Any Peloton bike users here?
Discussion
gangzoom said:
Zwift has 3 million subscribers!!! How did manage that, most people outside the cycling world hasn't heard of it.
By not throwing half a billion at s
t boom/bust growth marketing with hilariously high CPA. And of course aiming at the general public, who notoriously quit things because they're lazy and have no willpower, add in a pandemic and boom/bust was inevitable tbh. The key difference is Zwift have aimed mostly at people whose passion is cycling (and to an extent running too, but not here), those people are far less likely to give up/novelty wear off, of course in summer usage drops off, but to my mind it is a far better long term proposition albeit with a smaller universe of potential customers. They have just moved into the hardware side so will be interesting to watch this space.
My wife bought one before Christmas and has used it a lot - we've got an eight-month-old and it's great to be able to just jump on and do 20 minutes.
I've got a separate Zwift setup with a smart trainer and a Wahoo Headwind in the garage which I prefer but have quite enjoyed the Peloton rides I've done too. It's a very different experience but the app is clever and the leaderboard and achievements are better than Zwift for motivation.
Expensive, yes, worth it, probably yes!
I've got a separate Zwift setup with a smart trainer and a Wahoo Headwind in the garage which I prefer but have quite enjoyed the Peloton rides I've done too. It's a very different experience but the app is clever and the leaderboard and achievements are better than Zwift for motivation.
Expensive, yes, worth it, probably yes!
gangzoom said:
2 million subscribers paying $40/month, that's $80 million revenue per month to pay for online class delivery/production. Not sure what that isn't sustainable as a business?
It's the shops. Something like 120 retail stores across several countries, plus more concessions in big department stores. Just in London - currnetly they have a 2300sqft shop on King's Road in Chelsea, similar on Marylebone High Street, another in Canary Wharf, another in Spitalfields, then concessions in John Lewis on Oxford St and Harrods. Plus a big new studio/office/production space.
Need to sell a lot of bikes and subscriptions just to cover rent, fitting and staff costs for that lot.
dirtbiker said:
My wife bought one before Christmas and has used it a lot - we've got an eight-month-old and it's great to be able to just jump on and do 20 minutes.
I've got a separate Zwift setup with a smart trainer and a Wahoo Headwind in the garage which I prefer but have quite enjoyed the Peloton rides I've done too. It's a very different experience but the app is clever and the leaderboard and achievements are better than Zwift for motivation.
Expensive, yes, worth it, probably yes!
I've just gone in a slightly different direction to come down to 1 setup, as the missus has no intention of going back to the gym for spin classes. I've got a separate Zwift setup with a smart trainer and a Wahoo Headwind in the garage which I prefer but have quite enjoyed the Peloton rides I've done too. It's a very different experience but the app is clever and the leaderboard and achievements are better than Zwift for motivation.
Expensive, yes, worth it, probably yes!
I've sold the smart trainer, and we've bought the Stages SB20 bike. I can use that for my stuff and she is using it for online spin classes, so far, so good.
sjg said:
It's the shops. Something like 120 retail stores across several countries, plus more concessions in big department stores.
Just in London - currnetly they have a 2300sqft shop on King's Road in Chelsea, similar on Marylebone High Street, another in Canary Wharf, another in Spitalfields, then concessions in John Lewis on Oxford St and Harrods. Plus a big new studio/office/production space.
Need to sell a lot of bikes and subscriptions just to cover rent, fitting and staff costs for that lot.
In John Lews Kingstom Pentolon announced a new concession not long ago.Just in London - currnetly they have a 2300sqft shop on King's Road in Chelsea, similar on Marylebone High Street, another in Canary Wharf, another in Spitalfields, then concessions in John Lewis on Oxford St and Harrods. Plus a big new studio/office/production space.
Need to sell a lot of bikes and subscriptions just to cover rent, fitting and staff costs for that lot.
In reality when it opened it's just a bike and a treadmill sat on a black mat in the open store with one or two sales people.
So zero fitting costs, don't take up much space so less rent and commission based sales.
Tight asses who are promising tight asses
So it turns out the Peloton web portal lets you mass export all your sessions as a CVS file with various variables attached....
A few interesting points*
1: Even though nearly all my sessions are done on the bike, and every session is uploaded to Strava successfully there is 'missing' data in the CVS file, not a surprise but thats databases for you.
