After some thought, it also occured to me that it could be possible to rig small ground stations to track the rocket instead of the rocket tracking points on the ground. That is if the competetion rules allowed it. The math would be almost identical, but all the electro mechanical aspects of the system wouldn't need to be constrained by needing to ride on the rocket. Depending upon what the rules allow, even the calculations could be done on the ground and the on-board microcontroller would just need to recieve an ejection signal.
radman:
The only thing I can think of to account for the 9m was that I was walking at a different time and so possibly triangulating with different satellites. I just looked up Garmin (it was not Garmin I was using) and they say their receivers are accurate to 15m on average. I was expecting 5-10m.
Remember all the US GPS satellites are still partially owned and operated by the Dept. of Defense, in conjunction with the Dept. of Transportation, thus pin-point accuracy for non-US military recievers isn't a priority for roughly half the people involved in making day-to-day decisions about the system. Furthermore, I'm sure many in the Pentagon would view such a thing as a potential threat, given the abundance of relatively inexpensive GPS devices not entirely without reason. At least it's much better than the late 1990's after the DoD was ordered to make GPS a dual-use technology (did you think the Pentagon would grant public access to one of it's most beloved "toys" out of the goodness of its collective heart? 
). For the first few years they were allowed to introduced enough intentional error in the civilian-grade signals so they accuracy was at best within 100 m, not the 20 m or somewhat less you can get today.
Although there will always be some error in the system, GPS units with 5 m accuracy might exist right now. Yet, if that's so I'd imagine they are restricted to the US military and certain three letter agencies.