Hacker News Reader

The trainers used by many people on Zwift (eg. the Wahoo Kickr) actually read out pretty accurate power numbers.

The classic Peloton is basically a mechanical device with sensors attached - and none of those sensors are a power meter. The power estimates are widely reported to be inaccurate (they exaggerate output to varying degrees). On the other hand, the Kickr precisely modulates resistance and is able to accurately estimate power. I believe the newer Peloton Bike+ has a better system that produces accurate power readings as well.

Of course, as long as the competitors are bringing their own hardware, there will be ways to fake the data. I’m just pointing out it’s possible to measure power accurately and hardware that does it is widely available.

Personally, I find power readings to be useful because they help me understand the effort of cycling in different conditions. I can compare the readings I get in my garage to those from the climb up a steep mountain road, or into a gust of wind. Measuring my training based on my power output has helped me to go have experiences that I couldn’t have before.

I think people are overblowing his investment in making it competitive

It's quite easy to cheat in Zwift as you can put any weight on your profile (and the metric for measuring speed is generally W/kg). As well as different trainers have different power measurement accuracy, etc

But to some degree Zwift and other 3rd party Zwift race-tracking platform find the cheaters based on outliers and other factors and remove them from the list

All he did is figure out a way to cheat which should be simple to detect by the platform itself (changing weight mid-ride -> ie make yourself lighter uphill and heavier downhill) and demonstrated it. And he got banned for it

The way I see it, it's the same as someone demonstrating an LLM prompt injection or a cheat in a video game. A geek hacking a thing they like and showcasing it

The most effective way of "cheating without cheating" is just using something like Sauce for Zwift or a head unit that displays your rolling 20min power, so you can effectively "sandbag" and keep your power under the threshold for the next category. This is arguably not cheating at all, just smart racing, but it is not a great racing experience when you're in the top third or so of a category and you race someone who very clearly is far superior and only staying in your cat to rack up podiums.

To be fair, though, 1) this makes a lot of sense for team racing series, and 2) it's a really crappy experience to be at the bottom of a category and get dropped within 2 minutes of the start of every race.

I'm 100% convinced the vast majority of people either outright cheating or gaming the system are Cat C riders trying to avoid getting upleveled to B.

For heavier riders this is almost the only way to enjoy racing on Zwift. Even with a high FTP, I will never compete in A. So I jump in B races, use them as sweet spot training and get some sprint in at the end.

> I was repairing my Peloton the other day and getting inside made it really obvious how else you can manipulate

That's because Peloton's hardware is ...to put it nicely... "somewhat economical". :P

Peloton's never been intended as a serious sports science tool though, so it really doesn't matter how inaccurate it is. Its experience is much closer to gym spinning classes than outdoor cycling.

A spinning bike is even cheaper in design, it'll just have a "make it harder or easier" screw to tighten a strap round a spinning drum. That's the level they're competing with.

Despite the name you're unlikely to see anyone in the professional peloton warming up for a Grand Tour on either a spinning bike or something from Peloton Inc's current range.

A high end smart trainer like the Tacx Neo family will use electromagnets to generate its resistance. 100 watts of resistance is 100 watts of electrical power to about 0.5%-1%. No calibration necessary because it's connected to a source of electrical power[1].

-

1. You can actually use a Neo without connecting it to the mains but it's less accurate and you don't get some of features like downhill simulation.

I started trying to think of reasons your times might be systematically different (other thin bike calibration, as you noted) and every one I could come up with -- altitude, CO2 level, noise level, how you get there -- could affect anyone at any time. If you're not racing next to me we're not competing to any degree of useful accuracy.

The Peloton doesn't have a power meter. They "calibrate" it at the factory with a curve based on resistance and cadence and ship it out.

Meanwhile a good trainer has a power meter reading actual torque in real time and giving a power accuracy around +/-1% or better. There's really no comparison.

For example, my FTP is 342 watts as tested on my Stages SB20. It's slightly higher outdoors. A buddy got a Peloton and asked me what my FTP was because he knew I rode a lot and was surprised that his was soooo good because the Peloton returned him an FTP of 280. I was suspicious so I asked if I could give it a go. My FTP on the Peloton was 560 watts. I'd best doped up peak Lance Armstrong.

It wasn't long ago all Power Meters cost more than a Peloton, some probably still do, but I haven't checked in a while.

They are still expensive, so obviously they're not going to put anything decent in a Peloton.

Yeah who cares lol literally go bike outside and breathe in the crisp autumn air

I wouldn’t last thirty minutes in Hong Kong Island’s traffic.

I’m sure many others are in similar situations.

The Strava heat map shows a lot of cyclists surviving in Hong Kong.

https://www.strava.com/heatmap#9.83/114.11252/22.26615/hot/r...

I know, I've seen them and I always admire their fortitude -- cars do not leave them much space when they over-take.

I said I wouldn't survive.