Several people have commented about their solutions to cheating. Most have more holes than a colander and a few have tended to be bogus in that they have no idea how a powermeter works.
However this idea kept coming up. I was initially dismissive of it because it didn’t work. You can’t just stick a scale under a trainer. The front wheel doesn’t float! It takes weight too and it depends on your position. But why not just add a second scale? Oh that is good. So I built a prototype concept around this as a showcase of my ADC board that I want to redesign and put up on tindie or something. It’s still a work in progress, but given the alternative is terrible, it should hopefully improve things.
It’s simple. Continuously weight the rider and report that back. Measure the trainer and the bike before hand, subtract that off the weight with the rider/bike/trainer and you have your rider weight!
Lets look at some possible suggestions on cheating with this and how it would thwart them!
Sandbagging
How about we hang a weight while you’re bike it just sitting there. Okay, so this is probably an OE play and not aftermarket. Given that it knows the trainer weight then you probably got max of 4kg for a 6 – 10kg road bike. The trainers cost 1000 dollars CAD min usually, and this would be “competition level” so you’re not putting your 500 dollar 15kg antique MTB on it.
Even if you did try and fake it it has to be perfectly synced taking it off and putting it on. One slip up on timing for a few seconds and you’re caught!
Replicate
Like I simulated the HRM and power. However this gets harder because there is some tells that you can extract. Like mentioned, cadence sync. A capture and rebroadcast of pm cadence will cause a 0.5 – 1 second delay. Noticeable. The rocking motion on the feet wouldn’t have that. Weight scale cadence sync delay = cheater!
Replicating all is more possible, but using the powermeter challenge code idea and poof! Third party validation with keys, and that’s solved.
Have my daughter weight in
She’ll have to ride the bike too, or you both do? Anyway. This fixed that.
Have my 3 time Olympic champion cyclist friend ride it
Power models! I mean, it’s how Zwift power should have caught me, but couldn’t. HRM is trash as an idea, but a power critical curve model is easy and deviation from it is simple. Also, since live weight and people don’t gain or lose fast you have historic weight, so more than a few kg movement between a day or two becomes less feasible.