I'll check it out, but the second portion of your comment seems questionable. Trending to a local minimum or maximum in modeling can be caused by small changes

I'll check it out, but the second portion of your comment seems questionable. Trending to a local minimum or maximum in modeling can be caused by small changes at certain points dependent on time and be completely negligible at other points. Without understanding the model: 1. We can't even pretend to understand causality 2. We can't infer the complex state at an individual time or even the magnitude of a push.