achoo

Yeah, the new Rotor definitely looks different from the Power2Max units.

I'll wait for a DCRainMaker or similar review before forming ideas about new products. Things like that VeloNews article are little more than press releases.

Still, though, gotta like the choices from the competition.

DXchulo

There are lots of possibilities. Some people like the wheel option better (each location has its own pros and cons). Maybe he likes a certain crank that Stages isn't compatible with. Maybe he likes that PowerTaps have been around a lot longer and have a good track record.

Maybe he does trust PT's data more. But you can't really assume that from what he said.

furiousferret

It's not a data thing, Powertap has been around for years and has proven to be a reliable product with great support. Stages is still new to the game and hasn't built that reputation yet (although they seem to be on their way).

RChung

That's not true if you know what you're doing. If you do, you can measure the aero difference between two front brakes, or two helmets, or the rolling resistance difference between two tires (you can't measure rolling resistance differences in a wind tunnel, which is why I didn't just restrict my comment to aero drag testing). Which isn't to say that high precision field testing is easy to do. You need good technique and good measurements of total power. I've tried to do some estimation with a Stages and the data quality isn't nearly as good as for a Power Tap, because the power from the Stages isn't consistent enough across the range of speeds and powers that I need to get good estimates. Other power meters also show the kind of consistency needed for precise estimation: for example, I've also seen good results with Quarqs and SRMs. The first generation P2M was terrible for drag estimation but the current generation seems much better. For that matter, the current model of the Stages is much better than the first generation. The differences in data quality between the early P2M and current P2M and the early Stages and the current Stages are pretty clear -- so is the difference in data quality between the current Power Tap and the current Stages.

[Edit] Which is not to say that I don't think there's a place for the Stages. I think it can be a good gateway drug.

simonaway427

Just curious....so any pro teams use Powertap?

RChung

Do you mean at the WorldTour level? I don't think so. Lotto used to but not anymore. Power Tap is a partial sponsor for Trek Factory Racing but only for trainers, not for wheels. I think it's probably because if you get a flat in a WorldTour race they just swap wheels. This is particularly so for neutral wheel support but even if you were swapping from the team car you'd have to pair the hub with the head unit and no one has time for that.

reef58

Above beat me to it

Don't know, but since wheel changes during the race are common I would suspect it is not the best method. Sometimes you will need neutral service. There goes your powermeter

woodcraft

If, as has been suggested, left/right power varies day to day or more often,

Then you could register your 45% left leg reading on Tuesday, and your 55% left leg reading on Thursday,

Suggesting a 10% increase that would not have been measured at the rear wheel.

achoo

But the only way you can know that is if you're already measuring it. If you're not measuring it, you're not going to know one way or the other.

And do you really think your left power is going to be 45% one day and 55% the next? Assuming you're going anywhere near threshold for any length of time, that would be like your left leg's FTP varying by almost 20% day-to-day. Your entire body's FTP doesn't vary anywhere near that much on a day-to-day basis, so why would one half of your body's FTP vary that much?

If those variations did happen with one-leg power, those using a Stages power meter would be reporting 20% swings in FTP on a day-to-day basis. I've seen no evidence that's happening. And yes, the numbers you used are a 20% variation in one leg power. You could even call it 22% as 55/45 = 1.22222...

Again, there's no evidence supporting assertions of day-to-day fluctuations anywhere near that magnitude.

You will make the same training decisions based on power data from either a PowerTap or a Stages unit. In neither case will you know your left/right power balance.

Sure, it seems like total power should be better than just using left power doubled. But then when you start digging into what would actually make it better, there's really nothing there. The "signal" of the recorded power data is no more consistently correlated with your actual power output for either a Stages unit or a PowerTap.

And all that matters for training purposes is the consistency of that correlation. Every time you generate X watts your power meter should record "300W", and it really doesn't matter what "X" actually is.

Yeah, the actual engineering and quality of the units can and will affect that correlation.

RChung

You keep saying this. This isn't true. That's like saying a differential GPS receiver is no more consistent in measuring the height of a hill than the GPS receiver in a phone, or that a laboratory scale is no more consistent in measuring mass than a bathroom scale, or that a laser is no more consistent in measuring distance than a wooden yardstick. It's not hard to show that the differential GPS receiver is both more accurate and more precise than the GPS in phones, it's not hard to show that a lab scale is more accurate and more precise than a bathroom scale, and it's not hard to show that a laser ruler is more accurate and more precise than a wooden yardstick. It's not hard to show that you can get better measurements of drag from a Power Tap than from a Stages. Whether that's important is a different question and perhaps one worth addressing -- whether the PT is more accurate and more consistent really isn't.

[Edit] Which is not to say that I don't think there's a place for the Stages. I think it can be a good gateway drug.

gregf83 

Some interesting data from a recent study on pedaling asymmetry. What the authors found was that crank based systems (i.e. SRM) were not able to measure pedaling asymmetries effectively compared to pedal based systems. This is relevant as some of the conclusions from previous studies (some of which used crank based measurements) has been equivocal. it's also interesting to note the significant change in pedaling asymmetry as power increases.

TexMac

Have you used one?

achoo

So if it's so easy to show how a differential GPS is better than the GPS receiver in a phone, it's got to be just as easy for you to tell us EXACTLY, with no knowledge of left/right power balance, how measuring total power is more accurate than measuring just left power and doubling it.

"Because it is!" isn't an answer.

This has nothing to do with lab scales vs. bathroom scales. What the hell does the accuracy of any scale that measures mass have to do with this discussion? And GPS units? You phoning this in from orbit? The fact you have to resort to irrelevant analogies instead of actually supporting your claim is somewhat, ummm, telling.

This is about whether or not using a fraction of an unknown input signal to produce a measurement is less consistent than using the entire unknown signal.

Guess what? If that fraction is constant enough that any variation in it isn't significantly greater than the other sources of errors in the resulting measurement - and there are a quite a bit - it's not any less consistent.

And a consistent number is all you need for training purposes.

And how do you KNOW the drag result you got from a PowerTap is better than one you get from a Stages unit UNLESS YOU MEASURE IT AGAIN WITH SOMETHING ELSE? Here's a hint: you DON'T.

achoo

I'm curious about this test - in theory, an SRM with its high sample rate should be able to pick up power asymmetries. And topping out at 350W really isn't a high power level - but it looks like only 6 out of 10 cyclists in the test could hit 350W. How good are the cyclists in the test? How long did each test run?

So the test seems to be blowing cyclists up? It doesn't really matter what happens to your power when you blow up, cuz you get dropped.

And I also note that this test doesn't claim the dominant pedal changes from side-to-side for any cyclist.

gregf83 

The problem with the SRM is that it can't separate the torque from the left and right pedals regardless of its sampling rate. The SRM can't tell if you're pushing down with your right or pulling up with your left leg. This is illustrated in fig 2 above.

The data was from a step test 25w/min increases.

What the data shows is that for this group of cyclists their asymmetry changed significantly with power and the asymmetry can be quite high. At 300W the asymmetry index(AI) was 36%, equivalent to a 41/59 split. The AI wasn't consistent with changes in power. Other studies have also found changes in AI with fatigue during the course of a time trial. That could make pacing based on power difficult.

thill454

You might want to read up on the Chung method of Cd determination before you make a complete ass of yourself.

Measuring on one side only could cause you to favor one leg over the other. Working harder with just the side being measured would show double the real change in overall power, ie you see a change of X in power, but the metabolic cost is only 1/2X. For people that train to the numbers, this could cause them to build in a long term imbalance that favors the leg being measured.

RChung

Too late for that -- but it's occasionally kinda amusing watching guys work their way up to it. Every once in a while someone will quote something about power meters or CdA estimation or something like that in one of these discussions to "prove" that I'm wrong -- without realizing that I was the one who did the analysis they're quoting.

Some guys have such a high opinion of themselves that they think if they can't figure out a way to test the accuracy and precision of a device then no one else could possibly have figured it out either. achoo has gone one step beyond this: he thinks if he can't figure out a way to test the difference between any other power meter and the Stages then not only can no one else do it but this also means that there is no difference between any other power meter and the Stages.

Which is not to say that I don't think there's a place for the Stages. I think it can be a good gateway drug.

Weatherby

Analogy is a means to convey a message to a simpleton.

It is telling that you did not get that......umm.

canam73

Speaking of high opinions of themselves, I'd like to question you on the value of your drag testing. And if you would indulge me in this, please direct your answers towards my situation. Specifically, a 42 y.o. father of 2 who works full time, races Cat 3 and rarely finds 10 hours a week to train, much of which ends up being on a trainer. During the off season, I don't ride much at all. I essentially never feel as if I am maximizing my cycling abilities do to lack of training time and consistency.

While I realize plenty of lucky souls manage to dedicate more time and effort to this, I don't think I am far off the typical person you will find considering a Stages or other PM in the Road Cycling Forum.

Anyway, I do have a quarq and a PT. My question is this: knowing that whatever time I spend drag testing will directly reduce the time I spend training, how much time should I spend on it? And for somebody in a similar situation to me, how much should the ability to this sort of testing factor in to their PM buying decision?

And if it matters, my race bike is cheap, by far the cheapest of my 4 road bikes. It is a cheap AL frame with Sram Rival and chinese carbon tubular rims I laced to novatecs. I simulated the closest race set up I feel comfortable in for an hour crit. I am 5'10, ride a 54cm frame and have about 4" saddle to bar drop. Basically, I look like your average crit racing doufus. I do not have big plans to replace anything with more aero components.

