Accidental cheating

Racing
1

I race using Assiomo pedals as power source and Zwift Hub Trainer as resistance. I didn’t want to use the pedals, but it appears that the Hub under reports power above 300w once the trainer is warm by up to 8%.

There also appears to be a delay in the power I push into the pedals and the power recording in Zwift and moving my avatar. This delay appears to be around 3-4 seconds. This makes it very difficult to ride at a ‘constant’ power output. If there is a slight speed increase in the race and you push a little harder on the pedals to stay in position, the delay means nothing happens so you panic and push harder. Then when the power registers in Zwift you have ‘overcooked’ things, so you soft pedal a little to not fly off the front. It’s annoying (TTT is almost impossible) and you end up spending far too much time in Zone 6 and 7.

Since I was promoted to the highest racing category, I have found that I need to constantly accelerate to not get dropped. I’m also a former outdoor racer from the Netherlands and still have the mindset to constantly overtake riders to reach the front, then recover and repeat (crosswinds here mean staying near the front is essential).

I’m finding my race power profile is very spiky and the amount of accumulated time I spend above 500w is really intense.

But now I also have people accusing me of ‘sticky watts’ and yesterday I came 35/40 in a Zwift racing league and apparently have been flagged as Code 19, ‘unconventional racing technique’. Weird to finish almost 5 minutes down on the ‘improbably light’ winner and get some sort of cheating accusation.

It’s odd, because I push myself dangerously hard in races and have serious issue with anyone accusing me of cheating, especially as so many of my races seem to be won by riders who weigh exactly 50.0kg or who claim to be 65kg but have significant upper body muscle mass in their profile photos…

On Friday I actually threw up after managing to finish in the lead group in the Zeal Japan race with 349 AP. I wondered if I had actually done myself some serious harm for a few moments…

I am wondering if switching to ANT+ instead of Bluetooth might help this issue?

Also, I am using Zwift Cog 2 and virtual shifting - does anyone know if this is an issue with the Assiomo pedals?

Does Trainer Difficulty also play a role?

I’m busting a gut to finish races here, spending up to 28%(!!) of races in Zone 6, and getting accused of cheating and it’s really annoying me.

I totally get that some people take advantage of this, but I don’t, and I’m not even sure I know how to. I’m not constantly sprinting and then not pedalling, my cadence is fairly constant and while my power profiles are spiky, I don’t have the ‘flat tops’ that you see in the classic sticky watts profiles.

Anyone have any advice on how I can fix this?

1 Like
2

That might make it worse

This requires bluetooth

No

Are you calibrating the pedals? They should be a very good power source.

3

Try to stop using sticky watts would be a start.
Use the trainer in 10ghz race mode rather than the pedals and take the ego hit on the power.
Stop pedalling hard and then stopping as ultimately this is the cause of it.

Il be honest, WTRL set a high bar to code 19 people as they simply dont want the hassle and would rather keep their head in the sand along with Zwift, so if you are getting pinged by it, its pretty strong indicator you are using using Sticky watts and getting an unfair advantage.

2 Likes
4

I don’t stop pedalling.

Trainer under reporting power by 8% and me wanting accurate power isn’t anything to do with ego.

‘Unfair advantage’? The power I produce in races is totally aligned with the power I produce in my interval training sessions.

Thanks for the unpleasant feedback, by the way. Cheered my morning.

1 Like
5

Yes, totally calibrated.

1 Like
6

Can I have the pedals send power data to Zwift via ANT and the trainer and virtual shifting run via Bluetooth?

Apologies if this is a daft question, I am not tech savvy.

2 Likes
7

You can record the pedals to the head unit for personal preference, but switching to 10ghz and improving your pedalling technique will remove the issue in play.

Do you want to fix the issue or just moan?

When you are gaining the sticky part of sticky watts, i.e. when the power holds on that value, you are not producing that power through physical effort - Ive seen instances of users getting 3-4-5 seconds of sticky power, regardless of your view, it does provide an advantage at that time.

And further more, you have been DQ due to it, so yes it is cheating and getting an unfair advantage.

1 Like
8

I didn’t get DQd.

Ok, so the power stays high for a couple of seconds after I ease down. But what about the couple of seconds where I am pedalling hard and the power isn’t registering? Doesn’t this even out?

I certainly am not smashing it for 3 seconds, resting for 3 and doing that over and over again. I am trying to ride at a constant tempo. My cadence only ever drops to zero briefly in descents. My profiles don’t have the ‘flat tops’ of riders who are doing this on purpose.

1 Like
9

Hi James, I would like to help you understand, but I’ll be honest and say that the data is telling a bit of a different story about whether you stop pedalling. I’m not judging, I’m just want you to know what I, and others, can see in the data.

Sticky Watts happens with “event based” (means when you pedal) power meters - like the Favero Assioma pedals. When you stop pedalling Zwift doesn’t receive any power data and so it repeats the last read it got 2-more times (for a total of 3 identical power readings in 3-seconds).

I had a look at your Zwiftpower power graph, and when I zoom in what I can see is that there are lots of occaisions where your power is identical for 3-seconds before it drops down low - and these are the flat tops. This is physiologically impossible to do probably even once - and certainly impossible hundreds of times. To the best of my knowledge the only way to trigger it happenning is to cease pedalling (rather than just slow cadence) and hence send no power data to Zwift.

zpgraph
zpgraph3253×1181 138 KB

I know your race cadence graph doesn’t drop to zero very often, however I wonder if that is taking cadence from the trainer rather than the pedals?

