How to calibrate a treadmill

I recently discovered that my treadmill doesn’t report its actual speed. At first, I was disappointed. Then I calibrated the machine and found that it under-reports its speed. Phew!

The other day at the gym, I warmed up on a treadmill that I had used many times in the past. The speeds that I set the treadmill at were the same that I use at home, but they felt much easier.

Self-serving bias kicked in, and I assumed that my training must be working. But my growing distrust of gut reactions made me wonder:

“Although the displayed speeds are the same, what if the actual speeds are different?”

Calibrating a treadmill

To find out, I measured each treadmill’s belt and then timed 100 revolutions. A hundred revolutions seemed like enough for the calibration to be reasonably accurate.

Here’s how to calculate the actual speed in kilometers per hour:

[Belt length in meters] x 100
/ [Time in minutes]
x 60
/ 1,000

Here’s how to calculate the percentage difference in speed between the track and the display:

[Actual speed] / [Displayed speed]

Which treadmill is more accurate?1

At home, I have a NordicTrack Incline Trainer. The gym I go to has a Woodway 4Front.

NordicTrack Incline Trainer: FAIL

At flat grades, the NordicTrack has consistent speeds. The track usually runs within 1% of the displayed speed.

But at an incline, it has sudden and noticeable changes in speed, especially at grades over 25%. It feels like the belt slips. Or it’s computer-regulated and the algorithm sucks. Either way, it’s common for the speed to change a lot. The worst performance is when inclining the treadmill before running the track.

Here’s a typical calibration for my NordicTrack:

Belt length: 128.5″ (3.2639m)
Speed at “3.2 kmh”:

3.2639m x 100 = 326.39
/ 5.14583′ = 63.43 m/min
* 60 = 3,805.68 m/hr
/ 1000 = 3.81 kmh

And the percentage difference from the displayed speed:

3.806 / 3.2 x 100 = 118.9%

Woodway 4Front: PASS

In contrast with the NordicTrack, whether the Woodway is at a flat or steep grade, the speeds are reliable:

Belt length: 3.6m
Speed at “3.2 kmh”:

3.6m x 100 = 360m
/ 6’44.54″ = 53.39 m/min
* 60 = 3,203.64 m/hr
= 3.204 kmh
/ 3.2 x 100 = 100.11%

And at that close a ratio, the difference could be from the inaccuracy of my measurements.

Under-promise and over-deliver

When I first realized that my treadmill was lying to me, I was bummed. I assumed the worst, that my training was not going the way that I thought. But actually, the opposite is true.

It’s ridiculous that the NordicTrack is so inaccurate. But it’s gratifying to know that the speeds are under-represented rather than over-. I’d rather be training faster than I thought than slower.


  1. If you want a treadmill, and if you can afford a Woodway, I recommend buying one over a NordicTrack. The NordicTrack has a lot of bells and whistles. And for people that hate exercise, being able to watch YouTube is probably welcome. But for athletes that rely on accurate metrics, the Woodway is much, much better.