Have you ever wondered what it takes to ride a thirteen minute 1 in 20?

In 2002, one of our Aussie pros, Trent Lowe, did just that, clocking 13:02 before going on to win a Junior World Mountain Bike XC Championship title. Now, with the help of a video of the ride, a bit of physics and a lot of guesswork, I’ve had a go at reconstructing the ride.

The short answer is: you need to be able to put out about 6.2W/kg, which is 395W if you weigh 64kg. And, if you can do that, you’re probably being paid to ride a bike.

View the ride here.

For anybody who isn’t aware, the 1 in 20 is a very popular climb near Melbourne, named after its fairly constant gradient (although there’s a false flat in the middle). The current Strava KOM time is 13:20, which is based on about 16,000 ascents so far.

My approach to generating the data for this ride was to:

• Ride the 1 in 20 myself with a helmet camera and a GPS.
• Find places where both rides were in the same position, thus providing checkpoints for Trent’s ride (i.e., after 2:12, he had gone 1.142km).
• Use a physics model to work out what constant power output would be needed to get to the each checkpoint in the required time (speed data was generated here).

There a number of caveats worth mentioning:

• This heavily relies on the elevation and gradient data being accurate, but the data I have doesn’t appear to be as good as it could be. One issue is that the gradient is a lot spikier than it should be, which, in turn, causes the speed to be spikier than it should be. Compare the generated data with a real 1 in 20 ride.
• The model is assuming that there was no wind on the day.
• There’s no guarantee that the drag coefficient or frontal area (used to calculate wind resistance) is accurate.
• The system weight (bike+rider+clothes+everything else) is modelled at 73kg. That’s not necessarily true.
• I’m pretty sure the 15s drop to 215W starting at 7:40 wouldn’t match the real data.

All in all, take the data generated with a grain of salt, but I think that it would broadly agree with the real data produced.

By the way, from looking at the start and finish of the video frame by frame, Trent’s time was actually 13:01.28 by my reckoning. If anybody wants to check, I’ve got the start happening at 0:22:20 and the end at 13:24:02 (the last numbers in those times are frames, and the video is 25fps).

Update: Trent Lowe got in contact with me to confirm that my numbers are okay.

@cyclist_dave Hi Dave, if memory serves me correctly, those numbers look spot on, somewhere I have an SRM file….

— Trent Lowe (@trentlowe_pro) October 31, 2012

@cyclist_dave … I think I did about 375watts at 60kgs, I was about 19YrsO. Incredible how you worked that out. Good work!

— Trent Lowe (@trentlowe_pro) October 31, 2012

I used his Wikipedia weight of 64kg, but if I reduce that by 4kg, my model estimates his average power as 379W, which is surprisingly accurate given how much room there is for errors to creep in.