Some safety management thoughts. Below is a very basic and generic risk management matrix, something many of us will be familiar with and probably far more basic that the management system Boeing use.
Playing hypothetical, let's put ourselves in Boeing's safety management empire and apply this simple matrix to the issue of trim runaways. For the 737NG we may have assessed the likelihood of a runaway to be Possible with the impact being Low. This would 'score' a Yellow 6. We may then choose to reduce (mitigate) this score by applying a set procedure (stab switches etc) for an 'average' crew reducing the score to a Green 3.
Now add MCAS. The first thing we must do is assess the new likelihood of a trim runaway. It doesn't take the brains of a rocket scientist to realise that the likelihood has increased as you have an additional mechanism/bit of software that could instigate a runaway. Add-on the single-source AOA piece and the trim runaway likelihood probably jumps up quite a bit. So, one could now envisage a new likelihood score of Probable but still Low impact as, after all, it's still only a trim runaway. However, you now have a scored a Yellow 8. Add exactly the same 'average' crew and stab switch mitigation procedure and you have only reduced the score to a Yellow 4.
For sure, my scenario above is extremely basic but it is meant to offer that MCAS probably brought increased likelihood of a trim runaway but there appears to have been no effort (certainly prior to the LionAir crash) to put enhanced measures (hardware, software, training) in place to retain the original risk assessment score for such a failure. Introduction of MCAS without additional mitigations increased the overall risk associated with a trim runaway event. Boeing should have number crunched this one and it would be very interesting to see the numbers.