HalfSpeed wrote:
I'm not on Zwift, but a female friend is. She's been riding a mountain bike for a few years, but recently got a pretty sweet Cervelo which she keeps on a dumb trainer. She puts up both real and Zwift rides on Strava...
Typical real ride on mountain bike, on streets - 11.59 miles in 1:18:56; 1361 feet climbing; 8.8mph average, 27.3 max
Recent Zwift ride - Giro di Castelli Stage 4 - B Group: 16.68 miles in 44:25; 1099 feet climbing; 22.5mph, 53 max.
mmhmm... right!
This is a pretty ridiculous and pointless comparison.
You're comparing speeds from 2 potentially utterly different real and simulated rides and apparently trying to use that as a basis for questioning the credibility of the simulation.
Why do I think this is ridiculous?
Let me give you a few obvious reasons.
- Well for starters, you don't tell us if she was putting in a similar effort. There's a limit to how fast I can go, but I can go as slow as I like! Was she wearing a HR strap or using a power meter to quantify the effort for each?
- Were these routes both closed loops? It makes quite the difference if they started and finished at different altitudes or if the outdoor ride was predominantly into the wind.
- What dumb trainer was she using and is it set up correctly. Does she have a power meter or is she using power curves for "virtual power"? If you select the wrong power curves for your trainer on Zwift, you can get VERY inaccurate data. That's not Zwift's fault. If it's a magnetic type trainer, did she use the correct resistance setting?
- Did she ride continuously outdoors or were there junctions and delays included in the time and average speed given?
- She was on a mountain bike for the road ride and a road bike on the trainer and in the sim. You don't think there might be differences here? What sort tyres on the MTN bike? Smooth or knobbly, and race or durable types? Position can make a very significant difference to speed. Zwift assumes a good road or tri bike position. How's her mountain bike position?
- Her max speed outdoors was only 27.3mph. With the figures you gave she must at some point have been descending decent gradients unless it wasn't a closed loop route. Is she a very nervous descender? That's a very low speed. I doubt she was pedaling and she was probably braking during some or all descents. You can't brake in Zwift.
- Was it windy out?
- The outdoor ride has a very significant amount of extra climbing. 1361ft climbing in 11.59miles is very different to 1099ft in 16.68 miles. Average gradient is about 80% greater.
- Did she enter accurate weight data in Zwift?
Seriously, the number of obvious flaws in drawing the parallel you did is stunning! Perhaps you have more information you didn't share, but as you posted it, the anecdote was worthless.
For me, Zwift may be slightly optimistic on speed but not dramatically so, and it doesn't matter anyway, so long as it's the same for everyone. My Zwift rides approximate pretty well with outdoor rides on the tri bike, but I'm slower in reality on the road bike. If I held the position the avatar does for the full ride, that would probably be close too but I rarely stay low on the road bike.