If you did just slow cadence but kept some power going through the pedals in the low-power phased then Zwift should record a low power and low cadence … not the exact same repeated power.

The truth is whether you intended to gain an advantage or not, Zwift repeating those powers for 2-more seconds every time did give you a reasonable amount of extra power. It is perfectly possible to ride with Assiomas without getting Sticky Watts, but you do have to ensure you don’t drop from a high power to no power. EDIT - I’m afraid that bluetooth or Ant won’t make a difference to what Zwift see so its not a connection-fix.

If you still feel that you’re getting this Sticky Watts effect in Zwift without dropping the power/cadence abruptly, then the only way to prove that there’s a new technical issue would probably be to video yourself doing it and send into Zwift.

3 Likes
10

Thanks for your reply.

My cadence is recorded via the pedals.

I always felt that the ‘advantage’ of the power holding for a couple of seconds after you ease off was offset by the ‘disadvantage’ of smashing power for 2 seconds before the power registers in Zwift?

I actually feel as though this delay disadvantages me, rather than giving me an advantage. I am a ‘diesel’ type rider and I want to ride at a steady output and this latency is forcing me to fluctuate excessively and spend too much time in anaerobic zones that are really not my strength.

11

At the level you are racing, you are going to run into more of an expectation that you will use an accurate trainer as your primary power source. Assiomas would be nice to have as a backup source to confirm the numbers. Some other racing platforms would not even let you race with your current setup. On Zwift you can race, but when people are scrutinizing your performance, which they will do in events they care about like ZRL, the suspicions will probably keep coming up because of how the pedals work.

I’m curious whether the accuracy problems you’re observing when using the trainer as the power source are affected by the use of virtual shifting. If you start doing some dual recording and try riding with a cassette on the trainer and virtual shifting turned off, that might help you understand whether the disparity between the pedals and the trainer is in any way affected by enabling virtual shifting. It may not make a difference. It’s possible the trainer is just too cheap for a serious athlete.

3 Likes
12

There should be no delay when putting the power down - certainly not 2-seconds. I have Assimoas as well, and their power registration is as instant as a normal turbo trainer. Maybe check out your log files through Zwiftalizer and see if there are any connection issues being thrown out. Feel free to post here and @ me if you want another look at those.

1 Like
13

Yeah, I ride the Zwift Ride, so a cassette is going to be difficult.

I do have an unused Tacx Flux 2 I could set up with a cassette and use a bike, but far from ideal in terms of space as my missus uses the Zwift ride.

It’s really annoying that using expensive and accurate power meter pedals is problematic, but trainers with lower accuracy are apparently fine.

I don’t know if it is related, but in my training, ERG mode is impossible to use as power fluctuates all over the place and I spend the whole interval trying to find the correct level. Is this related to the pedals or virtual shifting in some way?

1 Like
14

If I do a free ride in Zwift and start pedalling, the wattage doesn’t register for a couple of seconds and my avatar doesn’t move for a few seconds.

Is this normal?

15

For the very first movement, probably yes. If it happens again like you go from coasting to pedalling - then that’s not usual.

16

It depends how much oscillation around the target power number we are talking (a little is expected) - but one thing that can cause this is a device that is connected to multiple sources trying to control it.
Ensure there is nothing trying to control the trainer when you are in ERG mode.

10Ghz Race mode on the trainer should remove the sticky part of the issue as the trainer reports 10 times a second to zwift over the 1 time a second the pedals do - If zwift misses the second it holds the power, so the trainer has then sent 20 power measurements to zwift over the 1 missed and next 1 from the PM and so on.
(you can see this in the fit file if you were to open it up)

Edit to add - I say this as Assioma Duo user, they are good power meters, but they are the check & balance indoor, not the source.They are my outdoor PM

17

In ERG mode virtual shifting should not affect anything, but you could test that by not pairing the Ride controls and doing an ERG mode workout.

The suggestion to run logs through zwiftalizer.com is also a good one if you want to look for any problems with connectivity to the paired devices. It’s free and won’t take much time. You can click the Share link on the results and post the link here if you want feedback on that (you may need to break the URL by sticking some spaces in it in order to post it).

In your situation I would probably sell the two trainers to raise money for something better to put the Ride on. It’s a good time of year to get rid of old trainers and many people would be happy to ride on the Hub and Flux.

If the Hub is still under warranty you could also do some dual recording with the purpose of trying a warranty claim over its inaccuracy.

18

Is it possible for me to consciously ride ‘smoother’ in a way that doesn’t trigger the sticky watts issue?

It isn’t easy, as there are still times in a race where you do soft pedal or even stop and then the latency is going to have an influence, but you could certainly try to avoid a spiky profile, right?

19

Without seeing cadence its really hard to tell, throw up the strava activity or the fit file.

There is a difference between soft pedalling which everyone does, and having sticky watts kick in. Its more about sudden changes to input.

1 Like
20

It’s weird, because to be honest this feels more like racing IRL, rather than the ‘everyone riding a TT’ feel of some Zwift races.