The other issue with live tracking is the latency – it usually seems to lag 5-10 seconds behind the TV. I’m not sure how much of this is down to server-side processing but I suspect a fair chunk of it is a result of the trackers using SMS for backhaul.

It would be very interesting to compare the performance of the specialist GPS trackers vs just giving each athlete a cheap Android phone to run with (e.g. Moto e15, €80).

Urban canyons (multipath errors)

Sprint maps are often in city centers, with tall buildings, walls, and narrow streets.

GPS signals bounce off surfaces (multipath), so the receiver “thinks” it’s somewhere else.

This can lead to errors of 10–50+ meters, which is huge when you’re navigating alleys only 2–3 meters wide.

Dense vegetation and structures

Parks with trees, tunnels, or underpasses block satellite signals.

GPS units may lose lock completely, then “snap” to the wrong place when reacquired.

Update rate & smoothing

Many GPS trackers used in orienteering update at 1 Hz (once per second) and use aggressive smoothing to save battery and bandwidth.

In sprint, where athletes change direction every few seconds, this creates delays and makes tracks “cut corners” unrealistically.

Limited constellations / old hardware

Some trackers still rely mainly on GPS only (US satellites).

If GLONASS, Galileo, or BeiDou aren’t used, accuracy suffers — especially in cities.

Transmission constraints

Real-time tracking usually relies on GSM networks (4G/3G/2G). In crowded city centers or rural areas, connection can lag, producing “jumps” in live tracking.

🛠️ What should be done to improve it

Use multi-constellation GNSS

Modern receivers can combine GPS + GLONASS + Galileo + BeiDou.

More satellites = better geometry = fewer urban canyon errors.

This is a big upgrade over older trackers.

Increase sampling rate

Switch from 1 Hz → 5 Hz or 10 Hz logging.

Even if transmission is downsampled, higher raw resolution allows post-race smoothing and correction.

Sensor fusion (GPS + IMU + map-matching)

Combine GPS with inertial sensors (accelerometer, gyroscope).

Apply map-matching algorithms (like those used in car navigation), which snap tracks to passable routes on the sprint map.

This avoids the common “running through buildings” problem.

Smarter live corrections

Apply real-time filtering that respects impassable barriers (walls, buildings, fences).

Example: If GPS shows an athlete inside a building, the algorithm corrects to the nearest valid path.

Better antennas and placement

Trackers often sit in vests or harnesses, where body/shielding causes signal loss.

Placing units higher (e.g., on headbands or lightweight chest rigs with sky visibility) improves reception.

Use of differential GPS (DGPS) or RTK in major events

Overkill for small competitions, but in World Championships or World Games, organizers could use ground reference stations to correct GPS drift in real-time, achieving sub-meter accuracy.

✅ Summary:
GPS tracking in sprint orienteering is bad mainly because of urban canyons, multipath errors, and slow trackers. The fix is a combination of modern multi-constellation hardware, higher update rates, and map-aware filtering algorithms. With today’s GNSS + AI techniques, it’s possible to make sprint tracking look far more realistic — but it requires investment in new equipment/software.