Don't be fooled. There is a lot of confusion around live tracking of athletes at endurance events.
The issue has been compounded in recent years with the advent of event-provided checkpoint tracking.
No solution is perfect. Here's the lowdown on what you should know.
There're 2 basic forms of live tracking at a race:
- athletes pass by checkpoints along the course. Once their pace over
previous segments is known, an estimate of their current position and ETA
2. GPS or location-based
- athletes carry a GPS sensor that also has a live data connection to a server
online. GPS positions are then interpreted by an engine, translated into
course progress, etas, and other metrics about their race, then shown to
fans in real-time.
Event-provided tracking is usually checkpoint-based. They have timing chips
and mats to use as checkpoints so adding an app is relatively easy. Timing
chips are usually simple devices that only work when athletes cross timing
mats or antennas instrumented along the race course. They make good
Almost every other tracking device that athletes bring themselves are
GPS-based. Examples include dedicated live tracking devices, specialized
athletic watches, bike computers, smart phones, or smart watches. Many are
capable of live tracking, although some require the use of a phone for
internet connection, and each option has its pros and cons. I'll have a
look at the pros and cons of these different kinds of devices with regard
to live tracking in another article.
Now that we have the high-level differences established, let's have a
deeper look at each. There are some important considerations, pros and
Race progress is estimated and measured based on when an athlete passes
known checkpoints - usually timing equipment. Then their current position
and ETAs are calculated by an algorithm somewhere along a course. Often the
current position is shown animated as a moving pin on a map or even looking
like GPS-nav style animation.
- usually provided by the event as part of the entry fee and included
at no extra charge.
Nothing extra to carry
- it uses the timing chip, which they need for official times anyway.
Useful race info
- Event apps usually provide other useful information for fans like
Race map doesn't have to be accurate to work
- The map showing in the app can contain many errors and still give
an ETA. Since the app functions based on crossed checkpoints at known
distances it doesn't have to be accurate.
Easiest way to track everyone
- everyone is carrying timing chips, so everyone can be
tracked via checkpoints. Rankings and comparisons can be made as
athletes pass each checkpoint. It is much more likely that you can compare
an athlete to their friend, to a group, or to the leader since they
all have timing chips.
Alerts based on checkpoints
- Fans get push notifications as their athletes' progress over
checkpoints. If the entire system is running efficiently this can be
within seconds, although sometimes there's a delay of minutes.
- Because the sample size is everyone, it is easy to observe race-day
conditions effect on athletes. It could be an exceptionally hot, wet or
windy day, and everyone slows down. This kind of trend is easy to see
as course segment times drop through the day, especially if the race is
long and they progress over the same path multiple times.
"Live" or "real-time" is misleading
- In checkpoint-based tracking, animated map pins are a simulation. The
only real-time reports are when athletes pass a checkpoint. Other than
those moments, their fans won't really know where they are. The problem
is, because of the simulation, fans often think they do. This can lead them
to worry needlessly, go to the wrong place, wait much longer than need be,
or miss out when their athlete has a major problem.
Stops or jumps
- When the estimate is off then the athlete will not reach the next
checkpoint at exact moment of the estimate. This results in the
tracking simulation either stopping, because the estimate was faster
than the athlete, or jumping forward when the athlete hits that
checkpoint ahead of schedule.
- Timing mats are relatively far apart when compared to GPS data. This
varies from course to course, but it can be difficult for organizers to
field a lot of timing mats or other checkpoint devices.
Imagine that you're watching the race on an old television. You'll get
an idea of what's going on, but it can be fuzzy. They can add more
checkpoints, but it simply can't compete with a good GPS device.
- Because of the above they don't have frequent updates to provide
accurate ETA updates with. This is especially true if athletes' don't
progress at a consistent pace. Speeding up, slowing down, or stopping
causes a lot of issues with checkpoint ETAs.
Relies on continual pace
- so long as an athlete is consistent then checkpoint "live tracking" looks
fine and the ETAs are ok. However, if an athlete speeds up, slows down,
stops for a flat, or any other reason, no-one following will know about
it. In addition, the algorithms typically do not include adapting to
environmental conditions like hills, headwinds, or inclement weather.
Fans won't know when an athelte is in trouble
- Whether stopped for a flat, potty break, gotten off-course,
or any other of a number of other issues, checkpoint tracking
can't tell anyone.
- sometimes timing mats don't capture an athlete crossing. Other types of
checkpoints may register athlete's passing multiple times which can
No SOS button
- timing chips don't have the ability for the athlete to press SOS to
report when an they have an issue.
Checkpoint data can be delayed
- Timing mats also require a data connection, usually cellular, to
transmit to timing systems. Cell networks sometimes have issues
especially during events. Or timing mats are placed in a location where
no cell data connection is available, and the data has to be sent in
via other means. Often these timing systems have software of their own
to process data before sending it off to live tracking apps. All of
this adds to delays and not always alert fans in a timely manner.
