the one found by the OP, I would look first at the virtual power as the one more likely to be wrong. The Stages is a device designed and sold exclusively to measure power. Virtual power is an estimation offered for free by a software company, based on data supplied by a third party. Sure, the Stages has the whole left/right issue, but then the virtual power has many things to consider too: variability between trainers of the same model, ambient temperature, tire used, tire pressure, weight of the rider, etc., just as Nick posted up there:
The power-to-speed ratio (power curve) for each trainer has been calculated with certain tire pressures, different rolling resistances, different trainer wheel tensions, and in some cases different power meters to calculate those values (which we know can have discrepancies from device to device). So it should be kept in mind that the discrepancy between VirtualPower and real power could range from just a few, or in your case a little bit more. 

#20
Senior Member
Thread Starter
Join Date: Jul 2012
Location: California
Posts: 128
Bikes: Holland Exogrid Jet
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 27 Post(s)
Likes: 0
Liked 21 Times
in
9 Posts
Perhaps it would help to clarify things if I explained my concern. Specifically, I'm concerned about the adverse effects of overtraining.
If my power meter is consistently under reporting my level of effort, it means that I'm consistently overshooting the target. I am far from an expert in these matters, but my guess is that the cumulative effect of constantly overreaching will interfere with my body's ability to adapt. So, for example, if I'm doing a workout consisting of several intervals at 130% of my 200 W FTP, but I'm actually doing 340 watts instead of 260, I don't think that's necessarily reason to rejoice. More likely than not, I won't finish the workout. In fact, looking back at the last six months, there are a number of the tough workouts that I did not finish. There could be any number of reasons for that, including the possibility that I'm just a wuss, but I'm just sayin'.
If my power meter is consistently under reporting my level of effort, it means that I'm consistently overshooting the target. I am far from an expert in these matters, but my guess is that the cumulative effect of constantly overreaching will interfere with my body's ability to adapt. So, for example, if I'm doing a workout consisting of several intervals at 130% of my 200 W FTP, but I'm actually doing 340 watts instead of 260, I don't think that's necessarily reason to rejoice. More likely than not, I won't finish the workout. In fact, looking back at the last six months, there are a number of the tough workouts that I did not finish. There could be any number of reasons for that, including the possibility that I'm just a wuss, but I'm just sayin'.
#21
Senior Member
Join Date: May 2013
Posts: 749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 26 Post(s)
Likes: 0
Liked 0 Times
in
0 Posts
Perhaps it would help to clarify things if I explained my concern. Specifically, I'm concerned about the adverse effects of overtraining.
If my power meter is consistently under reporting my level of effort, it means that I'm consistently overshooting the target. I am far from an expert in these matters, but my guess is that the cumulative effect of constantly overreaching will interfere with my body's ability to adapt. So, for example, if I'm doing a workout consisting of several intervals at 130% of my 200 W FTP, but I'm actually doing 340 watts instead of 260, I don't think that's necessarily reason to rejoice. More likely than not, I won't finish the workout. In fact, looking back at the last six months, there are a number of the tough workouts that I did not finish. There could be any number of reasons for that, including the possibility that I'm just a wuss, but I'm just sayin'.
If my power meter is consistently under reporting my level of effort, it means that I'm consistently overshooting the target. I am far from an expert in these matters, but my guess is that the cumulative effect of constantly overreaching will interfere with my body's ability to adapt. So, for example, if I'm doing a workout consisting of several intervals at 130% of my 200 W FTP, but I'm actually doing 340 watts instead of 260, I don't think that's necessarily reason to rejoice. More likely than not, I won't finish the workout. In fact, looking back at the last six months, there are a number of the tough workouts that I did not finish. There could be any number of reasons for that, including the possibility that I'm just a wuss, but I'm just sayin'.
BUT, this is why you should be keeping a diary and writing down notes of your workouts.
Based from the sentences I bolded I feel that you need to re-test. It's ok to fail 1-2 tough workouts but when you say a number that means you can't hit your targets. I'm guessing your FTP is set a bit too high.
#22
serious cyclist
Join Date: Apr 2009
Location: Austin
Posts: 20,666
Bikes: S1, R2, P2
Mentioned: 115 Post(s)
Tagged: 1 Thread(s)
Quoted: 9021 Post(s)
Liked 3,511 Times
in
1,907 Posts
I have not. All I am saying is that between trusting a Stages power meter 'with all the things that can affect that' and trusting virtual power, also with all the things that can affect that, I would trust the Stages power meter. If in your particular case the virtual power was closer to a power tap, which makes it better in your eyes well, that's good.
