Fitness Data Analytics


Simple Analytics – the analysis we offer based on individual post-training reports


  • Recovery HR – we track the pattern of the HR during the short breaks between the series. In this case, we can discover the general form of the trainee and the difficulty level of the particular training plan. If the HR values during this period are going down sharply, lower and more quickly, the trainee still has a fuel for the training (and vice versa). It is typical for each consequent break to show the signs of strain and the HR values will drop just slightly less and slightly slower than during the previous break. This is also a normal indicator because the trainee is getting tired after each series. But if the training plan is too difficult for the trainee or general fitness level is too low, after each series the recovery HR will show the very noticeable difference – it will fall down very slow and just slightly. In other terms, the difference in HR between each consequent break will be larger.


  • MIN and MAX HR thresholds – the desireable HR zone for each trainee can be personalized. When a trainee is exercising above or below this zone it indicates that something is wrong. If the HR is often below the lower threshold it indicates low effort of the trainee or that the training plan is too easy for this particular client. If the HR is above the threshold it indicated the opposite, that the trainee is trying too hard for his fitness level (age, condition and etc.) or that the training plan is too difficult for the trainee. In very rare cases it can indicate some kind of serious health issues and can be an indicator that doing health check is necessary before any intensive exercise.


  • HR peak(s) and patterns during each exercise – since we can have the context of which exercise is done in particular time of the training and we know the patterns and peaks of the HR during that exercise, we can make conclusions about the level of performance of the trainee during each exercise. The HR behaves differently if the exercise is cardio than if it is strength exercise. We can detect if the trainee is showing a high level of strain during particular exercise which targets particular muscle group and based on the pattern of HR we can make conclusions about weak points in musculature that particular trainee has.


Figure 1 – typical HIIT training session with 10 exercises in 3 main series and 6 exercises in 2 “finishing” series


The combination of parameters creates deeper context and minimalizes mistakes. The analysis of the HR in the context of, these two previously described parameters, recovery HR, and HR thresholds can tell us a lot about fitness level of each trainee and about the level of difficulty of the particular training plan for each trainee. The multiple parameters create a context which gives more credibility to any conclusion that our analytics makes. We do not follow just values but patterns of HR which gives us far more useful information than measuring single value before, after or during the exercise. This helps the instructor to tailor the training plans for each trainee and it enables the instructor to achieve a very high level of personalization.



Figure2 – The same trainee with the same training plan as Figure 1 recorded a 6 months before. An example of unstable training


  • Stability of the HR (pattern) during the training. In the Figure 1 graph, we can see the repetitiveness of the HR patterns in each series. Also, the recovery HR shows very gradual changes. This means that the trainee, that this graph belongs to, is training regularly and for at least a couple of months. This is an example of a “stable graph”. We discovered that only trainees which have stabile exercising HR patterns can have efficient and well-balanced progress. The goal is to design personalized training plans to take the quickest route from “out of shape” and instability training (Figure2) to stable training sessions (Figure1). This is crucial for the effective results, motivation, and satisfaction of the trainees.


Advanced Analytics – Clusterization of trainees according to their performance


With the help of the clusterization of clients by their performances during the same training session. We can compare them and use this knowledge to create different groups according to their performance and track the progress of each individual over a period of time inside this cluster, whether or not performances are increasing or decreasing.