Measuring training load for accuracy and quantification issues with Garmin

Training load is how hard your heart works. If you don't get your HR up, you don't accumulate points towards training load. Remember, these are triathlon watches first and foremost. Training load for sprints can be fixed with a good heart rate monitor. I personally use either a Polar H10 or Verity Sense (optical), which are superior to wrist based systems. For running training load, I use a Stryd power meter. Since its connected to your foot, measuring power, it doesn't matter what your heart rate is. For what its worth, the Stryd solution is fantastic. 

I don't know if this post is really just a vent or what, but the 945 is the cream of the crop for fitness watches. There is currently no better technology to capture these sorts of metrics. There are very real limitations to GPS and wrist based optical heart rate. I think maybe adjusting your expectations might be necessary to enjoy the watch for what it is - a tool to help you get better. 

Guruyay said:

Speed work - useless! 20m, 40m, 60m sprints are non accurate in terms of time. The GPS can not catch up or accurately measure the speed (important for field athletes). The HR data won't be able to accurately record the training load or stress.

Show us any other GPS watch with optical heart rate that can.

Furthermore, heart rate measurement even with a strap will never be able to respond fast enough to provide data for training load or stress for short sprints. it's a physiological limitation of the human heart.

I posted months back how the watch told me to do 15 second sprints at 2:30/km and I did them at 2:03/km but the watch said 2:45/km because the GPS takes 10 seconds to catch up (my FR10 can do it). Also my heart rate never went over 150 because im a high mileage runner it takes longer than 15 seconds to get my heart rate up the sprints felt like nothing. Now if I had the explosive power of a sprint I could get my heart rate up but I don't train sprints. I retested with a polar h10 and my hear rate still never went over 150. It will if I do 400s I can get it to my max. 

So I got 0 anaerobic points for that workout suggestion. 

I purely run and I do enjoy the training status for base, tempos and 300 - 1200m rep work. I don't do the sprints I just tack on strides after an easy run once a week. 

Thanks for the reply. The post is out there as both feedback for Garmin (although probably the wrong place) and also hoping that someone knows more about the watch than I do with regard to accurately measuring Training Load (TL) for sports and fitness other than endurance based sports (Triathlon, distance running, cycling, etc). I bought this watch as it was the "cream of the crop for fitness watches" however, as I am not a triathlete, rather into field sports (strength and conditioning) in which I have found its limitations; and it's these limitations that I was directing this post about. The Stryd foot pod that you mentioned seems like a great option for measuring power... nice one. But does this information then convert into an algorithm that measures how much stress was placed on the body (i.e. training load) as a result of that output. Similar to how I mentioned that the device does not accurately measure the stress on your body from a Strength training session.

I do appreciate the watch for sure, but my expectations of wanting more are not too high at all and I'm sure Garmin would want to hit the market of more than just 'triathletes'. 

There have been dozens of studies that validate accurate measurement of TL using the calculation of RPE x Duration (i.e. rate of perceived exertion scale of 1-10 multiplied by duration of session in minutes). This would be such an easy add-on for Garmin watches and it would cater for virtually any training session (i.e. Power training, strength training, sprint training, endurance training, etc).

Show us any other GPS watch with optical heart rate that can.

I recently worked in a professional rugby union team in strength and conditioning; we monitored player training load through GPS devices (Catapult wearable technology). We were able to extrapolate data including Total Distance, Max Velocity, High Speed running metres (HSR = distance above 5m/s) and Very High Speed running metres (VHSR = distance above 7m/s).  HSR and VHSR are arbitrary numbers above a speed threshold.There have also been studies that validate the devices (by Catapult) as reliable in capturing accurate measures of both HSR and VHSR, although they found in these studies that acceleration, deceleration, change of direction (and collisions) measures should be used with caution. As it can measure individual Max Velocity, then even better you can have individual HSR and VHSR (e.g. HSR ~ =60% of Max Velocity). 

Agreed that the heart rate measurements after sprints are unable to provide TL, however disagree that the technology and algorithms are not available to accurately quantify TL's for sports other than endurance. 

Garmin have an opportunity to step up IMO.