In any case, the main point was that when looking at large discrepancies such as the one found by the OP, I would look first at the virtual power as the one more likely to be wrong. The Stages is a device designed and sold exclusively to measure power. Virtual power is an estimation offered for free by a software company, based on data supplied by a third party. Sure, the Stages has the whole left/right issue, but then the virtual power has many things to consider too: variability between trainers of the same model, ambient temperature, tire used, tire pressure, weight of the rider, etc., just as Nick posted up there:
Again, I am not suggesting virtual power is useless or always inaccurate or anything like that, simply that I would assume it to be wrong before assuming the power meter is wrong when seeing 40+ watt difference. I am sure there will be cases where the virtual power will be more accurate (such as the one you described, if we assume the PowerTap reading is the 'real' power) but I bet than more often than not, the Stages would be closer to the 'real' value than the virtual power.
In any case, the main point was that when looking at large discrepancies such as the one found by the OP, I would look first at the virtual power as the one more likely to be wrong. The Stages is a device designed and sold exclusively to measure power. Virtual power is an estimation offered for free by a software company, based on data supplied by a third party. Sure, the Stages has the whole left/right issue, but then the virtual power has many things to consider too: variability between trainers of the same model, ambient temperature, tire used, tire pressure, weight of the rider, etc., just as Nick posted up there:
Again, I am not suggesting virtual power is useless or always inaccurate or anything like that, simply that I would assume it to be wrong before assuming the power meter is wrong when seeing 40+ watt difference. I am sure there will be cases where the virtual power will be more accurate (such as the one you described, if we assume the PowerTap reading is the 'real' power) but I bet than more often than not, the Stages would be closer to the 'real' value than the virtual power.
#23
Senior Member
That is your single experience. Both Dan and I had cases of virtual power (well, the Zwift equivalent, based on the same power curves) giving us extremely high power estimates, so 'slightly off and fairly consistent' might apply to your case, but not to ours. I'm well aware of the single sided power measurement disadvantages, I just don't see how, in a case where you have a 40-70 watt discrepancy between virtual power and a single sided power meter, you would look at the power meter as the more likely one to be wrong. It could end up being the case of course, I just think that more often than not it will be the other way around. That's it, anyone can troubleshoot as they please.
#24
Senior Member
Perhaps it would help to clarify things if I explained my concern. Specifically, I'm concerned about the adverse effects of overtraining.
If my power meter is consistently under reporting my level of effort, it means that I'm consistently overshooting the target. I am far from an expert in these matters, but my guess is that the cumulative effect of constantly overreaching will interfere with my body's ability to adapt. So, for example, if I'm doing a workout consisting of several intervals at 130% of my 200 W FTP, but I'm actually doing 340 watts instead of 260, I don't think that's necessarily reason to rejoice. More likely than not, I won't finish the workout. In fact, looking back at the last six months, there are a number of the tough workouts that I did not finish. There could be any number of reasons for that, including the possibility that I'm just a wuss, but I'm just sayin'.
If my power meter is consistently under reporting my level of effort, it means that I'm consistently overshooting the target. I am far from an expert in these matters, but my guess is that the cumulative effect of constantly overreaching will interfere with my body's ability to adapt. So, for example, if I'm doing a workout consisting of several intervals at 130% of my 200 W FTP, but I'm actually doing 340 watts instead of 260, I don't think that's necessarily reason to rejoice. More likely than not, I won't finish the workout. In fact, looking back at the last six months, there are a number of the tough workouts that I did not finish. There could be any number of reasons for that, including the possibility that I'm just a wuss, but I'm just sayin'.
My advise would be to use your power meter and forget about the virtual power. Do an FTP test with the Stages and base all your training on the numbers you get from the Stages, not from numbers you get from anywhere else.
#25
Perceptual Dullard
Join Date: Sep 2009
Posts: 2,275
Mentioned: 35 Post(s)
Tagged: 0 Thread(s)
Quoted: 810 Post(s)
Liked 958 Times
in
413 Posts
Specifically, I'm concerned about the adverse effects of overtraining.
Many (but not all) power meters allow you to check their accuracy fairly easily. Sadly, the Stages isn't one of them, so you don't know whether the problem is you, your trainer, or your Stages. There are ways to check the Stages but they're considerably more trouble -- or they require a calibration rig or another power meter that is known to be accurate.
Fans of the Stages often point to Team Sky and say if it's good enough for Sky it's good enough for them. Team Sky has enough other equipment to check their Stages. Most people don't. If you really want to know, you'll have to find someone with a known accurate power meter who is willing to help you determine whether it's the Stages or your trainer. If both are okay, then it's you: you have a huge left-right discrepancy.