Zwift vs Reality

If you ride a bike you can’t avoid hearing about Zwift. Virtual Reality riding linked to your smart trainer to make that winter training more bearable. For some, including me, it’s a massive win keeping me motivated to sit on my turbo for over an hour this evening trying to set some new PRs on some virtual hills.

For others, they can’t understand why we’re not out in the cold wet and dark getting killed by winter motorists as well as cold wet and miserable.

But each to their own…

Big bone of contention is that Zwift racks up your mileage and climb totals for the year on Strava, which annoys the hell out of people.

So, should it count? Is it close enough to reality?

I only really care about going up hills, so let’s look at that. First, I remember Ben at posting an article about how to find segments you’ve ridden that are like a climb of interest you won’t get to ride. It’s here. So we can use that to find a segment that suits us for comparison.

To save me the bother, Andrew pointed out that Half China is quite a lot like Carr Lane. Which is a segment B&DCC, the club we ride with, runs a hill climb up so we’re quite familiar with.

How to compare? Well I figure VAM is a good one. I’ve only ridden Carr Lane 7 times, which is nothing compared to Andrew who is over 80 repeats. My VAM ranges from 652.4 3 and a half years ago to just over 1000 for my PR and last April.

My VAM for Half China goes from 844 a few days ago to just over 1000 just before I got ill late last year and slumped my fitness.

I’d say that’s a pretty fair comparison and climb on Zwift counts then.

I checked one or two other segments, seems close enough to me.

Strava Centurion

If you use Strava, have a premium account and don’t  have a power meter. Strava Centurion might be of interest.

It’s an open source app that I wrote with two guys from work into cycling to take a TCX file from a garmin and perform power calculations, producing an output TCX file that contains wattage points as if you’d ridden with a power meter.

If you then upload this file to Strava, instead of getting the limited charting/graphing/analysis of Strava’s own power estimation, you’ll get the full premium user power meter based data.

Now, this is still power estimation. It’s based on the same physics and handwaving as the Strava power estimation detailed here. Only you have finer control over the simulation of reality that is provided.

This doesn’t mean it’s any more accurate. Just that we had more fun playing with the physics. And you can use all the strava power charting with the estimated data.

So, this is how you use it:

  1. Run the app
  2. Click the reality tab
  3. Configure your reality (only need to do this once, or if you want to change bikes/temp)
  4. Go back to the TCX tab
  5. press the … button to load a TCX file
  6. Press Power Xtreme

Then in the same folder your input TCX file was found in, will be a file with the same name but a load of numbers (they’re the current date and time). That has the power. Then you use the Strava upload from file to put it into your Strava account.

Configuring Reality:

Bike Weight and rider weight are in KG and are self explanatory I hope. I record my bike weight as the actual weight of my bike with the pump, spares, full bottles etc in place. Obviously the weight goes down as I drink the water. But it’s not going to be much more inaccurate.

And my weight might vary a bit as I wear different kit and carry more or less gels/bars and eat them etc. But. It’s not going to be much more inaccurate. It’s estimated power.

The temperature won’t have much effect, it’s used to estimate air density, and thus resistance, but, seriously do you notice on a cold day when the air is heavier it’s harder to push through? It’s there as we were having fun with physics.

That leaves the interesting fields. These are the coefficient of rolling resistance, how much power you lose to your tyres on the road surface. This is going to vary massively over a ride. The potholes and smooth surfaces and rough surfaces are going to vary. I’ve got a nice default in there for thin, slick tyres on average road surfaces.

The frontal area and drag coefficients affect the calculations of pushing through the air. It’ll be different for the Fat Lat At the Back on the Mountain Bike with the non-aero position to the climbing sparrow on the full TT bike. I’ve got an “on the hoods” average road bike figure in as the application default.

However, here is the chart I cribbed it all from, and you can probably find other figures round the internet to use:

Drag Chart

Drag Chart

And here is the Release of the app to play with (7zip archive, just download and extract). It’s a bit techie. But. It might be of interest. Source code is available on github here.

Why Centurion and Power Xtreme button? Blame my childhood.