Your question is Evaluate Model Calibration. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
You have a model that ranks cases well, but the team is asking whether its predicted probabilities can be trusted. The scores are used to decide when to auto-approve, route to review, or hold back a decision. You need to judge whether the model is well calibrated.
How would you evaluate whether a model is well calibrated?