Measuring walks.

I am fascinated by the metrics of my walks. When I started walking in 1999 I had a GPS, but back then I only turned it on occasionally in order to record my position in a notebook. At that time, I would measure walks using a map measurer on paper maps. I spent many a happy evening measuring the route of my coastal walk by this method.

However, map measurers would frequently underestimate the distance, mainly due to the wheel skidding on the surface of the paper and the difficulty of accurately following a route. I also started using a pedometer, but I found that this overestimated distances on hilly routes or rough terrain as my stride length reduced.

Nowadays, I use mapping programs such as Anquet or Memory-Map to measure walks before I do them. I also leave my GPS on for the entire walk in order to get a detailed log of where I walked.

Of course, none of these systems will be 100% accurate. In the past I have carried two GPSs with me on some walks, and the resultant measured distances, even when spurious track points have been removed, have been over 1% different.

However, I thought that I would have a little fun by comparing the following methods of measuring a walk:
1) The pedometer.
2) The GPS
3) A map-measuring wheel.
4) Measuring on a computer map

Setup
My first task was to calibrate the pedometer. To do this, I did a flat, level road walk for a fifth of a mile, the distance measured using a GPS. I averaged 3.1 MPH with no pack.

This was using a Silva Pedometer Plus. The 0.2 miles was completed in exactly 400 steps.
0.2 miles is 1056 feet, therefore my stride length is (1056/400) = 2.64 feet. My stride length when I last used a pedometer in 2000 was about 2.6 feet, showing that my stride length has not changed over the years.

For the main test, I thought that I would use a route local to me along relatively flat roads. I decided to measure using the following methods. All measurements were performed on a 1:50,000 map:

1) Measure the distance with a map measurer;
2) Measure the distance using a mapping program;
3) Walk and measure using a pedometer;
4) Walk and measure using a GPS (same time as (3) above).

Test
After the walk, my pedometer showed 11723 steps, or 5.86 miles.
The GPS showed 5.92 miles
The distance measured on a 1:50,000 computer map was 5.87 miles
The distance measured on a 1:50,000 paper map using a measuring wheel was 5.9 miles

Analysis
I am actually surprised by how the values from the GPS, pedometer, map measurer and computer map are all within a hair's breadth of each other. This was not what I expected from the test.

So for short walks with no real hills, there is little to be chosen between these systems.

Because the route was so short and relatively flat, I managed to keep a good pace and maintain a constant stride. My guess is that these would not be the case if I was carrying a full pack or on a longer walk over mixed terrain. These factors shall form the basis of my next series of tests.