You're right. I meant 16% for 60% condition loss. (4% on average at every session consuming 15% of condition)

Al Svanberg had empirically estimated the TEF (training efficiency factor) to 0.28 for a 5* player.

Just scored a 18yo DC repeatedly gaining 5% attributes for each 18% condition loss (6 times press the play)
TEF = 5/18 = 0.278