I agree it was translated correctly in the KJV, but not for the same reason you do:
The reason is was correct in the 1600s was because the English term "servant" did not imply, as it does now, the specific concept of a free-man employee (that is a convention of Modern English)... it could be applied to a free-man employee, as opposed to the term "slave", which could not... but it could also be synonymous with "slave". Attesting to this fact is that the term "servant" was used in the Colonies and early days of the US synonymously with "slave".
From etimology.com:
The word 'servant' has the implication of free will desiring to serve, the word 'slave' doesn't.
Again, I agree with your statement, but only in regards to how it applies to MODERN English.
There are words in Greek that correlate to the modern English "Servant". The word from which we get "DEACON" is one such word, and, like the English word "Servant" can be applied both to chattel slaves as well as free-person employees.
But if one is speaking of a free-person employee, the English term "slave" is the wrong word to use. The Greek word "doulos" is the wrong word to use, in the exact same way.
Respectfully, I would ask you why you seem so antagonistic to the concept of having a slave-master relationship with God?
So, the difference is the WILL, which is free and can choose to serve or not.
MacArthur, being a Calvinist, wants to remove the will as a real issue in serving God.
As for the word being translated into English with it's modern connotations, the Greek word 'doulos' is translated as 'servant' by both the NIV and NKJ in Jn.15:15.