If you haven't read this year, you should: SDSU coach Rocky Long, inspired by Pulaski Academy (Ark.) coach Kevin Kelley, is considering going for it at all times within the opponent's fifty. Kelley has had magnificent success with even more extreme tactics at Pulaski; he in fact tries an onside kick every time he kicks off, in addition to disliking punting. Will Long have similar success? I don't know, but I might have to stay up late to watch a couple of SDSU's games just to find out.
The big question with this concept is whether or not a place like Pulaski that has found success with it did so because the idea is sound on a universal level (there's some statistical evidence that suggests that it is), or whether it's because they have the players to make it work. In high school, jettisoning conventional wisdom oftentimes works because some teams have so much more talent than others that they're able to impose their will on most opponents in almost any situation. Why would you want to punt if you're more likely to net a 20-yard gain than to turn it over on downs?
Still, the idea does seem to have some legitimacy. I've long thought that it's unwise to punt the ball on a manageable fourth down when you're within the opponents' 40. A bad punt means you end up gaining less than 20 yards. Of course, if you have a great punter who is good at pinning the ball on the goal line, that's one thing, but there aren't too many of those lying around. Similarly, kicking a field goal on the goal line seems strategically dubious (remember the one we missed against Nebraska?), because while you've got a good chance of making the field goal, you've also got a pretty good chance of scoring a TD, and if you turn it over on downs, the other guy gets the ball with his back against the wall and you have a good chance at setting up another TD opportunity for yourself, while if you miss the FG, he gets it on the 20. I'm curious to see if Long proves some of these notions right.