While widespread and available at many events worldwide, the biggest
downside of checkpoint tracking is it leads fans to believe that athletes are being GPS tracked. The
most common visual is a smoothly animated opaque pin or dot. This is
deceptive. Many fans don't see, understand how/why, or simply ignore the
little warnings that say it is an estimate only.
My takeaway is that visually treating checkpoint tracking like GPS tracking
is generally not a good approach. There are more accurate and honest ways
to present this information that will lead to a better experience for those
For now, my advice to fans is to be very cautious about believing what they
see with checkpoint-only solutions. They may be accurate sometimes but are
often off by minutes or more.
When trying to meet an athlete along the course, arrive early and be patient.
They probably won't pass by when the pin moves past your location, even if
it sometimes worked in the past.
: You'll know if an event's tracking is checkpoint-based if they say
"checkpoint" or use the word "algorithm." In case
you don't know: algorithm is nothing more than a big word for
describing how to do something. A cooking recipe is an algorithm. They
range from very simple to extremely sophisticated. Classically, it was used
most by mathematicians and computer-scientists. Marketers and sales people
use it to make tech sound fancy. In the case of checkpoint-based tracking
the algorithm may as simple as an average pace from previous course
segments, forecast to the present. At that basic level it can be very
inaccurate for a wide variety of reasons.
GPS tracking far more accurate and useful for live tracking purposes.
Athletes regularly carry race-legal GPS devices. Some devices also have a
live data connection and can be used for live tracking. When live tracking
is turned on these devices persistently report their current position at
regular intervals (e.g. every 30 seconds). Since it is being carried by the
athlete it follows them no matter where they go. Course progress, ETAs and
estimated real-time positions can be provided in real-time more accurately.
However, there are some challenges with GPS tracking as well.
- GPS tracking typically has more than 10 times the data points than
checkpoint-based. It's like watching a race in 4k vs watching on an old
Accurate and updating ETAs
- ETAs are based on more data and this data is updated much more
regularly. Changes in pace are reflected much more quickly and the
overall ETA is more accurate.
Accurate real-time estimates
- GPS is typically updated periodically, say every 30 seconds or 2
minutes. This means that most of the time athletes are ahead of the
last reported position. An estimate of their position can be much more
accurately made than checkpoint.
Adapts to changing conditions
- If an athlete slows down because it is 107 deg F outside, speeds up
because they're having an awesome day, has a flat tire, or encounters
strong head or tail winds, fans will know. The athlete carries this
sensor so their changes in pace are accounted for by default.
Fans know where they are too
- GPS-tracking apps usually show the location of fans at the event too
using their smartphone's location service. This is often not provided
by event or checkpoint-based apps since they aren't GPS accurate.
Freedom! - more than tracking on the race course itself
- Because GPS is global and unrestricted, it will track an athlete before or
after the event. Allowing fans to find them after they've finished
the race, when they're off-course, or doing just about anything.
Continues to track after Finish/DNF/DQ
- Fans sometimes need to find athletes when they're off-course. After
the race it can be hard to find an athelte in the finish area packed with others.
Or perhaps they missed a cutoff along the course and are heading back
in a van. If an athlete has DNFed they will want their fans to find them
too. Especially if they've made it to the Medical Tent, where race
organizers don't always let fans know if their loved one is inside.
Timing systems not necessarily needed
- Creating your own event is entirely possible and eventually even professional
events can run without the need for expensive timing systems or services. It won't be
as certain or reliable as timing systems for a few years, but it doesn't always
have to be, and eventually centimeter-level, accurate, and reliable devices
will be common.
Race support and SOS
- SOS can be tied into event organizers' first response teams. The
support vehicles can be centrally tracked and monitored in real time on
the same map as everyone else. We regularly help support vehicles at
events, so they can keep track of their vehicles in real-time.
Even for individual fans support is possible as they can find their athlete
and help them at events where this is permitted.
Other sensors and data are possible
- live tracking devices can include other sensors like accelerometers
or g-force sensors. Alerting people when an athelte has fallen over.
They can also include ANT+ or Bluetooth to collect and transmit data
from power meters, cadence sensors, heart sensors and more.
- Alerts are not tied to specific checkpoints and can be sent as
athletes complete regular splits, course segments, hit objectives like
transition areas, or are temporally close to fans or objectives. These
can be delivered as push notifications to fan's devices, SMS messages,
emails, or social network updates.
Your own device
- Athletes need to provide their own live tracking device or rent one.
This device needs to connect to a live tracking app or service.
We provide both.
Carrying 1 more thing
- there's no way to get around it. Since timing chips don't have GPS
and cell data athletes have to carry a location-aware + live data device.
Carrying one varies a lot in difficulty depending on several factors,
and there are pros and cons to each category of device and course type
- so I'll leave this to other articles.