2: One thing I want to 'check' is the calibration of the bikes 'power' reading, so I plotted the average power vs average HR to get an understanding of correlation. Not surprisingly its wasn't a perfect correlation, but I will repeat it in another 12 months - any changes in the correlation coefficient 'may' indicate a change in calibration?? Though that doesn't take into account training adaptation between HR and power???
Need to think about the hypothesis there abit more, there is also time as a variable, so you could do a time based analysis, but again training adaptation isn't accounted for?? - Any suggestions welcome.
3: Excel really is awful at handling any large(ish) dataset, crashed twice on me just plotting the graph - Need to get this setup as a script in R or similar for future ease of analysis.
*4: I need to get out more (in the real world)
A few interesting points*
1: Even though nearly all my sessions are done on the bike, and every session is uploaded to Strava successfully there is 'missing' data in the CVS file, not a surprise but thats databases for you.
2: One thing I want to 'check' is the calibration of the bikes 'power' reading, so I plotted the average power vs average HR to get an understanding of correlation. Not surprisingly its wasn't a perfect correlation, but I will repeat it in another 12 months - any changes in the correlation coefficient 'may' indicate a change in calibration?? Though that doesn't take into account training adaptation between HR and power???
Need to think about the hypothesis there abit more, there is also time as a variable, so you could do a time based analysis, but again training adaptation isn't accounted for?? - Any suggestions welcome.
3: Excel really is awful at handling any large(ish) dataset, crashed twice on me just plotting the graph - Need to get this setup as a script in R or similar for future ease of analysis.
*4: I need to get out more (in the real world)
Edited by gangzoom on Wednesday 26th January 18:35
Well I cannot figure out why my wifes average HR is so poorly correlated to output.
I wondered if its because she did no exercise before so there is alot of adaptation going on. So I divided her data from the last 10 months into 4 blocks. The hypothesis would be correlation coefficient should improve as time goes on if the vartion is due to training effect.
However as the data shows although 2 months after starting Peloton there appears to be a massive jump in correlation that trend isn't sustained.......As ever with data more will help, but drawing any conclusions is never easy. Probably need to look at activity types and/or time but this will significantly lower sample size. Will have another 'play' with the data tomorrow, who knew Peloton would work the brain as much as muscles
.
I wondered if its because she did no exercise before so there is alot of adaptation going on. So I divided her data from the last 10 months into 4 blocks. The hypothesis would be correlation coefficient should improve as time goes on if the vartion is due to training effect.
However as the data shows although 2 months after starting Peloton there appears to be a massive jump in correlation that trend isn't sustained.......As ever with data more will help, but drawing any conclusions is never easy. Probably need to look at activity types and/or time but this will significantly lower sample size. Will have another 'play' with the data tomorrow, who knew Peloton would work the brain as much as muscles
.Chris Stott said:
Leaderboard is ranked on KJ.
There a measure called strive which is a calculation based on ave heart rate over time.
KJ is just work done which is based on watts they then will use a standard efficiency multiplier like Strava to get calories. The whole thing rides on watts being vaguely right. They’re not because it’s a toy. There a measure called strive which is a calculation based on ave heart rate over time.
Gangzoom - I’ll leave you with that ever popular phrase - garbage in. Garbage out. The only thing you have here if any use is yours and your wife’s heart rate.
Edited by okgo on Wednesday 26th January 23:07
okgo said:
Gangzoom - I’ll leave you with that ever popular phrase - garbage in. Garbage out. The only thing you have here if any use is yours and your wife’s heart rate.
You haven't worked much with real life data than Edited by okgo on Wednesday 26th January 23:07
.The correlation coefficient between out put and my hr is actually remarkable. R2 value of near 0.8 based on real world data is astonishing, so hence my interest in why my wifes data is so variable......more sub analysis needed on that, but sub analysis is never great once you start chopping the main data set.
We've got a few friends with Peloton as well, I've warned them already next time I see them I'll be asking them to login to the webprotal and downloading their CSV files....yes am that kind of fun guy at parties
.I’ve got ten years of power data. From a powermeter from a powermeter company, and many of those years with HR too. I would not waste a single second analysing data from a tool that has no way of producing reputable data. Piers Morgan tweeting 380w was just the icing of the cake for me.
okgo said:
Piers Morgan tweeting 380w was just the icing of the cake for me.
