One of the major learnings, mostly the hard way, I have had in modeling is that if I don't make my outcomes and criteria explicit, I might get just that. In other words, do the golfers you model always have the same performance? Probably not. Maybe you should define your criteria for success in this part of the process over a longer time frame? I am not trying to take you off the hook for setting a high standard, but by looking at the micro you may only see the exception. Another option may be to continue your modelling by finding an examplar who is good at "recoving" after their swing has fallen apart. Model their strategy and now you've got the opportunity to enjoy getting your swing back quickly whenever you "forget". In other words, and I think this is the position presented in WITW, you don't make your pattern explicit until you have hit criteria. You can't hit criteria until you know what it is. May sound obvious, but this is one that has slipped by me before. Hope this is helpful. Have a great project. Todd