[CTcard]

Has anyone sent Shaw this handy guide?

While I have no particular complaint with the recommendations of that chart overall, it somehow bothers me that people create these charts then describe them with statements like "the math says you should go for it on fourth down here".
The math presented says that under certain sets of assumptions one should go for it on fourth down with a certain confidence level.
That set of assumptions may or may not be true for analyzing averages across entire leagues and game situations, but by definition not so for a particular team for a particular situation.
It is a useful exercise to understand what these models say under various assumptions, but the chart-for-all is not necessarily the best approach.

Note none of this is a comment on what Stanford should do differently on fourth downs. I don't think these mathematical/statistical limitations have anything to do with Shaw's approach to fourth down situations, and it seems fairly clear to me Stanford would do better with more attempts to complete first downs.

[CTcard]

Has anyone sent Shaw this handy guide?

While I have no particular complaint with the recommendations of that chart overall, it somehow bothers me that people create these charts then describe them with statements like "the math says you should go for it on fourth down here".
The math presented says that under certain sets of assumptions one should go for it on fourth down with a certain confidence level.
That set of assumptions may or may not be true for analyzing averages across entire leagues and game situations, but by definition not so for a particular team for a particular situation.
It is a useful exercise to understand what these models say under various assumptions, but the chart-for-all is not necessarily the best approach.

Note none of this is a comment on what Stanford should do differently on fourth downs. I don't think these mathematical/statistical limitations have anything to do with Shaw's approach to fourth down situations, and it seems fairly clear to me Stanford would do better with more attempts to complete first downs.