Very good question... For me, this is like religion: you can choose the way you believe in God. So, I would suggest to put both in your results. I made this in some of my last papers:

https://www.researchgate.net/publication/233839090_ACUTE_EFFECT_OF_DIFFERENT_HALF_SQUAT_SET_CONFIGURATIONS_ON_JUMP_POTENTIATION?ev=prf_pub

https://www.researchgate.net/publication/233739174_Cardiac_Autonomic_Adaptations_in_Elite_Spanish_Soccer_Players_during_Pre-season?ev=prf_pub

Sometimes, some colleagues forget that statistics is a tool and not a purpose in itself.

Cheers,

Daniel

Article The Acute Effect of Different Half Squat Set Configurations ...

Article Cardiac Autonomic Adaptations in Elite Spanish Soccer Player...

I would agree and we have done this with some recent papers too. However, my one concern is that this can be confusing in a paper particularly when, as quite often happens, the two methods suggest different outcomes!?

OK. In such a case, academic industry rules: you would select the method depending on the journal (editors) you will submit your manuscript. There is however one important aspect that should not be ignored: inferences are more "permissive" which is an important advantage for finding positive outcomes when analyzing data on small samples as usually occurs in sport. Hope be clearer now.

I agree with Daniel. And I think that the outcomes of both statistics have different implications. Thus, it is not problematical that the two methods may suggest different outcomes. Whereas the conventional statistics will give a probability whether a possible effect is due to chance or a real effect, the magnitude-based inferences give us some idea to what extent a finding may have clinical or practical relevance. Thus, I agree that it is appropriate to report both stats. Moreover, we must be aware that stats are only a tool which may help to find the best conclusions. Stats will never substitute our subject-specific expertise.