Does anyone else find that the average power they are able to sustain on Alpe du Zwift is significantly lower than their FTP?
I also use BigRingVR and have a fairly good idea of my real sustainable power over 50 mins or so, but it seems that the power I can sustain on Alpe du Zwift is about 10w lower than I would usually expect on a long, continuous climb of similar length.
My theory is that there are very frequent small gradient changes in the way that Zwift has modelled Alpe du Huez, and that combined with the generally pretty steep gradients this stops you getting into an efficient climbing rhythm. In my case I have trainer effect set to 85% - wondering if I might be able to sustain a better average power and thus get a faster time with a lower trainer difficulty, but this seems contrary to the spirit of the route!
are you able to calculate weighted power? may be closer to ftp.
BTW, how about an experiment with trainer at 0% difficulty so see if you can hit ftp?
Well, I just installed Golden Cheetah to check and xPower (their equivalent of NP and weighted average power), is within 1w of averege power for the climb…
I assume that if I did the ride with 0% trainer difficulty I’d be able to hit FTP (given that I can, or almost can, in other situations).
Just wondering if others have noticed the same thing really - the odd thing is that I usually find it easier to maintain FTP on a long climb (simulated or real) than on the flat, it’s just on Alpe du Zwift I have problems!
Every corner you need to shift chainrings or several gears. This costs time and power as you have noticed. The corners throw me too. 0% trainer difficulty, as has been mentioned, will remove this issue.
If you did the alp in 40 minutes I’d bet you’d do it with a higher power than your FTP just because you’d be able to go slightly harder for shorter. Have you tried that?
From what I understand is the Zwift alpe is a copy of the real thing.
“ The length, elevation climbed, distance between switchbacks, number of turns—all of it is near-identical to that iconic cycling course,” Eric Min, Zwift’s CEO and co-founder, told Bicycling .
If you did the alp in 40 minutes I’d bet you’d do it with a higher power than your FTP just because you’d be able to go slightly harder for shorter. Have you tried that?
You mean do some of it for 40 mins…? Can’t do all of it in 40 mins, I’m not that good… 
Takes me about 50 mins.
Either can I. I was only joking.
1 Like
It does bring up a wider issue - if you are racing or otherwise riding competitvely against someone else on Zwift, trainer difficulty does make a difference, even if you are going just as fast with the same power.
You can always tell on a climb when someone is doing it properly (with near maximum trainer resistance), because of the way they ride. Most people do short, hard efforts on steep sections and ease off a bit on the flatter bits. I’d love to see zwift make everyone’s trainer difficulty setting visible - that would then open up the possibility of meaningful and engaging racing on hills.
2 Likes
Then you probably would like to see what their trainer actually is (brand, type, model, etc…) as not all trainers are “equal = comparable”. I mean - first one is able to simulate incline up to 7% while other 10% and another 20%. As far as I can understand this difference it’s quite similar to “difficulty setting” … (correct me if I’m wrong)
1 Like
Yes, you’re right - if a trainer can’t simulate a 10% gradient for a 70 kg rider (or whatever), then the resistance won’t get any greater after a certain gradient whatever you set the trainer difficulty to.
But most controllable trainers these days can simulate decent gradients, even if all can’t do the steepest. A bigger problem I think is accuracy of simulation when it comes to gradient. I use an old Tacx Genius Smart (non direct drive), and I’m pretty sure it over-resists, which is why I set it at 85% rather than 100% (I use a PM2 power meter for power data, the Tacx is way off for power estimation!).
Increasingly we will see most people using one of two or three models of direct drive trainers with known accuracies for both power and resistance however. I’d upgrade if it was a requirement for certain events.
Then we also get into what bike gearing are you using? Climbing the Alpe using a 48/26 is very different from a 36/34.
2 Likes
Yes, but that’s a valid personal choice - a difference that could also exist in the “real” world. Unlike a hill being a hill for one person and a ride on the flat through treacle for another…
Another theory - heat. Much more likely for body temp to rise (and watts to fall) on longer indoor session than outdoors. Less sustained overheating and impact in shorter session when measuring FTP (e.g. ramp test).
No, not that, because I can sustain my FTP on BigRingVR for similar lengths of time - the difference being that the climbs are usually steadier (and perhaps just a little shorter, maybe 45mins). Also I can sustain 104% of my FTP for 20-25 mins on the Epic KOM on Zwift.
Doesn’t decreasing trainer difficulty have the same net effect as choosing easier gearing? I assume you have to do the same amount of work, but you’ll just go slower and have more pedaling to do to if you set trainer difficulty below 100%. Or is there more to it than that and by decreasing trainer difficulty you actually do less work?
I always ride at 100% because I want a simulation of how I would ride in the real world (or close to it anyway).
3 Likes
Doesn’t decreasing trainer difficulty have the same net effect as choosing easier gearing? I assume you have to do the same amount of work, but you’ll just go slower and have more pedaling to do to if you set trainer difficulty below 100%. Or is there more to it than that and by decreasing trainer difficulty you actually do less work?
No, not the same thing - decreasing trainer difficulty flattens out the gradients in effect (in terms of how they feel, if not how fast you go on them). You might be able to get the same feel with a different gear on a steep hill at 100%, but you would still experience the differences in resistance between slopes of different gradients. It’s the reacting to the gradient changes that has a lot to do with how hills feel. Also, you’d need a really big sprocket to make a 12% hill feel flat…
1 Like
Ahhh. That makes sense. Thanks!
Thinking about it more - I’m pretty sure that on most trainers a 10% gradient with a gear that allowed you to put out 250w at 90rpm would still feel different from a 0% gradient with a gear that allowed you to put out the same power and cadence. There will be less momentum in the pedal action, it will still feel jerkier (I’m guessing).
What I cannot seem to find anywhere are attempts with the front wheel raised 8% versus flat. Different muscle demand should affect watts. Without raising the front wheel you are essentially doing a flat TT into a head wind in terms of muscle recruitment. Just my opinion
Well, I assume that when you have a 0% gradient, you have the full advantage of the weight of your legs. The opposite would be the theoretical situation where you would ride on the ceiling and would have to combat the full weight of your legs when pressing…ummm…up. The correct answer is based on sin or cos and left as exercise for the reader to google 