Investigate an "early aggressive pick" strategy

[blchelle]

Description

Because the NFL usually has a preliminary set of game (TNF), it is possible that some weeks it is more optimal to make an aggressive pick on the Thursday game even when it would normally not pick that game. The logic here is that if the upset doesn't pull through, it can still bet extra aggressively on Sunday to have a shot at winning. And if it gets the pick right, it can choose to bet conservatively or slightly aggressively, depending on the nature of the games on Sunday.

I think this would be done as follows:

1. Simulate the event where we make the aggressive pick for TNF and
   a. It succeeds
   b. It fails
2. Run the "best picks" for the remaining games, given both events.
3. Weight the two outcomes by their probability of occurring.
4. Compare the results to the best picks when the early game is not assumed.