So while I’ve been using Zwift, the virtual cycling environment, for a few months, it has always been at the back of my mind as to how closely Zwift reflects conventional cycling on the road.

Clearly it is not possible for a cycling simulator to exactly match a ride on the open road, but I wanted to know how close the two environments were from a data point-of-view.

Recently, due to weather and time restraints the opportunity to make a side-by-side comparison came about.

My equipment comprised a set of power-meter pedals (power and cadence) and a heart rate monitor. The Zwift session used a standard fluid turbo trainer communicating with the Zwift software. The road version used a Garmin Edge 1000 to store and calculate data. Both sessions were uploaded to and analysed by Strava and StravistiX.

I chose some local roads that I thought best reflected Zwift’s Richmond course, and set the test distance to 32km. Unfortunately I got the road elevation wrong but I think the comparison still stands when power, suffer score and intensity are taken into account.

The results came as a minor surprise, the two environments were closer than I thought they’d be.

I admit my test wasn’t exactly empirical, but it did show, for me, Zwift cycling is a very good, and accurate, training environment for continued improvement on the road.