Thanks for sharing:-)

Guruyay said:

Catapult wearable technology

That and others of similar ilk are very different devices to to the 945 and indeed any other consumer focused GPS watch.

There are a number of similar accelerometry and GPS based wearables used in data collection for professional and elite athletes that collect far more data than any consumer focused GPS watch. Such devices are generally not designed to collect data for long periods and provided for a very different market to Garmin watches. Garmin collects data points once per second hence can collect and store data for much longer periods than Catapult which collects data at over 1000Hz. Doesn't have heart rate either if I remember correctly. 

And since you brought up player training load there's some considerable academic debate about player load based on accelerometers as that does not necessarily quantify what the athlete is doing. Perhaps a better approach is to use instantaneous force measurement devices for intermittent movements as that is likely to better inform limb loads and injury risk. So maybe that's where Garmin et al needs to be heading?

I digress. I never suggested

Guruyay said:

that the technology and algorithms are not available to accurately quantify TL's for sports other than endurance. 

What I said was

Dr Phil said:

Show us any other GPS watch with optical heart rate that can.

and you will not currently get the same level of precision from a Garmin (or other similar watch)  that you can expect to get from a wearable designed to track performance metrics for professional athletes in team sports.

Guruyay said:

Garmin have an opportunity to step up IMO.

Depends on whether they decide to change their focus to the small number of professional athletes and  away from the mass market of weekend warriors. The key with any instrument is to understand its limitations. Often times, people don't and hence have a permanent disconnect between capability and ability.

The key points to take away from this exchange are:

• Heart rate, whether recorded from the wrist or a strap will never respond quick enough to the change in effort for very short sprints. That is a physiological limitation. Nothing to do with the technology.
• GPS on its own is unlikely ever to be able to instantly recognise rapid pace changes. Data has to be collected from multiple satellites and analysed before being presented to the watch for use. That takes time. Moreover, nobody accelerates in a step-wise fashion from 6 min/km to 3 min/km. For a 10s or 15s sprint it's almost time to slow down before the sprint has even started. Hence the support of sophisticated accelerometers is required. But whether these can be applied to a consumer device is another question. The Catapult devices are bigger and clearly have more space. But who knows how this might change in the future?
• There is indeed perhaps an opportunity for Garmin to calculate a simple training load using RPE and time. However, what RPE is used? RPE for the whole session or just the efforts? Are the steps of a workout evaluated separately? With multiple steps there could many RPE entries required for accuracy.

Hmm. I think I have too much time on my hands today!

The Catapult system doesn't solely rely on GPS like Garmin does. It comes with an array of sensors that allow for instantaneous location of players. Vector | Catapult (catapultsports.com)

Don't get me wrong, I'm not trashing on their technology. It's a hyper-optimized system for field use. Comparing the two is like comparing Grapefruit to Kumquats. They are very different technologies that happen to share location finding capabilities.

Local positioing systems are far superior to Global positioning systems for tracking the activities you mention. Validity and reliability of GPS and LPS for measuring distances covered and sprint mechanical properties in team sports - PubMed (nih.gov)

The evidence suggests this isn't a Garmin problem, but hard limitations to the technology. 

The link you provided to Catapult Vector does look like it has better technology (i.e. GNSS + LPS) to measure accurately metrics such as high speed running, max velocity, total distance, etc. However, the devices that was I was comparing to were the ones we used at the club I was working at. They were the Catapult OptimEye S5 and not the Catapult Vector. The OptimEye S5 uses GPS + GLONASS which is actually the same technology in the Garmin 945. Therefore, the comparison is relevant from my view. 

100% agree that the most recent technology is even more accurate and something that they could move toward. I also agree that Garmin have a fantastic device from a hardware perspective, but I still believe they haven't tapped into the technology that is available to them to create the 'ultimate device' nor have they done the best they can to provide a means of measuring TL for activities other than endurance event training. Something as simple as an option to enter in Session Rate of Perceived Exertion (SRPE) at the end of a training session would be a massive step forward. (Validity and Reliability of SRPE).

I'm not trying to be defiant or argumentative; i have put a lot of research into this and strongly believe Garmin are lacking big time in this area. 

I appreciate your comments.