Small sample size, but towards the end of the round-cell era I was running NiMHs straight from bulk packaging and comparing them with matched packs on my own test equipment. The "unmatched" cells were better.

Matching may have mattered with 80s cell manufacturing technology, but in the 00s the manufacturing was so consistent, all the matchers were doing were putting a few extra-heavy load cycles on the cells before the customer got to use them.

You'd think there was money to be made in "matching" LiPos, I wouldn't be surprised if someone has tried it but found that the numbers you get wouldn't make a persuasive sales pitch.