The biggest inaccuracies of the 305 is the altitude/elevation profile information. However, it seems to be better when your runs are not completed along the ocean. For some reason, all my runs along the coast end up with excessive errors, claiming I have climbed a few thousand feet in 10K. Clearly, I barely left sea level.
If you're using the new Training Peaks WKO+ software to monitor your running, with rTSS, NGP and IF, you are in luck! You can actually correct the elevation profile of your Garmin 305 files, to get the right data.
How does it do this? WKO+ takes the coordinates of your run, and connects to a database of your choice, (I like the USGS website), to get the exact profile of the run you completed. This becomes very important, because a profile which says you are running more up and down than you are will give you a Normalized Graded Pace which is faster than you actually did. This in turn affects all your other calculations, such as run Training Stress Score, and Intensity Factor.
Just how much of a difference does it make? Here's an example from my run, a 10K loop around the bay and on the ocean, mostly at sea level...
Data from Garmin before correction of elevation profile:
Elevation Gain: 1917 ft
Elevation Loss: 1879 ft
Grade: 0.1 %
rTSS: 77.7 IF: 0.967
Elevation Gain: 255 ft
Elevation Loss: 259 ft
Grade: -0.0 % (-3 ft)
As I said, the problem seems to be more exacerbated near the ocean, and as I go inland it seems to get better. However, I have not tested this theory, and certainly haven't seen if it is any better or worse at significant altitudes.
So this one example how the Garmin 305 as a training tool is not perfect, but it's inaccuracies can be accounted for.
Next time I will discuss the accuracy of the 305's distance measuring.