What is self-reported data?
Self-reported (or declarative) data is a measure you produce yourself: a rating of your smash, your morale, your sleep, the effort you felt. Every scientific approach starts this way: observe, record, count, before explaining.
A rating you give yourself after each match is exactly that: a regular observation, taken under the same conditions, on the same scale. It’s the first floor of measurement, the one everything else is built on.
Forty years of sports science
In 1982, Gunnar Borg laid the foundations of the rating of perceived exertion scale 1. A meta-analysis later showed that this rating is linked to heart rate, blood lactate and oxygen uptake 3. In 2001, Carl Foster proposed multiplying a session’s effort rating by its duration: that’s session-RPE, which has become a standard measure of training load 2. It has been validated in many sports, for both sexes, at every age and every level 4, 5, 6, 7.
Wellness questionnaires (fatigue, stress, muscle soreness, sleep) follow the same logic 16, 20, 21, 22, as do mood profiles 17, 18, 19 and sleep questionnaires 24, 25, 26, 27. International consensus statements on athlete monitoring recommend them alongside physiological measures, from the International Olympic Committee to the European and American scientific societies 28, 29, 30, 31, 32, 33, 34, 35.
And in racket sports?
In tennis, perceived effort during a real match is linked to blood lactate 8, 9. In professional players, load measured by session-RPE tracks load calculated from heart rate 10, and it has been used to guide training at Roland-Garros 11 as well as with young players 12. In squash, it distinguishes between session types in professionals 14. One study compares the load of padel with that of singles and doubles tennis 13, and a review covers racket sports as a whole 15.
To be honest: we haven’t found a study that validates session-RPE specifically for padel, pickleball or table tennis. The results from tennis and squash are the closest.
More sensitive than many sensors
The systematic review by Saw, Main and Gastin (2016) compared 56 studies: subjective measures reflect training load, both acute and chronic, with more sensitivity and consistency than commonly used objective markers 23. Across more than 2,500 questionnaires in Australian football, wellness ratings track changes in load over the week and over the season 21.
An important caveat: these studies concern self-reported effort, fatigue and wellness, not the technical rating of a shot. For shots, the argument is different, and it comes just below.
What if I rate myself wrong?
We judge ourselves imperfectly: on average, self-assessment is only moderately linked to actual performance (a correlation of about 0.29) 37, 38, and it is often off, most often upwards 39. But it becomes more accurate when it concerns a specific domain, a familiar task, and when you’re used to assessing yourself 37, 38. In junior tennis, the shots estimated by players do not differ significantly from video analysis 36.
Above all, a constant bias doesn’t get in the way. If you always over-rate yourself by two points, all your ratings shift by the same amount, and the gap between your good and bad days stays the same. What matters isn’t a rating’s accuracy but its reliability: judging the same thing the same way every time 42, 43, 44.
That’s why the app compares you with yourself, never with others. What holds for a group often doesn’t hold for an individual 48, 49, and it’s by measuring the same player repeatedly that you separate a real trend from chance 50, 51, 52. Repeated self-reports, recorded in the moment, are reliable and limit memory errors 45, 46, 47.
The limits, no beating around the bush
A rating doesn’t say whether your smash is objectively good: it says what you perceived. It’s sensitive to your mood at the time and to the result. And it describes associations, never causes: “I win when I rate my smash high” doesn’t prove that the smash makes you win. It helps you spot things, not draw conclusions.
What MySportAPP does with it
Post-match ratings are treated as consequences of the result, never as causes: they don’t go into “what makes you win”. They’re read by comparing your wins with your losses, with minimum match counts, a displayed reliability level, and a pull back towards your usual level when matches are scarce.
Your level, meanwhile, is calculated as in the best tennis prediction models: a rating that goes up or down after each match depending on the opponent and the score margin, along with its uncertainty 53, 54, 55, 56. The points in a match aren’t entirely independent 57, and individual statistics help predict the outcome of a match 58.
The question is never “is this shot good?”, but “when you perceive it as successful, does your result change?”. And that relationship can be measured very well.
Tomorrow: how you feel and your body, side by side
Sports science distinguishes external load (what you do) from internal load (what it costs you), which includes perceived exertion 7. With the watch, your heart rate and your measured shots will join your ratings: the effort you feel, next to the effort your body records.