Is there any public data how the "autopilot jail" decision was decided? I.e. - 80% of the prior accidents were caused by drivers that got >5 autopilot disengage

Is there any public data how the "autopilot jail" decision was decided? I.e. - 80% of the prior accidents were caused by drivers that got >5 autopilot disengagement strikes?