Yeah, gps-based speed is solely off of 2 dimensional position (latitude and longitude) divided by time between the data points. I wonder how far off a GPS device's calculated speed would be if you're biking up or down a (very) steep incline. I guess it would be a pretty easy test if you set up 2 head units: one that measures speed from GPS and another that uses a wheel-based magnet to measure speed and then compare the results. Unfortunately for me, living in Chicago pretty much prevents me from doing any sort of test like that. I wonder if Ray from DC Rainmaker has done or could do a similar test (hint, hint).

EDIT:

Okay, I got bored and did the math. The steepest road in the world is apparently Baldwin Street in Dunedin, New Zealand. The average slope is 19 degrees or 35%, which corresponds to 1 meter up for every 2.86 meters across. This puts the ratio of the horizontal (GPS) distance to slope (real) distance at 2.86:3.03, which gives an average difference of 5.6%. I.e., if you were going 30 mph as measured by a wheel based sensor (your "real" speed), your GPS device would display 28.32 mph, which is a calculated displacement.

Sorry to sidetrack this thread. Really cool skydiving data, by the way. Too bad you broke some serious FAA regulations by having a GPS device turned on in a plane :) Hope no one reports you!