Socionics trumps MBTI imo because it's more likely to lead smn into actually learning the IEs and understanding whether the typing is correct and how IEs manifest, whereas MBTI seems less organised.
And I too do know about a case of a friend who was evaluated at work by paid for MBTI staff who told people by the end of the session - "you know, look at the results, but also at all the other types, cause we tend to adapt our behaviours and your test results might not be reliable." This kind of undermines the whole exercise, lol.
Although it proves to be right sometimes, especially for people who were forced by life circumstances to act out of character in order to survive in unfriendly conditions (be it family-wise or work-wise).
However, when I think about people I know who took the MBTI tests - their results in most cases correspond with socionics (with j/p switch for introversion). Cases in which they don't match up are either people of introverted subtypes of extroverted types or the other way around in two-subtype Socionics divide and/or Enneagram 9s and/or instinct stackings are skewing results and/or struggling with depression.
(I'm not saying that this is your case, just saying that's how it works with people I know irl.)
I know a person who is in Socionics ESFp (sp/sx) who tests as ISFP in MBTI and a person who is in Socionics ISFp (e9 so/sx) who tests as ESFP in MBTI.
But yeah, in cases where things don't match up directly the explanations require employing other systems such as i.e. Enneagram and I can understand that smn might say this is too much of a hassle.
Also, imo the way a lot of free online MBTI-inspired tests are constructed is overtly simplified and they just work in a way that if you like to spend a lot of time with people = E, if not = I, if you report relying on intuition = N, if not = S, if you say facts trump feelings = T, if feelings trump facts = F, if you're judgemental = J, if not = P.
That's a pretty vast oversimplification and has hardly anything to do with MBTI functions.