Reading a GPS track of a training

A phone in your pocket records a point every few seconds for the length of a run. That gives a line on a map and a handful of numbers. The numbers look exact. Some of them are.

Distance

The distance of a track is the sum of the straight lines between points. On an open trail that is close to the truth. Under dense trees, in a narrow valley or when the phone sits deep in a sled bag, the points scatter and the sum grows. A track that says 23.4 km on a route you know is 23 km is doing well. A track that says 26 km on the same route had a bad hour of reception, not a longer run.

If you run known trails, keep the route's own distance as the reference and treat the track distance as a check, not as the answer.

Riding time and rest

The time from start to stop is not the time the team ran. Water stops, a tangle, a dog that needed its booties checked: all of it counts as rest, and rest can be a third of a long training. The number you want to compare between runs is riding time, the time the team was moving.

Splitting the two from a track needs a rule, because a team never stops cleanly. A standard way: rest starts once the team has been standing still for a set number of seconds, within a radius big enough that walking along the dogs with the phone does not count as moving. Nordspann uses exactly that, with the seconds and the radius set per kennel. A tight rule gives more rest and less riding time; a loose one does the opposite. Pick one and keep it, so the runs stay comparable.

Pace

Riding time divided by distance gives pace, and pace is what a track is for. It only means something on the same trail at a similar temperature, which is why the two numbers next to it matter: the route the track rode, and how cold it was at the start.

Temperature is the one number worth looking up rather than remembering. Dogs run faster in the cold until it is too cold, and the difference between minus five and minus twenty on the same loop is real. A log that has the temperature with every run can tell the two apart.

Matching a track to a route

A loose track is a line on a map. A track that is recognised as a run of a known route is a data point in a series. Once the app knows that this morning's track rode the lake loop, it can put it next to every other run of the lake loop: how often it was run, when the last time was, and how the times compare across the season.

That is also why a track that clearly rode a known route should not become a second copy of that route. One route, many runs.

What to do with a bad track

Throw it away and type the result. A training with a hand-typed distance and time is a good log entry. A training with a track that wandered through the forest is a wrong one, and it will sit in the averages forever. The phone was in the wrong pocket; the run still happened.