Procedure turn accuracy
I recently got my GPS to record my track whilst carrying out some published holds and instrument approaches. I then overlaid this track on the approach map (on a computer - I didn't sit there plotting points on a paper chart!). I was a bit surprised to see how far off I was in respect of the radius of the turns (particularly in a holding pattern - base turns weren't so bad as I was then tracking something back to the airfield so didn't do a strict rate 1 turn for 1 minute).
I was doing this in about 30knts of wind, but let's assume nil wind. The turns on the map, for a standard hold, are approximately 1.6nm wide (by wide I mean the diameter of the circle of which the turn is a half of). The base turn for a category A aircraft is a similar size. However, a rate 1 turn at 100knts is only about 1nm wide. So clearly doing a rate 1 turn isn't going to follow the track on the map.
My question is, therefore, am I correct in assuming that a properly flown procedure is one which follows the rules (a rate 1 turn for 1 minute) rather than one which follows the track on the map?