More importantly, both offense and defense have dynamic decision-making. Players are coached to react and change on the move. This enormously complicates any attempt to quantify the odds of success or failure, even if you could calculate them for a given play vs. a given defensive call. In reality, coaches don’t really try to quantify relative odds in their playcalling. The large numbers of parameters, many without calculable metrics, make it effectively impossible.

At best, you get some a feel or sense that some calls being riskier than others, and a sense that certain calls are better at exploiting the advantages your team has in a given situation. Unless you are confident you’re reading his mail, you can’t get too wrapped up in what specific call the other guy is making.

When I have more time, I’ll pass on some advice I heard the other day that Norm Chow gave another college OC a few years ago when he was with the Titans.

I’m trying to group plays in two categories.

Situation(conversion, red zone, third down keys).
Field placement.

Then we can have a play call sheet that combines the two. On the armband we can color higlight the situation, and have that placed with the center placement for balance.

Then we still have placement selections for wide field sides, and perhaps a restraint play or two built into those.

We can run a lot more items in effeciency terms from placements to our sideline, so the restraint calls, different plays, should be there with greater frequency.

Yes, if that’s what you mean by illogical then that’s fine. You also could just choose randomly, and that was part of the answer.

But wouldn’t that be picking logically? It’s a continuation of the guessing game. Picking illogically to me would be like choosing the bridge with snakes because you like snakes and your pursuer doesn’t know that you like snakes. She might be on the other side but it would be because of luck and not because you were outsmarted.

One takeaway (and not the only one) is that, play-calling is not easy, and a mixed strategy “kinked” in a direction of your strength but otherwise random might be the best, or at least the best starting point.

I get the point of the bridges example, but couldn’t one arrive to the conclusion to just randomly choose a bridge? If you know what the other person is probably thinking, and you know that they probably know what you are thinking, wouldn’t randomly choosing negate that? Or perhaps even better, wouldn’t illogically choosing a bridge work too?

http://www.consoleclassix.com/info_img/Tecmo_Bowl_NES_ScreenShot4.gif