I think it would be really cool if there was a way to build a QBERT-like model for coaches. Often the coach gets the blame for losses (and sometimes the credit for wins). I think it would be amazing for someone to quantify it.
Since coaches don't switch teams that often, I guess the data is hard. But players *do* switch. So maybe the model is: how much better is a given player for a given coach.
I think it would be really cool if there was a way to build a QBERT-like model for coaches. Often the coach gets the blame for losses (and sometimes the credit for wins). I think it would be amazing for someone to quantify it.
Since coaches don't switch teams that often, I guess the data is hard. But players *do* switch. So maybe the model is: how much better is a given player for a given coach.