You are am sure 100% right on absolute accuracy, but am not at all interested in the absolute accuracy of the power reading which am sure a proper power meter is better. Am more interested in the accuracy over time of the numbers to help me judget my own efforts (essentially test-retest reliability). So its the consistency of the readings thats far more important for me. But looking at the data its the difference in correlation between me and my wife I simply cannot explain, and that has nothing to do with absolute power accuracy readings of the bike(or does it?).
These paper implies the correlation isn't linear....
https://pubmed.ncbi.nlm.nih.gov/10527323/
Where as this paper suggest in traimed athletes the correlation coefficient between VO2 max, HR, power is nearly one!!!....The implications of this is the fact you can be nearly 100% sure of an athletes VO2 max and power output based on purely HR data alone.
https://pubmed.ncbi.nlm.nih.gov/7960315/
.....If you have 10 years+ of power readings and HR, fancy sending them to me to run some basic correlations on? With date stamps ideally as I have a theory I want to check
.None of this is very serious, I simply like data, and when I see trends in data I cannot explain it makes me want to understand why
.If we follow your hypothesis that the Peloton bike readings have poor test to retest reliability, and especially at the lower end of power where my wifes figures are it might make sense of the data. That's actually quite an easy null hypothesis to check, but it'll take me a few weeks to do the low threshold hold rides to find out, or I could a sub analysis of my own data at the lower power range......I'll let you know, regardless of if you are interested or not
.The other two 'unknows' is the inter and intra test reliability of users, however you would persume HR is a physiological response that shouldn't be affected by human perception??? Accuracy of the HR monitor could play a role, but again any equipment errors would be obvious for results on all individuals not just one. I guess this why I 'love' real life data analysis, you really have to work hard to understand the meaning versus lab based experiments.
100% accurate data is never present in real world data, but understanding real world data in my opinion anyways is as important as lab testing, as biases for 'human factors' or another 'real world' factor have a massive impact on outcomes that you will simply miss on lab-based testing. Though am not 100% sure real life ultitly of understanding Hr and power correlation is, apart from my own curiosity
.Edited by gangzoom on Thursday 27th January 06:55
Just to jump in. We got a plus about 6 months ago, my wife and I have used it at least 3 times a week since. Although I'm having january off as I did the powerzone training at the end of the year and it turned it into a chore.
Yes it is expense but we think it's worth it. Although I had been back on a bike for a year before we got the peloton I did notice a difference, especially after the powerzone training.
Working hard to lose that fitness so I can start again in feb!
Yes it is expense but we think it's worth it. Although I had been back on a bike for a year before we got the peloton I did notice a difference, especially after the powerzone training.
Working hard to lose that fitness so I can start again in feb!
'Serious cyclist' dismisses Peloton as a toy 
We have 2 Pelotons, which give pretty much identical outputs. One of them is the + version, which has a power meter.
Having said that, I did one ride on my BIL's bike and was over 30% above my typical output on my own bike... and yes, there are some people like Morgan who will doctor their bikes to give higher outputs. The 1st generation bikes aren't infallible.
But I'm not sure why 'serious cyclists' get so worked up about them... a Peloton isn't about cycling, it's about health and fitness... the only similarity is the use of pedaling. The vast majority of Peloton users have no interest in 'real' cycling and certainly don't want an old push bike on rollers in their spare bedroom/home gym.
Surely anything that gets people exercising and improves their health is a good thing.
We have 2 Pelotons, which give pretty much identical outputs. One of them is the + version, which has a power meter.
Having said that, I did one ride on my BIL's bike and was over 30% above my typical output on my own bike... and yes, there are some people like Morgan who will doctor their bikes to give higher outputs. The 1st generation bikes aren't infallible.
But I'm not sure why 'serious cyclists' get so worked up about them... a Peloton isn't about cycling, it's about health and fitness... the only similarity is the use of pedaling. The vast majority of Peloton users have no interest in 'real' cycling and certainly don't want an old push bike on rollers in their spare bedroom/home gym.
Surely anything that gets people exercising and improves their health is a good thing.
The whole platform is a waste of time if the numbers aren't broadly correct, and repeatable, your calories burned, your supposed fitness increases/decreases, everything. Moving your legs is better than not, I agree, just seems peculiar to me to build an eco system against something that isn't uniform.
It's not that I particularly care, just don't want to see others buy into it expecting x when they're really getting why thanks to fanboyism.
It's not that I particularly care, just don't want to see others buy into it expecting x when they're really getting why thanks to fanboyism.
Gassing Station | Pedal Powered | Top of Page | What's New | My Stuff