Closed models refuse more at max reasoning effort, skewing evaluations

Here's a quirk when evaluating closed models on tasks that trigger refusals (cyber, bio etc):

low reasoning effort => few refusals, high accuracy
max reasoning effort => many refusals, low accuracy

This gives you test-time compute curves where max effort does way worse than low because the refusals tank the performance.

What is the fairest way to report performance in these cases?