The document discusses using dynamic programming to determine when an NFL team should attempt a one-point or two-point conversion after scoring a touchdown. It models the problem as a series of decisions based on the score differential and remaining possessions. The dynamic programming approach considers all possible outcomes at each stage and guarantees an optimal solution, without enumerating all possibilities. It formulates the problem as a longest path problem to maximize the probability of winning.
Related topics: