Which numbers are correct?

[SCM]

Hi folks, new guy here, first post.

I am brand new to a power meter (powertap c1) but not new to cycling data as I have been training with HR for a while and always analyze what information that I have had. With the power meter, I upload the information to goldencheetah and plan to use that information. My data is uploaded to numerous sources automatically and seems to differ slightly between the group. My question is, how do I know which numbers are accurate?

A lot of images but I can show you what I mean.....

Goldencheetah

more...

strava...

stravistix...

GarminConnect...

More Garmin...

Wahoo Elemnt app...

So right off the bat..garmin looks to be innacurate, not sure why but it doesn't appear correct. One giveaway is TSS over 300. It is my understanding that typically TSS does not exceed more than 100/hour. This was a 2 hour ride. Also, NP at 354 is too high. Average power of 287 seems to coincide somewhat with the image directly above from the elemnt app. My winter trainer is a TACX Vortex and my FTP all winter is 288. That may not be entirely accurate but it was calibrated and acts as a baseline.

The elemnt app seems to be fairly close with wattage in the high 200s which puts my FTP in the high 200s or so. It states my average as 240W but that makes sense as that's for the entire ride and counts every time my watts were 0.

Stravistix puts my weighted average power at 277 which seems to be somewhat in line with what I'm seeing from the elemnt app as well.

Strava has my average power right on with the elemnt (that's where it gets its data uploaded directly from over wifi.) The weighted average seems to contradict the others though at 251W.

Lastly, the one that I plan to use exclusively is goldencheetah. Golden has my FTP estimated at 254W which is more along the lines of my weighted average through Strava as well as the elemnt app. This also lists CP and Xpower both around 250W.

Now, I apologize for the long post but this is consistent over 4 rides. Id like to start getting into this information and forming a game plan for training going forward. Do I throw them all out and put my faith in golden cheetah? One more piece of information is that I download the strava file (including power) and then upload that info to goldencheetah. Big first post and I appreciate your help.

 

[Power Meter City]

Congrats on the C1 - those are cool. Wow...lot's of data here. It's normal for there to be some variability as different software programs parse the data files differently... however you sure wouldn't expect such large differences. I download my data to TrainingPeaks and Strava, and they're usually pretty close. What head unit are you using and which software program best matches the data that is on the head unit at the end of the ride?

 

[dkrenik]

Are you sure that you have your FTP set the same for all the app's?

Looks like GC and Strava agree nicely. I'm not familiar with Garmin and "stravistix" and only use Wahoo for running.

I'm thinking that something is whacky with the Garmin download as the kJ's (and other stuff) are really out of whack with GC and Strava.

I've always found close agreement with TP, WKO4 (and WKO2, WKO3), and GC. You need to do some "root cause" work on the Garmin stuff. In the mean time I'd go with what GC is showing you.

 

[SCM]

Thanks a lot for the replies. The powertap is linked directly to a Wahoo Elemnt. It then uploads directly to Strava from the elemnt. From there, I download the file from strava and upload it to GC. Those three seem to be pretty close. Garmin is in its own universe so I'll take that with a grain of salt. You may be onto something when you say that it could be a personal setting in the garmin app.

 

[smaryka]

My understanding is that Strava's weighted and average power numbers won't include any stopped time, i.e. it takes away all the zeros from sitting at traffic lights or whatever. That's why they are often higher in rides with lots of stops vs GC. It will include the zeros from coasting or descending, just not any zeros from when the speed is 0.

Golden Cheetah takes all the data that it gets, so includes all the zero-watts time including stopped time (unless you physically stop the Garmin from recording). GC will only estimate your FTP from the data it has, so unless you've actually done some good max efforts it won't be very accurate. If you've only done 4 random rides, that data is pretty bad for estimating FTP. What's better in GC is the critical power tool where you enter your best watts for a few durations and it will calculate a power curve for you. Critical power and FTP are pretty similar, but are both quite influenced by your short term power and W' (watt balance). The higher your W' the lower your CP will be and vice versa. Again, use the CP tool and put in some real numbers to get a better estimate.

The Garmin Connect app data looks really off, is it including the zeros from non-pedalling (coasting) time? My guess is that it isn't.

TSS is completely based on FTP so any TSS number that's either using poor data (missing the zeros) or a wrong FTP will have TSS all wrong.

 

[SCM]

That makes a lot of sense about GC. I was aware that strava didn't count the zeros but never thought that GC didn't. I forget that it's not a function of the power meter or my GPS unit, it's a strava algorithm. When I download all that data to GC it takes ALL the data. I suppose I was a little surprised the numbers were low but also on my trainer I would be in a sterile environment and would pedal hard for a test or an hour (I did both for accuracy.) I realize that this was a real world ride and the 5th of the week and 3rd in two days so I'm likely not going to see FTP type numbers unless I output that type of effort.

Going forward: I'll use that info about the CP tool to start getting some value out of these numbers. More importantly I will start getting trends and paying attending to things like W' so I don't overdo it as well. Great forum, thanks. I'll be here a while as this is some high level knowledge required.

 

[SCM]

What's better in GC is the critical power tool where you enter your best watts for a few durations and it will calculate a power curve for you. Critical power and FTP are pretty similar, but are both quite influenced by your short term power and W' (watt balance). The higher your W' the lower your CP will be and vice versa. Again, use the CP tool and put in some real numbers to get a better estimate.

This was a good tool. I just plugged a few numbers into it from an older ride and was able to surmise some useful data that I hadn't seen before. Through this tool it puts my CP a little more closely in line with the numbers I was seeing this winter. Looks like GC will be a good tool to analyze what I have. Just need to not get too hung up on minor details and see the big picture trends for my training.

 

[Power Meter City]

Just need to not get too hung up on minor details and see the big picture trends for my training.

Wise words. Many people (myself included), get too hung up on a watt here or there.

 

[pass4sure guide]

thanks for sharing this type topic i get knowledege....thanks

 

[cyclenthusias44]

I love the speed (29.8mph). You rode for 2 hours 6 min and that's really awesome.

 

[dkrenik]

I love the speed (29.8mph). You rode for 2 hours 6 min and that's really awesome.

His average was 19.4 mph.

 

[cyclenthusias44]

His average was 19.4 mph.

I see it but I am talking about the maximum speed. His maximum speed was 29.8mph. See it.

 

[RapDaddyo]

This was a good tool. I just plugged a few numbers into it from an older ride and was able to surmise some useful data that I hadn't seen before. Through this tool it puts my CP a little more closely in line with the numbers I was seeing this winter. Looks like GC will be a good tool to analyze what I have. Just need to not get too hung up on minor details and see the big picture trends for my training.

The Critical Power calculator is a great tool and I use it all the time. It is, by the way, available from sources other than GC. For example: http://www.cyclingpowerlab.com/MonodCriticalPower.aspx

You do need to choose your max power/duration samples carefully when you use the CP Model. For example, the model will overestimate max power for long durations unless you have at least one data point >= 60mins.

 

[certificationsdesk]

I'm not familiar with Garmin and "stravistix" and only use Wahoo for running.

 

[Rich from StriveMax]

Hey there, this is Rich. I am new to the forum. I might have some insight. Two friends of mine and I decided to develop our own analysis tool since we weren't happy with any of the existing ones. Not enough functionality, too much (irrelevant) functionality, too expensive, crappy design, etc. So now we but ourselves out there to see if we can do it better. In the early development process we noticed the same thing. Different softwares, end up showing different values for the exact same data file. That has to do with how the data is processed. Everyone does it a little differently.

But anyways, we just launched StriveMax. It's completely free and free of ads. No obligations arise from using it. I am not trying to sell anything. I'd simply love to get some feedback and constructive criticism from knowledgeable and critical users like you all seem to be so we can keep improving our little project. Here is the link: www.strivemax.com

Thanks a bunch!
Rich

 

[Balajee2220]

thanks for sharing this type topic i get knowledege....thanks

 

[swampy1970]

Core temperature - where do you stick that sensor?