- Data transmission requires a lot of battery power, even if a watch
will last 24 hours with GPS turned on, it may only last 13 with data
turned on, and probably not even 5 or 6 with cell active. Battery
life should be considered within the context of each
kind of device and so we'll discuss this in future articles.
- Most devices rely on cell networks to transmit data back live. Cell
networks have coverage gaps that vary by area, protocol, and carrier.
The common misconception here is that GPS is everywhere - which is true
- but to live track the device also needs a data connection, which is
where cell or satellite data networks come in.
: At the time of writing there is a major transition in IoT cell
networks. US carriers have generally degraded and will soon be shutting
down the classic 2G and 3G networks. The transition to specialized IoT
LTE protocols will have many benefits. It will be better coverage area,
building penetration, battery life, and supported for the foreseeable
future. If you're in the market for a cell-based device, be sure it is
using LTE Cat-M1 or LTE NB-IoT for coverage in 4G countries, with 2G as
a backup if you wish to be tracked globally.
: Some devices cache GPS data when in a blackout area, sending it to
the platform when a connection is restored. We've observed that this
capability adds a little extra complexity and may cause a short delay
in information being shown live, but it also improves accuracy.
Cellular network challenges
- Sometimes cellular congestion is an issue when too many devices
connect to the same cell pole. These issues vary from location to
location because they depend on the quality of the cell networks in
each area. When the event is in a remote location, expect cellular issues.
They don't usually have enough coverage or capacity to handle
We also have a couple of techniques to cut through this with our
service for live tracking devices, but fans' phones may be affected
Terrain effects accuracy
- When near tall buildings, indoors, in a lot of trees, or in a canyon
- accuracy can be off or lost entirely. Usually this is due to 2
reasons: satellite signals are blocked or reflected. Tall buildings
both block and reflecting GPS signals, which cause inaccurate results.
Because water also blocks GPS and data connections, it is most notable
when tracking swimmers. For it to work while swimming the beacon must
have visibility to the sky or have large antennas dragged behind. In
most races the large antennas are not feasible, so the best option is
to place the beacon in the back of a swim cap or secured to goggle
straps. If the device is only water-resistant it must also be protected
Course maps must be accurate
- if the event course is plotted inaccurately, or the map itself is
off, problems with calculating progress, ETAs and current positions
arise. Drawing accurate maps is tricky so it is important to pay
attention to details. This most commonly occurs when an event changes
their course days before the event and doesn't update online maps or
doesn't provide an accurate map to begin with. Personally, I've updated
more than a few maps as the event is unfolding due to inaccurate event
When done right this is a pro for fans. They will be getting an
accurate representation of the course which is often lacking in
If you're creating a race map, it is often better to switch to "draw
lines" for sections where the "follow roads" feature doesn't work. Also
try switching between walking | cycling | driving. Sometimes follow
works better with one setting over another.
100% distance accuracy is not realistic
- More something to understand than a con: There is almost always a
discrepancy between a course's claimed distance and the distance of the
line plotted on the map. That 112-mile bike ride might plot out to
111.52 miles in our system. This can be due to a wide variety of
reasons, like the course itself isn't at the claimed distance, or
technical issues like map inaccuracy, plotting errors, or even the type
of distance algorithm used to measure a line (flat, spherical, etc.,
there are several). We allow some slack for this in our system so fans
have a better experience during the race.
Implementation is difficult
- GPS data is agnostic to race courses. At the most basic level, it is
nothing more than a pair of coordinates, a timestamp, and sometimes
speed and bearing. There's no inherent knowledge of how far along a
course they are. And GPS data points are never exactly on the course
line. Events with multiple segments, loops, and out-and-backs are even
more difficult to interpret. This makes it work for event live tracking
is hard. Live tracking platforms must interpret GPS data relative to
the provided course map, race start times, athlete prior progress, etc.
Events can be complex, containing multiple disciplines, transition
areas, and segment types. They often include loops, out-and-backs, and
data blackout areas. All of these make it a much more challenging
problem to derive their course progress, ETAs, and predicted real-time
position from GPS data than it is with checkpoints.
That last point is probably the biggest obstacle to widespread live GPS
tracking at events. It is more difficult to build the engine that will
always give accurate statistics using GPS data alone than it is for
checkpoint solutions. Even with GPS tracking it is very important
to move beyond pins or dots on a map, which raises the bar even further.
Our platform uses patent-pending AI and a few other techniques to assist
with many of the cons. It is robust to many of the above challenges. We'll
be adding features soon and more statistics now that we have this
The best of both worlds? Checkpoint and GPS combined
The best possible live tracking experience in the near-future could use
both checkpoint and GPS data. By incorporating both many
of the downsides can be mitigated while getting the best of both. As such
we are working on integrating checkpoint data into our system so that your
next race could have the best possible tracking experience. Our first
integration will be with race|result's systems.
There is much more to come on this and will be discussed in future