PT 43 results:
RC: -2
LR1: -4 (Sigh)
LR2: -1
LG: -0
Raw: 94
Scaled: 175
On the other hand, results of PT 17, 4 sections of which I mixed in the last PTs as experimentals:
LR1: -1
LR2: -1
LG: 0
RC: -1
Raw: 98
Scaled: 180 (

- My question from yesterday about the reliability of the experimental scores became suddenly very relevant

)
And current averages with 37 PTs left behind:
LR: -5,11
LG: -0,92
RC: -2,24
Raw: 92,68
Scaled: 174,00