RChung

Hmmm. You know, I worked up this approach more than a decade ago but I didn't really talk much about it until (um, checking) 2007 because I figured other people were doing it, and I just thought it was kind of cute parlor trick but not of much practical use. Anyway, I had a cold one week and couldn't get out to ride so I wrote it up. Then someone else tried it out, got really good results which I didn't believe so he sent me his data and I realized that the reason I was getting parlor trick level results was because I had poor experimental technique. Anyway, now we fast forward a few years and other people (not me) have monetized it and it's being used by WorldTour teams, national Olympic teams, manufacturers, and, of course, consultants and coaches who do testing in velodromes whose riders have set world records. I'm very happy for them but I have a real job and wouldn't have been able to spend the time to get this stuff to market so I'm pretty happy just to get invited to watch Olympic teams and pro riders test, and get introduced to Big Names and every once in a while someone thanks me by sending me a bottle of wine or a T-shirt they won from a race. Meanwhile, I do analyses for friends who publish reviews and analyses on their websites. Does that sound like someone with a high opinion of himself?

Although I don't need to make a living at this I do have friends who do so maybe you should direct your questions to them. I'm sure they'll be happy to take your money. If you're in California, try ERO Sports. If you're in Europe or Australia, you can try Aerocoach. There are others around the world but since you're on an English-speaking forum I'm presuming you're in an English-speaking country. My friends at Alphamantis in Montreal surely have a more complete list -- that's their job, not mine.

achoo

Said to the guy with an MSEE in signal processing....

You really don't know what you're posting about, do you?

FWIW, analogy is something used by someone who is desperate to show he's not in over his head because he doesn't really know what he's talking about.

You still haven't answered my question:

How do you KNOW the total power from a PowerTap is a more accurate and more consistent value than the value from a Stages unit unless you measure it some other way? Answer that question and demonstrate you know what you're talking about. Continue to fail to answer it and demonstrate you really don't know what you're talking about, despite your overinflated opinion about yourself.

Pulling irrelevant crap out of your ass about GPS and lab scales means nothing.

So you can read the intertubz, take a power number, and pull a Cda out of your ass. BFD. That doesn't display any understanding whatsoever about measuring several unknown values (torque, RPM, etc), processing those values, and producing a single value and calling it "power". The fact that you believe (I won't use the term "think"...) that posting about differential GPS and lab scales vs. bathroom scales demonstrates any knowledge about signal processing is ****ing pathetic.

canam73

I am directing my question to you because you are the person consistently coming to power meter threads on what isn't even a competition specific forum directing people away from a product that isn't good for the drag testing you developed. I don't see the folks at ERO Sports, WorldTour teams or the Olympics doing that.

For them, I absolutely can see why this could matter. But that was my point, they live in a slightly different world than I do. Which do you think the typical fella in this Road Forum is closer to?

Btw, the amount of name dropping you just did is the best indication I can see of your opinion of yourself. And again, in the 41?

achoo

You'd think though, with the high sampling rate coupled with measurement of the crank angle something would show up. Gotta wonder what's going on inside that SRM unit...

How long was each step? 30 second steps would be a whole lot different from 30 minute steps. The longer the steps, the more reliable I'd consider this.

If the steps in this study were "short" - I'll put off defining "short" for now because I don't really want to spend much more time on here right now - the higher power numbers would appear to be "failure" numbers from not-so-great cyclists. Getting highly asymmetrical power at 300W because it's a "high" power level isn't indicative of a decent level of competitive cyclist. That's, oh, about a cat 4/5 FTP for a 180-lb rider. In other words, the level of power this study is calling "high" is a level of power that good competitive cyclists can sustain for hours.

At those power levels, unless those power steps are each "long" - and by long I mean measurable with a sundial if not a calendar (100W?!?! I've averaged over 200W for a double metric century and I'm not all that fast...) - this test is showing us that recreational cyclists have one leg fail before the other.

And even then, it's not saying that left/right power balance varies on a day-to-day basis for non-failure levels of riding, or that someone who's left-dominant today is going to be right-dominant tomorrow.

I don't see that it's demonstrating any variance in the fraction of total power provided by the left leg for a given cyclist riding at aerobic power levels.

achoo

So you can take a number, process it, and produce another number.

In my work - computer programming and signal processing - we call that a "script kiddie".

achoo

Yeah, he almost seems like a shill of some sort.

And even then, drag Cda's are relative - even if the number from a PowerTap is presumed for some unsupported reason to be more accurate, any power meter that produces consistent results will consistently produce lower Cda numbers for what in reality are lower-drag positions/configurations.

Which is the point of drag testing, no? Find the lowest drag position/configuration.

He doesn't even appear able to understand THAT. "Script kiddie" indeed.